Servo Magazine 07-2017

D930SW

D954SW

Operating Voltage Range No Load Speed Range Peak Torque Range

4.8V ~ 7.4V 0.11 ~ 0.07 Sec @ 60° 8.0 ~ 13.0 kg-cm 112 ~ 182 oz-in

Maximum Current Draw

0.19 ~ 0.12 Sec @ 60° 18.0 ~ 29.0 kg-cm 251 ~ 404 oz-in

18.0 ~ 29.0 kg-cm 251 ~ 404 oz-in

5,200mA

Dimensions Weight

D955TW

40.0 x 20.0 x 37.0mm / 1.57 x 0.78 x 1.46 in 66g / 2.33 oz

68g / 2.40 oz

66g / 2.33 oz

Hitec continues to expand our innovative D-Series servo line with the latest introduction of our Premium trio, the D930SW, D954SW and D955TW. Powered by our 32-bit MCU (Microcontroller Unit) and 12-bit ADC (Analog to Digital Converter) industrial science, this high-resolution, ultra-durable gear group delivers exactly what you need and have come to expect from Hitec. Our D930SW is the perfect upgrade from the HS-8330SH; the D954SW replaces the HS-7954SH while the D955TW is the advanced version of the HS-7955TG. Servo Engineering ReimagIned!!! Hitec RCD USA, Inc. | 12115 Paine Street | Poway, CA 92064 | (858) 748-6948 | www.hitecrcd.com |

Full Page.indd 2

5/30/2017 7:38:52 AM

Full Page.indd 3

5/30/2017 7:37:28 AM

TOC - Jul 17_TOC SV Mar 15.qxd 5/30/2017 8:02 AM Page 4

s etail ore d ng, For m bscribi n u on s our ad o see ge 13. Pa

07.2017 VOL. 15 NO. 7

Subscription Information

Columns

SERVO Magazine — PO Box 15277 North Hollywood, CA 91615-9218 Call 877-525-2539 or go to www.servomagazine.com Subscribe • Gift • Renewal • Change of Info

08 Ask Mr. Roboto PAGE 60

with Eric Ostendorff

There are many kits out there to choose from which don’t require any programming that can help give a gentle nudge towards robotics. Here’s what Mr. Roboto recommends.

60 Then and Now by Tom Carroll

What is the Future of Humans Working Among Robots? What is the future role of robotics? Will robots displace all human workers? No one can say exactly for sure. The needed advancements that are mentioned here are just some of the stumbling blocks that will have to be addressed as we build better and better robots to work alongside us.

The Combat Zone

20 Choosing a Battery Charger 21 RoboGames 2017: The Rise of Whiplash

24 The Conversion of a Robot Skeptic: My First RoboGames

SERVO Magazine (ISSN 1546-0592/CDN Pub Agree#40702530) is published monthly for $26.95 per year by T & L Publications, Inc., 430 Princeland Court, Corona, CA 92879. PERIODICALS POSTAGE PAID AT CORONA, CA AND AT ADDITIONAL ENTRY MAILING OFFICES. POSTMASTER: Send address changes to SERVO Magazine, P.O. Box 15277, North Hollywood, CA 91615 or Station A, P.O. Box 54, Windsor ON N9A 6J5; [email protected]

4

SERVO 07.2017

TOC - Jul 17_TOC SV Mar 15.qxd 5/31/2017 10:29 AM Page 5

In This Issue ... 22 An Arduino Controlled Robot 50 The Multi-Rotor Hobbyist Arm — Part 2 by Ricardo Caja Calleja Last month, I showed how I had built my robot arm, describing the construction of the body and the design and assembly of the hardware. Now it’s time to get our arm moving with the software.

30 Animatronics for the

by John Leeman ESC Basics Just like an engine powers your car but is controlled by a computer, the brushless DC motors (BLDCs) we talked about last month power your quad but are also controlled by a computer. This computer — the electronic speed controller (ESC) — can make or break the performance of your quad. This month, we’ll go over the basics of ESCs and test one out with different firmware versions to see the difference between them.

Do-It-Yourselfer by Steve Koci Shine a Light on Your Creation What good is a fabulous prop if you can’t see it? Serious attention should be devoted to properly lighting your displays. The effort you apply here can make or break your scene!

36 The Wonderful World of Robot Mastermind, Tony Ellis by Dave Prochnow Prepare yourself for yet another British invasion. This time, however, it will be with robots. Roboticist, inventor, and entrepreneur Tony Ellis has done it again — he’s created a home-based autonomous robot that puts all others to shame. PAGE 50

40 Hardware Hacking a Power Wheelchair Control Module by Darby Hewitt On the surface, using a power wheelchair as a robotic base seems like a no-brainer. Unfortunately, it is not, so here’s what I did.

45 RoboGames: The Original Robot Amateur Hour by Camp Peavy This is not some slick trade show with suits, ties, and overthe-top corporate sales booths. It’s not a Hollywood production either, with predictable scripts and special effects where everything just works out. RoboGames (the Olympics for robots) is the biggest “open” robot event in the world.

Departments 06 Mind/Iron A Matter of Trust

07 Events Calendar 13 Showcase 14 New Products

15 RoboLinks 58 SERVO Webstore 66 Advertiser’s Index

PAGE 40 SERVO 07.2017

5

Bergeron - Mind-Iron - Jul 17_Mind-Iron - Feb15.qxd 5/30/2017 7:04 AM Page 6

Mind / Iron by Bryan Bergeron, Editor ª

A Matter of Trust ould you trust a robotic surgeon to operate on you or a loved one? Would you trust your life to a fully autonomous jet liner? Or, an autonomous driverless car? Would you trust a domestic robot around your newborn or your grandparents? Would you trust an autonomous swarm of attack drones to find the intended enemy and spare innocent civilians? Despite all of the media hype about driverless cars and the various experimental robotic systems that do everything from cook noodles to transfer patients from a gurney to a hospital bed, adoption relies first on human trust, and then the technology. For example, I eagerly await the availability of a driverless car — from a purely technological perspective. I can’t say I want to be a rider in the first-generation vehicles. As a cyclist, I definitely don’t want to share the road with vehicles that may not have been programmed to recognize a biker. There are simply too many unknowns for me to trust such a vehicle. As far as robotic surgeons go, I have even less trust in the first few generations of these devices. I’m not saying that I don’t think autonomous surgical robots are inevitable, and will be needed to reduce costs and increase availability of surgical procedures in rural and battlefield environments. I’m saying that I don’t want the robot operating on me. I don’t trust the technology — not yet. Perhaps it’s because I know too much about the variables involved in a typical surgery — no two patients are anatomically identical or respond to medicines in exactly the same way, for example. The latest coolest robot from Boston Dynamics — the SpotMini — is a technological marvel. As a robotics

W

6

SERVO 07.2017

enthusiast, I’d love to get my hands on one. However, there’s no way I would trust the robot — which looks like a dog minus the skin — around a toddler or even a pet. The robot looks like it could do some real damage with that face arm if the control system somehow failed. Given my reservation on these and other robotic systems, I can only imagine the acceptance hurdles — which I’ll label as trust issues — faced by the general consumer. Somehow, the robotics community has to address these trust issues or risk losing all technological advances. I suppose there will always be a military market for robotics, but even there, someone has to trust that an autonomous swarm of attack drones isn’t going to target and fire on friendlies. Developing trust obviously involves time and positive experiences with robotics. For example, consumers have to witness driverless cars accumulate nearly perfect driving records. Studies of surgical robots have to demonstrate both superior outcomes and pre-op experiences. Given the history of technologic development, these track records will be established first in the military — with autonomous vehicles transporting fuel and supplies, and pilotless drones airlifting injured soldiers from the battlefield. This is where technical excellence comes into play — as a necessary (but not sufficient) component in the trust equation. Do you have trust issues with robotics? Or, do you implicitly trust robots to outperform human operators? Either way, I’d like to hear about it. SV

FOR THE ROBOT INNOVATOR

ERVO

Published Monthly By T & L Publications, Inc. 430 Princeland Ct., Corona, CA 92879-1300 (951) 371-8497 FAX (951) 371-3052 Webstore Only 1-800-783-4624 www.servomagazine.com Subscriptions Toll Free 1-877-525-2539 Outside US 1-818-487-4545 P.O. Box 15277, N. Hollywood, CA 91615 PUBLISHER Larry Lemieux [email protected] ASSOCIATE PUBLISHER/ ADVERTISING SALES Robin Lemieux [email protected] EDITOR Bryan Bergeron [email protected] VP of OPERATIONS Vern Graner [email protected] CONTRIBUTING EDITORS Tom Carroll Kevin Berry R. Steven Rainwater Eric Ostendorff Steve Koci John Leeman Ricardo Caja Calleja Darby Hewitt Camp Peavy Dave Prochnow Ian McMahon Matthew Vasquez Anne Lillie CIRCULATION DEPARTMENT [email protected] WEBSTORE MARKETING COVER GRAPHICS Brian Kirkpatrick [email protected] WEBSTORE MANAGER/ PRODUCTION Sean Lemieux [email protected] ADMINISTRATIVE STAFF Re Gandara Copyright 2017 by T & L Publications, Inc. All Rights Reserved All advertising is subject to publisher’s approval. We are not responsible for mistakes, misprints, or typographical errors. SERVO Magazine assumes no responsibility for the availability or condition of advertised items or for the honesty of the advertiser. The publisher makes no claims for the legality of any item advertised in SERVO. This is the sole responsibility of the advertiser. Advertisers and their agencies agree to indemnify and protect the publisher from any and all claims, action, or expense arising from advertising placed in SERVO. Please send all editorial correspondence, UPS, overnight mail, and artwork to: 430 Princeland Court, Corona, CA 92879.

Bergeron - Mind-Iron - Jul 17_Mind-Iron - Feb15.qxd 5/30/2017 7:04 AM Page 7

EVENTS JULY 6-9

8-12

8

15-16

16-19

Robocore IMT, São Paulo, Brazil Events include Sumo, Mini Sumo, LEGO Sumo, Line Following, Hockey, and Robot Trekking. www.robocore.net

ASABE Robotics Competition Spokane, WA Autonomous Agricultural Robot Challenge. www.asabemeetings.org

24-30

Botball International Tournament Norman, OK This year's event is a Robot Assisted Agricultural Challenge. www.botball.org

International RoboSub Competition SSC Pacific TRANSDEC, San Diego, CA Autonomous robots must complete an underwater course. www.robosub.org

25-27

AUVS International Aerial Robotics Competition Atlanta, GA Autonomous robots perform a different challenge each year. www.aerialroboticscompetition.org

25-31

RoboCup Robot Soccer World Cup Nagoya, Japan Events include small-sized robot soccer, mid-sized robot soccer, legged robot soccer, humanoid robot soccer, and Rescue Robots. www.robocup.org

RoboBombeiro San Miguel Sports Hall, Guarda, Portugal Fire Fighting robot contest. http://robobombeiro.ipg.pt International Autonomous Robot Racing University of Waterloo, Waterloo, Ontario, Canada Autonomous robots navigate around fixed obstacles. http://robotracing.wordpress.com

Make your machine move MICRO LINEAR ACTUATORS · 10mm-300mm stroke · 25kg+ available force · 6v-12v power supply · 15g-100g net weight ACTUONIX . COM SERVO 07.2017

7

Ostendorff - Mr Roboto - Jul 17_MrRoboto - Sep 15.qxd 5/30/2017 7:06 AM Page 8

Ask Mr. Roboto

by Eric Ostendorff

Tap into the sum of all human knowledge and get your questions answered here! From software algorithms to material selection, Mr. Roboto strives to meet you where you are — and what more would you expect from a complex service droid?

Q

. My stepdaughter has excellent mechanical aptitude for a fourth grader and I’d like to give her a gentle nudge toward robotics. Can you recommend a simple inexpensive kit that we can build together? Preferably one which requires no programming, since we know nothing about that. I have a shop and tools, so we can handle mechanical construction, wiring, and even soldering. I’ve seen some robots which follow lines and some which steer towards light. Something like those might be fun to try. Anthony Powell Wheeling, WV

A

. Fantastic question! There are many kits out there to choose from which don’t require any programming. I’m a mechanical engineer and toy designer. I was into toy robots as a 1960’s kid, back before electronic and programmable toys when everything was done with brute mechanical cleverness. I’ve always been a huge fan and staunch supporter of simplicity and minimalism, so I may dive deeper into this question than you were prepared for. Elegant robotic walking mechanisms are fun to build and watch, and are a great first project. Two of my favorite snap-together kits are shown in Figure 1. The motorized “electric walking spider robot” (eBay # 401202985302) and Theo Jansen’s wind-powered “Strandbeest” (eBay #

Figure 2.

8

SERVO 07.2017

Our resident expert on all things robotic is merely an email away.

[email protected]

Figure 1.

252416741833) are each under $7 from China. These versions simply walk in a straight line, but I have found them to be quite hackable to build “real robots.” In fact, I wrote construction articles on each, in the March and April 2014 issues of SERVO. Videos are available at https://www.youtube.com/watch?v=bya4uAdNXos and https://www.you tube.com/watch?v= 5rcjZZreLFY. USA-based OWIKIT has been making all sorts of robot kits for many years. You can visit them at www.owirobot.com. OWI’s reconfigurable multi-tread allterrain robot is shown in Figure 2. Their current line is not electronic, so assembly is strictly electromechanical: switches, wires, motors, gears, and linkages. Some of their kits can be built in multiple configurations, including walking and rolling robots. Some use batteries and others have various alternative energy sources: hydraulic, solar power, compressed air, and salt water fuel cells using magnesium. OWI also makes kits to build a hydraulic robot arm and a new mag-lev train. Two of my favorite OWI toys not on that website are the Air Racer at https://www.you tube.com/watch?v= wa1pL5cwwo4 and their EDGE robot arm at https://www.you tube.com/watch?v= LivkkLzf82c. The vintage TAB “Build Your Own Robot Kit” shown in Figure 3 was sold in bookstores and electronics retailers circa 2001. Fairly revolutionary in its day, it can still be easily found in various stages of completion on eBay and Amazon for $25-$50. Reviews are polarized; people either

Ostendorff - Mr Roboto - Jul 17_MrRoboto - Sep 15.qxd 5/30/2017 7:06 AM Page 9

Your robotic problems solved here.

To post comments on this article and find any associated files and/or downloads, go to www.servomagazine.com/index.php/magazine/issue/2017/07.

Figure 3.

loved or hated it (or they misunderstood it). The fully-assembled PCB (printed circuit board) is the chassis, so building it is just light mechanical assembly of motors and wheels; 30-60 minutes max. The PCB has two infrared (IR) obstacle sensors and the PIC16C505 microprocessor comes preprogrammed with several routines (called “behaviors”) which may be selected using the included TV-style remote control. Behaviors include auto free-roam, seek light, avoid light, and follow wall (a.k.a., maze solver). The robot can also be driven directly using the remote control. No programming is required, but there is an empty socket which allows you to add an optional BASIC Stamp 2 chip if you ever want to try programming. It’s a noteworthy feature, but somewhat clunky as implemented since extra SHIFTIN and SHIFTOUT commands are required for the Stamp to talk to the PIC. I programmed my TAB robot to chug around a figure 8 and flash its LEDs, as shown at https://www.youtube .com/watch?v=PBe3EM5gLN0. The kit includes a vintage CD-ROM which has the assembly manual and tons of useful information, including a generous excerpt from everyone’s favorite book, “Robot Builder’s Bonanza” by guru Gordon McComb (http://budgetrobotics.com). I highly recommend that book for anyone interested in robotics. With dozens of years of experience behind him, McComb gives valuable info and insight on mechanical construction as well as electronics (hardware) and software (programming). You mentioned line followers and light seekers. I have built both recently from kits. These are great choices since they are interactive with lots of play value after assembly to keep kids interested. There are amazingly inexpensive ($5 and up) line follower robot (LFR) kits from China; search on Banggood or eBay for “D2-1 tracking smart car” (Figure 4).

Figure 4.

Various kits use AA batteries; some use AAAs — whichever battery holder your supplier has. The Chineseonly instruction sheet is not difficult to figure out, and there is an oval line course printed on the reverse side. Soldering is required; being able to read a schematic diagram and use a multimeter will help to verify resistor values. The circuit board is well marked and easy to solder, making the kit enjoyable to assemble. In fact, the PCB is the chassis. My only caveat is that the PCB is one-sided; the holes are not plated through. If you overheat the board, the copper trace can come off. This is easiest to do where the line sensors connect up front; you must solder two LEDs and two photocells (a.k.a., LDRs) on at just the right height, so be careful. There’s a helpful construction video at https://www.youtube.com/watch?v=BcUCDADzndw. The LFR compares light values using the LDRs to sense the line position, and an LM393 comparator chip controls the motors using 8550 PNP switching transistors. The oldschool cadmium-sulphide LDRs are most sensitive to yellow/green light, yet the kit uses garden-variety red LEDs to illuminate the line. In spite of the fact that nothing is optimized, the line follower works just fine; often without any adjustment at all to potentiometers R1 and R2. Here’s my empirical adjustment tip for the two pots. Set either one mid-range, and switch it on. Hold it up off the ground and adjust the other pot so both motors are half running. You’ll know the SERVO 07.2017

9

Ostendorff - Mr Roboto - Jul 17_MrRoboto - Sep 15.qxd 5/30/2017 7:06 AM Page 10

Figure 5.

Figure 7.

Figure 6.

sweet spot when you find it. The adjustment is coarse and easy. The very simple LFR control circuit has a few limitations: It only drives one motor at a time, so it’s forever zig-zagging; it never drives straight. The motors can’t reverse (only stop/go), so there’s a minimum radius for a curve. Finally, it’s fairly slow, so don’t expect to win any speed contests. However, if your stepdaughter likes it, show her the impressive speeds demonstrated at https://www.youtube.com/watch?v=7omDkur_fk8. I bought several extra of those $5 LFR kits to hack, since the “sum of the parts” is worth more than that. Stock up! There are similar robot kits which follow lines plus have obstacle avoidance and Bluetooth control. Search eBay for “smart cars” D2-1, D2-2, D2-3, D2-5, and D2-6. Naturally, I had to hack some of these cheap LFRs to speed them up by allowing both motors to run simultaneously; see https://www.youtube.com/watch?v=3CVrul3pyJg and https://www.youtube.com/watch?v=uJckB9PDNl0.

10

SERVO 07.2017

I used exactly the same components, just wired together differently. My simplest hack is shown in Figures 5, 6, and 7, where you break one trace and add one wire, leaving all components in their stock locations. I recommend you start by building the robot as designed, then make the mod. Breaking the trace to the IC pin 3 shown in Figure 6 is easily accomplished with a few scrapes of a hobby knife. Use a multimeter to verify the break. Then, add a wire from pin 3 to the outermost side of R7. The original circuit connected pin 3 with the centermost side of R7. The new arrangement provides an overlap zone in the sensor voltages where both motors run, which makes the LFR run faster and smoother. The only downside of this simple mod is that adjusting the modified LFR is trickier than the stock unit and requires empirical calibration, i.e., experimenting! Calibration depends on the battery voltage and the physical position of the LEDs and photocells which are hanging out in space and vulnerable to bending. So, be careful not to crash or bend them. I run my photocells ~ 0.100” AGL and the LEDs ~0.200” AGL. Use the small oval course printed on the back of the instructions to calibrate. Start with both pots adjusted to their center position. Per Figure 7, the left pot MOSTLY adjusts the right turns and the right pot MOSTLY adjusts left turns; they affect each other and you need to strike a balance. Calibration is an iterative process where you test a right turn circle (clockwise) and slowly slightly adjust R2 counter-clockwise until it follows the turn properly. Then,

Ostendorff - Mr Roboto - Jul 17_MrRoboto - Sep 15.qxd 5/30/2017 7:06 AM Page 11

Figure 8.

PCB sure makes a classy chassis with part outlines and numbers in red ink. The small fold-out book gives step-bystep instructions. Construction takes 30-60 minutes with two novel elements. First, the square metal motor cases are fastened to the chassis at oblique angles by solder only. You’ll need a big iron and some attention to detail to mount both motors symmetrically. Second, the tail roller uses a big fat barrel diode for an axle (LOL). It’s soldered in place for mechanical function only, but I think it would have been clever to use it in the circuit for something. After a smooth build, I was prepared to be wowed. Straight off the Shack’s website: “This photosensitive microbug follows light and stops in darkness.” To me, that clearly means that it will track towards a light source. Heck, I’ve already done that with two photocells and one servomotor: https://www.youtube.com/watch?v=4CaAq6gQbLY. Surprise, surprise, however. Here are the exact operating instructions given under “Running Microbug:”

test a left turn circle (counter-clockwise) and slowly slightly adjust R1 clockwise until it follows the turn properly. Then, go back and retest right and left turns. There is a best balance of pot adjustments which will Switch on to play. If you want Microbug to turn and let the robot track properly right and left. Once you get it circle left, fully turn the potentiometer on the right. This dialed in, you can try speeding it up. One of my videos then dims the LD resistor and stops the motor on the shows that faster speeds come from using higher voltage right. To run and circle in the opposite direction, fully turn Li-Ion cells (~4V) compared to the stock 3V alkaline cells the left potentiometer. The brighter your surroundings, the (recalibration required). How fast can YOU go? faster Microbug runs. Sadly, as I write this, RadioShack stores are having their final clearance sales, and I bought a “Running Microbug” YOU GOTTA BE KIDDING ME! This doesn’t begin to kit for $15 (cat #2770353, Figure 8). This minimalist robot describe or take advantage of this robot’s components or kit uses direct drive from two small square motors: no function. This is neither educational nor fun. Where’s the gears, no wheels. The motors tilt down so that their shafts touch the ground, becoming the world’s smallest wheels (Figure 9). Clearly not suitable for carpet, this is a fast robot for a smooth open floor. While quite functional, the openframe motors are practically dragging x Design New Ideas on the floor and easily pick up hairs, dirt, and anything metallic. I found this out the hard way and had to x Prototype Without carefully de-gunk two greasy the Wait commutators. I would recommend enclosing x Cut Real Metal the bottom of the motor with a piece of scotch or electrical tape to keep things out. x 120VAC - Plug in CNC Mill This must be done carefully to Starting at: Anywhere avoid taping the armature in place. Make a convex pocket in the tape and/or make a non-sticky spot on the tape (paper, tape dot) where it might TORMACH.COM touch the rotor. The white fiberglass

smallmachine BIGRESULTS

$4950

SERVO 07.2017

11

Ostendorff - Mr Roboto - Jul 17_MrRoboto - Sep 15.qxd 5/30/2017 7:06 AM Page 12

Figure 10.

Figure 9.

light tracking function? I found several videos such at https://www.youtube.com/watch?v=3eWeM6OQnMc which — at best — demonstrate random motion. After testing mine, I realized that the circuit makes the robot want to turn towards the dark and stop, which is pointless and futile — not to mention just the opposite of the website claim. I swapped the motor wires left and right, and aimed the LDRs forward and down (Figure 10). Suddenly, the robot came to life and began tracking towards a flashlight beam. It is squirrely fast and a handful to steer, but quite fun. Check it out at https://www.youtube.com/watch?v=d993QBPE6P4. Those two little tweaks made all the difference in the world, but most people may never know about them. That really annoys me. It’s as if no one at RadioShack ever cared

enough to build and test this item. Did they not get customer complaints? Another haphazard element in the instructions is that it gives you a choice of how to mount the two front LEDs: flush and pointing up as I did; or hanging out front like bug antennae. They are right in the crash zone, and also right where they can potentially be seen by the photocells. In that configuration, every crash probably moves the LEDs and varies their effects on the photocells, introducing random control factors. This is ROBOTICS, this is SCIENCE, darnit! THIS is why the Shack closed down. Well, at least you and I know how to fix this item’s shortcomings. (Sorry, I didn’t mean to get off on a rant.) I will miss the Shack for sure. I’ve spent lots of time and money there. It was a great standard resource for builders and makers to specify commonly-available parts, so it just got harder for the DIY crowd to get parts locally. Plan to stock up on robot parts for future builds! Well that’s all we have time for. I hope this info was helpful. Even the simple robots mentioned here can be lots of fun and very educational. Best of luck sparking interest in today’s youth! They will be the roboticists of tomorrow. Don’t forget! If you or they have more questions, send them to me at [email protected]. See you next time! SV

12

SERVO 07.2017

Page 13 - Jul 17_Bots in Brief Mar15.qxd 5/30/2017 7:08 AM Page 13

Stock Drive Products Setting Ideas Into Motion

One-Stop Shop for Mechatronic Components Explore | Design | Buy Online

Over 100,000 small mechanical parts

www.sdp-si.com

no minimum requirement

SERVO 07.2017

13

New Products - Jul 17_Mar15 - NewProd.qxd 5/30/2017 8:06 AM Page 14

NEW PRODUCTS Precision Spur Gear Motors

P

recision spur gear motors from ServoCity are available now for $34.99. There are eight available choices offering a wide selection of RPMs (from 6-90 RPM).

D-Series Hitec Servos

S

ervoCity is also offering Hitec’s D-Series servo line which sets new standards for advanced programmable technology. Powered by their 32-bit MCU (Microcontroller Unit) and 12-bit ADC (Analog-to-Digital Converter) industrial science, these powerhouses dominate the servo arena. Their potent combination of wide voltage capabilities, high resolution, rapid response, and patented Smart Sense technology brings maximum performance to all radio control applications. Members of this particular D-Series group are all designed with extremely durable titanium gears; the D941TW, D946TW, and D951TW trio boast full metal cases. Pricing is $157.99. For further information, please contact:

ServoCity

www.servocity.com

2,048 Microstep Resolution with New Cost-Effective Compact Stepper Motor Controller

PI

(Physik Instrumente) has released a higher performance model of its Mercury stepper motion controller. Stepper motors take up discrete positions in a revolution of a constant distance. Typical commercial models provide 200 to 1,000 full steps per revolution. Designed to deliver more than 2,000 times the basic motor resolution, the C663.12 Mercury controller is the newest addition to PI’s suite of motion control solutions. PI’s compact C-663.12 motion controller is designed for two-phase stepper motors in open-loop or closed-loop operation commanded by USB or RS-232. Up to 16 units can be combined via daisy chain to operate multi-axis motion systems. The Mercury controller includes a 48V wide range input power supply, and all cables required for operation. Programmable digital and analog I/O lines and input lines

14

SERVO 07.2017

for limit and reference point switches controlled via TTL signals are also integrated. The C-663.12 is equipped with a data recorder for high speed tracing, ID chip compatibility for quick start-up and on-the-fly parameter changes, and exchange of system components without recalibration. The controllers are delivered with extensive software packages, including drivers for LabVIEW as well as dynamic libraries for Windows and Linux.

Piezo Linear Motor

PI

also introduces a new robust OEM walking motor drive — the N-331 — with its configuration rooted in existing patented piezo actuator technology and a patented piezo stepping motion principle design. Based on the application of historically successful technology, the new design is ideal for use in high-end applications that require extreme precision and durability. Unlike traditional electromagnetic motors, piezo motors do not create magnetic fields, nor are they influenced by magnetic or electric fields — a characteristic that is

New Products - Jul 17_Mar15 - NewProd.qxd 5/30/2017 8:52 AM Page 15

advantageous in applications from e-beam lithography to MRI technology. When large optics, detectors, or camera setups in industrial applications need to be positioned with nanometer precision, it can be beneficial to cut the actuator power once the fine positioning process is complete. If a motor or actuator is dependent on a continuous current draw to hold a position, heat dissipation, electromagnetic fields, and thermal effects may have negative influences on the performance of the system. PICMAWalk motors are self-clamping; i.e., at rest or in steady state mode, they automatically hold the position without any need for power.

Extra 330SC — Fly 3D

D

esigned by World Aerobatic Champion pilot, Gernot Bruckmann, the Extra 330SC from Hitec Multiplex is a smaller version of his award-winning aerobatic flyer. Multiplex’s M-frame technology and CFRP-based/two-part wing system make for an extremely rigid and stable airframe, allowing for dramatic 3D maneuvers that can be performed like a pro.

PI’s PICMAWalk piezo linear motor uses 8 PICMA® piezo actuators arranged in a V-shape similar to a classical V8 engine to achieve high push/pull and holding forces along with nanometer precision. The spacetested PICMA actuators are highly reliable, providing 60N (13 lbs) power-off holding force and 50N (11 lbs) push/pull force. The maximum velocity is specified at 15 mm/sec (0.6 inch/sec) and loads up to 5 kg (11 lbs) can be positioned with nanometer precision. For further information, please contact:

PI (Physik Instrumente)

www.pi-usa.us

The detachable fin and tail plane make for convenient and simple transport. Available as a kit ($199.99) a RR ($379.99), the Extra 330SC also comes available in two eyecatching color schemes: a limited edition blue (RR only) and professional decal-designed orange (kit only). Kit and RR features include: • Rigid Two-Part Wing Based on CFRP Tube Spars • M-frame Technology for a Durably Strong Fuselage • Large Detachable Canopy and Detailed Instrument Panel with Dummy Pilot • 100% Factory Assembled ELAPOR™ Foam Airframe • Powerful Brushless Permax® Motor with Speed Control • Four Hitec HS-82MG Servos Specifications: • Length: 47.24 in. (1,200 mm) • Wingspan: 45.28 in. (1,150 mm) • Weight: 47.62 oz. (1,350g) • RC Functions: Rudder, Elevator, Aileron, Motor These planes are available exclusively at www.weekenderwarehouse.com. For further information, please contact:

Hitec MULTIPLEX











   

  




)

For the finest in robots, parts, and services, go to www.servomagazine. com and click on Robo-Links.

www.hitecrcd.com

SERVO 07.2017

15

Combat Zone - Jul 17_Combat Zone - Aug 15.qxd 5/30/2017 8:51 AM Page 16

To post comments on the articles included in this section and to find any associated files and/or downloads, go to www.servomagazine.com/index.php /magazine/issue/2017/07.

Choosing a Battery Charger ● by Ian McMahon battery charger will be one of your most important tools for quick turnaround of combat robots in the pits. For the Insect classes, a small 50W 6S charger or two will be sufficient, but when do you need a higher capacity charger and how do you pick your new charging system? To start with, what does 50W 6S even mean? The 50W refers to the charger having the capability to

A

Featured This Month:

16 Choosing a Battery Charger

17 RoboGames 2017: by Ian McMahon

The Rise of Whiplash

20 The Conversion of a by Matthew Vasquez

Robot Skeptic: My First RoboGames

by Anne Lillie

16

SERVO 07.2017

charge at 50 watts. Watts are volts multiplied by amps, and measure power. The 6S refers to the maximum number of lithium polymer battery cells that this unit can charge in series. These 6S LiPo iMAX B6 charger. batteries have a nominal voltage of discharge 5W 22.2 volts (3.7V x 6) and a fully Charge Current Range: 0.1~5.0A charged voltage of 25.2V (4.2V x 6). Discharge Current Range: 0.1~1.0A If we divide 50 watts by 25.2 volts, we get a maximum 6S charge rate of NiMH/NiCd Cells: 1~15 Li-Ion/LiPoly Cells: 1~6 2.0 amps. Applying this technique to Pb Battery Voltage: 2~20V the other cell counts, we get the Weight: 277g results shown in Table 1. Dimensions: 133 x 87 x 33 mm So, why are the values for 1S and 2S marked in red in Table 1? If we So, this charger can handle my look at the spec sheet for the iMAX Antweight, Physique Black, with its 3S B6, we can see that this charger has a 360 mAh battery. This battery (a maximum charge rate of 5.0 amps Hyperion G3 series) is rated to be regardless of voltage. This gives us the charge rate graph shown in Figure 1. charged at up to 6C. My maximum Here are specs for the iMax B6, allowable charge rate therefore is: 6 x 0.360A = 2.2A. This is below the 4.0 available for between $25 and $40: amp maximum charge rate for the Input Voltage: 11~18V iMax B6 when charging a 3S pack. Circuit Power: Max charge 50W; max Is it sufficient for my new 30 lber,

Combat Zone - Jul 17_Combat Zone - Aug 15.qxd 5/30/2017 8:16 AM Page 17

Table 1.

Maximum Charge Rate for IMax B6

50W Charger S

Max. V Amps

1

4.2

11.9

2

8.4

6.0

3

12.6

4.0

4

16.8

3.0

5

21

2.4

6

25.2

2.0

Figure 1.

Charge Rate in Amps

Speed Wedge 30? I want to be able to charge the battery at 2C for turnarounds in 30 minutes. For this bot, I have a 4S 4,000 mAh battery which is rated to be charged at 10C by the vendor. The wattage needs for the 2C charge are: 4 amphours * 2 charges/hour * 16.8 volts = 134 watts. The amperage needs are eight amps to meet my goal. If I were to use the IMax B6, I would only be able to charge at three amps for a full charge in one hour/20 minutes. This is tolerable if I have three batteries and two chargers available — especially since in each match I should only use about 2.2

Cell Count

amp-hours according to the Tentacle Torque-Amp Calculator (check it out at http://runamok.tech/squid/newtor quecalc.htm).

My theoretical typical charge time should therefore be 44 minutes, which is very tolerable. Nevertheless, I wanted a bigger and cooler charger

RoboGames 2017: The Rise of Whiplash

● by Matthew Vasquez

or the last few years, it seems that our team’s biggest challenge has been fighting in the larger weight classes. For over a decade, our robots were middle of the pack from the old Comedy Central BattleBots™ days through 2013. We were able to use the same old robots and do well. However, the level of competition drastically started to rise the last few years, forcing us to rethink our robots. At this point, we decided to throw our hat back into the heavyweight class and build something new. At RoboGames 2015, our first heavyweight in several years, Whiplash went 0-2 due to weapon problems and its inability to self-right. The robot was barely finished the night before the event, and was never fully tested. We also competed in season two of BattleBots with a new 250 lb robot, but fell short due to a few small problems and a short build time. For RoboGames 2017, we

F

started building about six months before the event, and spent most of our time fixing the bugs that arose previously. This gave us time to test and make sure that the robot was perfect. Because of this, we managed Whiplash vs. Sewer Snake (Credit: Jon Bennett.)

to take the Heavyweight Silver Medal at RoboGames this year. Our new version of Whiplash was an electric lifter with a vertical disk mounted in the arm. Having two weapons originally led to some problems, but if done correctly had potential for success. Our first match was against a new robot named Frankenbot which was armed with two small horizontal spinning bars. When the match started, I immediately drove across the arena, hitting it with our disk. I then pinned him against the wall with our lifter. It was a good first match for Whiplash, and calmed my nerves. For my second fight, I was up against Vlad the Impaler 2 — a robot that has been fighting for over 15 years. This fight was longer, but was very similar to the first. I got a few hits with the disk, flipped him over, and lifted him onto the arena bumper. Next was a fight that I had looked forward to for years. It was SERVO 07.2017

17

Combat Zone - Jul 17_Combat Zone - Aug 15.qxd 5/30/2017 8:16 AM Page 18

Whiplash vs. Swamp Thing

entered the arena. When the referee called “fight,” I (Credit: Jon Bennett.) rushed across the arena, getting to the side of Sewer Snake. After getting in one flip, I realized that our lifter had more reach, giving us the advantage. After several lifts and hits with our spinner, Sewer Snake had lost its lifter chain. Although Sewer Snake had no weapon at this point, the spinner on Whiplash had also stopped working. With a minute and a half remaining in the fight, Whiplash started to slow down. Because of this, I took it easy until I got behind Sewer Snake for the perfect opportunity for a Whiplash vs. knockout. I pushed Sewer Crash and Burn (Credit: Jon Bennett.) Snake at full speed all the way across the arena, lifting him over the bumper, out of the ring. Next thing I knew, Sewer Snake had climbed its way back over the bumper into the arena using its six-wheel drive. For the remainder of the fight, both of the with the four-time RoboGames damaged bots just tried to stay alive. champion, Sewer Snake. Driven by The judge’s decision came in, and the Matt Maxham, this six-wheel drive winner was Whiplash! monster is known for dominating its We quickly went back to the pits opponents with its electric lifter and to find that the weapon motor for the throwing robots out of the arena. spinner was burnt out. Other than Being the favorite, the crowd that, there was no significant damage. cheered as Matt and Sewer Snake We asked around, and thankfully

Whiplash vs. Touro Maximus (Credit: twitch.tv/robogames.)

18

SERVO 07.2017

Fredric Moore of the heavyweight “Boris Badenov” had a new motor to sell us. We installed the motor, and charged up for our next match against the steel wedge and flamethrower of Swamp Thing. Swamp Thing had gotten this far by using its superior pushing power and simply outlasting its opponents. When the fight started, I immediately flipped Swamp Thing onto its back. Shortly after, I hit it with the disk — launching the 220 pound robot a couple feet into the air — then hit it again, sending it into a flip. After this, one side of Whiplash’s drive stopped working. With two minutes left in the fight, Swamp Thing seemed unfazed. The two robots maneuvered around the arena for the remainder of the match, with no one taking the advantage. The decision was in, and Swamp Thing was the winner in a close judge’s decision. Being put into the loser’s bracket this far in the tournament meant that there were several possibilities for our next fight. The two bots in the loser’s bracket that caught my attention were Megabyte and the most feared robot in the tournament, Last Rites. I watched Last Rites next fight as it viciously tore its opponent, Megalodon into pieces, hoping that I would not be its next victim. On the other side of the bracket, the full body spinner Megabyte lost in an upset to the multi-bot wedge and flamethrower combo, Crash and Burn. So, my next fight was going to be against the winner of the matchup between Crash and Burn and the unstoppable spinning bar of Last Rites. I watched the fight from the builder’s area, as Crash Repairs in the pits. and Burn threw (Credit: Debbie Vasquez.) themselves into

Combat Zone - Jul 17_Combat Zone - Aug 15.qxd 5/30/2017 8:16 AM Page 19

Whiplash vs. Original Sin (Credit: Jon Bennett.)

Being told we have two minutes to get in the box. (Credit: Robert Masek.)

Last Rites spinning bar, taking hit after hit. Several huge hits later, Last Rites stopped moving. Crash and Burn were declared the winner in another huge upset. Having not had a great plan prepared for Last Rites, I was ecstatic that Crash and Burn took the win. I went back into the pits and watched as Kurtis Wanner and Travis Schmidt of Team Red Spark Robotics swapped out a new fork attachment to get under Whiplash. The fight started and both teams rushed out of their squares. For a little while, neither team was able to take the upper hand, until half of the multibot, Burn turned its back to me. At this point, I chose to go after that one, chasing it from behind. I hit its back end with the disk, just barely knocking it over the bumper. With Burn out of the fight, it was down to Crash. I hit it several times with the disk nearly getting it out of the arena, until smoke started coming from the weapon motor. Our spinner had died once again, so I was down to a lifter. We continued to maneuver around the arena, trying to get position on each other. I eventually got behind Crash and lifted it upside-down onto the bumper.

Being that Burn was out of the arena, and Crash was stuck, Whiplash won the fight by knockout. Whiplash was to move on to the next round but without its spinner, as we did not have another extra motor. My next fight was going to be against the RoboGames 2015 Champion, Touro Maximus, driven by Daniel Freitas. Touro Maximus is a drum spinner built by Team Riobotz from Brazil. In Touro Maximus’ previous fight, it launched the push bot Whoops into the glass so hard it broke the arena. Having an already disabled robot and being in the loser’s bracket, I went into this battle with nothing to lose. At the start of the fight, I rushed Touro Maximus as it spun up its drum. Touro immediately tore the front spatula off the lifter, sending it into the ceiling. I then got behind Touro, flipping it over with what was left of the lifter’s forks. Touro Maximus selfrighted and hit Whiplash on the front, sending it into the air. As I retreated, Touro got behind Whiplash, damaging the back of the robot. I then came back and lifted Touro Maximus onto the wall as it was tearing armor panels off of Whiplash. Both bots drove back to the

middle of the arena where I flipped Touro again from behind. I drove Touro into the wall, trying my best to lift it out of the arena. With no luck, I kept pushing and lifting, keeping Touro from spinning up. With Touro Maximus against the wall, I gave one more lift, jamming Touro underneath part of the arena bumper. I backed out, realizing that Touro Maximus was stuck. Touro was counted out, and Whiplash was declared the winner by knockout. Shocked by my victory, we threw the bot back on the cart, and rolled it back into the pits followed by several cheering builders. We got back to our pit with our damaged robot, and it seemed that our team had tripled in size. Builders from other teams including Matt Maxham, Fredric Moore, Will Bales, and Lester Nikolopoulos helped pull the robot apart while a large group of builders surrounded us, watching the chaos. With some help, we were ready for our rematch with Swamp Thing in less than 20 minutes. With no spinner, only half of our front armor, and just one out of two lifter forks, this was going to be a challenge. At the start of the fight, Swamp Thing and Whiplash went SERVO 07.2017

19

Combat Zone - Jul 17_Combat Zone - Aug 15.qxd 5/30/2017 8:16 AM Page 20

Heavyweight award ceremony. (Credit: Jon Bennett.)

head to head, with Swamp Thing getting under first. Realizing I would not win a pushing fight, I backed off and maneuvered to his back. I chased him from behind, getting in a flip. With Swamp Thing’s wedge upside down, I was able to get under it from any position. I rammed Swamp Thing into the wall several times, being careful to not flip him right-side up. For the last minute of the fight the drive on Whiplash started to slow down, and would even cut out every now and then. The fight had ended, and the decision was in. Whiplash had won in another close judge’s decision, earning a spot in the finals against sixtime RoboGames Champion, Original Sin. Original Sin is a four-wheel drive wedge that very few people have been able to outdrive. We rushed

back to the pits, followed by the same group of builders who had previously helped. Like before, we were given about 20 minutes to turn the bot around. We set up fans over the robot (provided by Matt Maxham) to cool down the drive motors, pulled out the batteries to charge, and added an extension to the one fork that was left on the robot. We rushed to the arena with seconds to spare, threw the bot off the cart, and were ready to fight. Gary Gin (the driver of Original Sin) and I hit our buzzers, indicating that we were ready. As Dave Calkins, the organizer of RoboGames called for the match to start, both bots immediately attacked. We started with a head-to-head collision, sending both robots back. Original Sin then got under the front of Whiplash, slamming me into the wall. Whiplash then came back, flipping Original Sin over. With Original Sin disoriented, I lifted him onto the wall, but could not get him over. Original Sin then retaliated, driving Whiplash into the wall once again. Both robots slammed into each other head-to-head with Whiplash launching over the top of Original Sin. After landing behind Original Sin, I drove to his side, lifting him up onto the wall. With Original Sin stuck on the wall, I tried my best to lift him over. After several attempts to lift him out, Original Sin escaped. After

missing my chance at a knockout, Original Sin gave another huge slam, throwing Whiplash into the wall. At this point in the fight, Whiplash had slowed down significantly. I tried to keep the robot alive as Original Sin relentlessly rammed Whiplash into the wall for the remainder of the fight. The judge’s decision was in, and Original Sin was now the seven-time RoboGames champion in an 18 to 15 conclusion. The success of Whiplash really reinforced the importance of originality. At RoboGames 2015, our heavyweight entry was a large vertical spinner that resembled a former RoboGames runner-up robot, Electric Boogaloo. Our robot did not do well because of spin-up problems and not being invertible. Even if the robot had worked perfectly, would it have been successful? In combat robots, originality is key. If you try to build a better bar spinner than Last Rites for example, it probably won’t fare as well if you go up against the original. So, don’t be afraid to try a crazy idea. Five years ago, if someone told me I should build a robot with a vertical spinner inside of a lifting arm, I would have laughed. However, it turned out this unconventional design was able to beat some of the best robots in the world. So, think outside of the box. Who knows, your crazy idea could be a champion. SV

The Conversion of a Robot Skeptic: My First RoboGames

● by Anne Lillie

didn’t know robotic combat existed until I met my husband-tobe 12 years ago. I’m not proud to admit that at first my attitude about his hobby wasn’t that great. I knew there were certainly worse things he

I

20

SERVO 07.2017

could do with his time and money, but I also secretly thought it would be nice if he had a hobby that was safer (something without the risk of amputating a limb or setting the garage on fire, for instance), or more

constructive (maybe building something that wasn’t destined to be torn to pieces), or a little more normal (something people have heard of, for example).

Combat Zone - Jul 17_Combat Zone - Aug 15.qxd 5/30/2017 8:16 AM Page 21

Of course, over the years, my interest in the sport has gradually grown. For a girl who didn’t know what a speed controller was, I’ve come a long way. I finally attended my first RoboGames this year, which cemented my conversion from robot skeptic to enthusiast. When my husband decided to build a new lightweight a few years ago, I was blown away by the problem-solving tenacity that building a robot from scratch required. Being at RoboGames exponentially multiplied my respect for that endeavor. I found myself thinking about the fact that every robot there — from the Antweights to the mighty Touro Maximus — was nothing at one time but a figment of someone’s imagination. If you’re a robot builder, you may take this process for granted. Every robot that exists starts in someone’s mind as an idea. What if I build a lightweight flamethrower? At this point, you as the builder contain the robot’s entire existence. If you die, it dies with you, leaving no evidence that it was even a thought in your brain. Then you make a sketch, or play around in a design program. For now, the possibilities are endless — the robot could be anything. No mistakes have been made. However, to transform the idea into a physical thing, you have to sacrifice the possibilities and commit to something. You may spend hours looking at pictures of other robots, consulting other builders online, and standing in the aisles of hardware stores. You search for elusive and possibly nonexistent parts on the Internet. Boxes from Amazon arrive on your doorstep. A person looking at your pieces of steel tubing or sheets of titanium would not suspect or envision the thing that’s shining in your mind, visible only to you. At this point, if you give up, no one will care because they don’t understand how cool this creation

could potentially be. The annoying thing about the world is that it doesn’t realize it needs a 60 pound flame throwing combat robot. Eventually, the robot is upgraded from a figment of your imagination to solid metal — a primitive skeleton version of itself. This is when the real trouble starts. Some of your plans don’t work, and end up being time and money down the drain. Building this thing is an emotional sine curve of successes and failures, with constant regrouping, rethinking, and retracing your route to the hardware store. By the way I’m glorifying this process, a person might expect that the finished product would go in a display case on the mantle (or maybe on the floor near the mantle, depending on the weight class). But no! As if troubleshooting this robot to a point of functioning completion was miracle enough, you then pit it against the most merciless weaponry that the world (and in particular Brazil) has to offer in a battle to the death. One of the first things I realized at RoboGames was that videos do not do justice to the terrifying power of these machines. I would compare watching a heavyweight fight on YouTube to looking at a picture of the Grand Canyon: it conveys the general idea, but it’s nothing like the emotional impact of seeing it in person. As I watched Last Rites spew bits of Sabre X across the floor, it occurred to me that it would be comforting to have a personal Lexan shield — nothing fancy, just something to flinch behind to protect my head and vital organs in the event of an arena breach. Another thing that surprised me was the amount of camaraderie in the robotic combat culture. People whose names I had been hearing for years — whom I considered to be celebrities of the robot world — stopped by our pit area to chat.

Some of them bought our team tshirt. One of them freely rattled off what I would have expected to be top secret information about the advantages and disadvantages of various flame-thrower set-ups. (Maybe he sensed it was safe to divulge this information to me because I could understand only about 30% of what he was saying.) I was impressed that teams shared tools with their rivals. Why help the competition? By the end of the weekend, I realized that — of course — everyone wants to win, but along with winning there is the secondary goal of seeing things get destroyed in spectacular fashion. If the thing getting destroyed happens to be your own robot, that’s okay as long as you put on a good show, and especially if it involves flying shrapnel, fire, and/or LiPo smoke. I left my first RoboGames full of respect for the ingenuity of the builders and the generosity with which they allow their creation to be destroyed for the entertainment of others. During the driver’s meeting, Event Master, Dave Calkins said, “The best engineers in the world are in this room.” At first, I wondered if maybe he was being melodramatic. But here’s the thing: Every person on the planet has ideas, and most of the time those ideas don’t make it out of our heads. It’s so much easier to watch a movie someone else created, or read a book someone else wrote, or sit in the stands at RoboGames and watch a robot someone else built get shredded by a drum spinner. Anyone can sit in the stands and think, I could build a better robot than that. The difference between the best engineer in the world and the person in the stands is that one of them decided to actually build something — even if it was destined to get thrashed to shrapnel bits — and one of them just sat there and thought about it. SV SERVO 07.2017

21

Caja - Arduino arm par 2 - Jul 17_Blank Rough SV.qxd 5/30/2017 7:16 AM Page 22

An ARDUINO Part 2 CONTROLLED ROBOT ARM

Last month, I showed how I had built my robot arm, describing the construction of the body, and the design and assembly of the hardware (electronic controller). In other words, I described the “physical” robot arm. So, now we have a robot arm fully assembled — which is pretty cool — but what can we do with it? Nothing yet! It still needs to be programmed in order to actually move and do other cooler stuff. he robot arm can be controlled manually using the “control box” with rotating knobs (potentiometers) that was built in the first article, as well as through the serial port (USB and Bluetooth) from a PC, Raspberry Pi, smartphone, etc. The possibilities are almost infinite! Especially when it comes to software, the choices are countless: Processing, Matlab, Python, ROS, just to mention a few.

T 22

SERVO 07.2017

So, I had to keep my feet on the ground and choose specific software to start with. Read on to find out the details!

Operation Modes Although I’ve already been experimenting with Bluetooth connectivity, for now we’ll focus on robot arm

Caja - Arduino arm par 2 - Jul 17_Blank Rough SV.qxd 5/30/2017 7:16 AM Page 23

By Ricardo Caja Calleja

To post comments on this article and find any associated files and/or downloads, go to www.servomagazine.com/index.php/magazine/issue/2017/07.

control using a PC via USB. I’ll come back to the serial communication via Bluetooth towards the end of the article. I defined three different operation modes to control the robot arm:

Figure 1. Control box.

• Mode 1: Manually control it with the rotating knobs of the control box (Figure 1). Every robot arm joint and the end effector “fingers” are controlled directly with the corresponding rotating knob (potentiometer). • Mode 2: Manually control it with the sliders of a GUI (Graphical User Interface) from the PC (Figure 2). In a similar way to the control box, the robot arm joints and the end effector fingers are controlled with the six sliders. • Mode 3: Automatically provide the end effector space coordinates (X, Y, Z) with the GUI from the PC (Figure 2). Here is where inverse kinematics come into play! Plus, the end effector fingers are opened or closed with the END_EFFECTOR slider. Once the robot arm controller is turned on, during the first four seconds it tries to find a device connected through the serial port. If there is no device connected, Mode 1 will start. If there is a device connected, Mode 2 will start. Once in Mode 2, the user can switch to Mode 3 by clicking on the INV_KIN toggle button in the GUI. Maybe you noticed that in Mode 3, I mentioned just “space coordinates” and not “pose” (space coordinates + orientation). This is because the end effector orientation will always be horizontal, which is a restriction that was introduced to make calculations easier.

Inverse Kinematics — Geometric Solution The main problem we will face while controlling robot arms will be to bring the end effector to some specific pose, which means that we will have to work with inverse kinematics. In other words, a set of joint angles will be calculated from a given end effector pose. As explained in the first article last month, different methods can be used for this: algebraic, numerical, and geometric solutions. For computing the inverse kinematics of my robot arm, I chose the geometric solution, defining a simplified model.

Figure 2. GUI (Graphical User Interface).

We’ll need to refresh some almost forgotten trigonometry that we learned at school! First, we need to measure the

Trigonometric Functions

• Pythagorean Theorem (only applicable to right triangles): The square of the hypotenuse (c) is equal to the sum of the squares of the other two sides (Xw, Z-p): c2 = Xw2 + (Z – p)2 • Law of Cosines: Side c can be calculated if the angle opposite (C) and the two other sides (a, b) are known: c2 = a2 + b2 – 2 • a • b • cos C • Law of Sines: Relates the side lengths of a triangle to the sines of its angles: a b c = = sin A sin B sin C

SERVO 07.2017

23

Caja - Arduino arm par 2 - Jul 17_Blank Rough SV.qxd 5/30/2017 7:16 AM Page 24

Figure 3. Robot arm link dimensions.

The geometric model is split into two parts: longitudinal and lateral. In the longitudinal part (Figure 5), the space coordinates X and Z are taken as input; for the lateral part (Figure 4), the coordinates X and Y are considered. All the space coordinates will be given in millimeters (mm). The lateral part of the geometric model is calculated first. If the space coordinate Y is not equal to zero, the robot arm base will rotate with an angle ϕ (see Figure 4). The outcome of the lateral part of the geometric model is Xw — the horizontal distance between the robot arm base and the wrist. To calculate Xw, we need to obtain the angle ϕ first with some easy trigonometry: tanϕ =

Y X

()

ϕ = tan-1

➜

Y X

With some easier trigonometry, we obtain the distance X’: cosϕ =

length of every robot arm link (Figure 3). Second, it’s important to define the zero position of every servo, as well as its positive and negative directions of rotation (shown in red in Figures 4 and 5).

X X’

➜

X’ =

X cosϕ

Subtracting the length of the end effector (95 mm, see Figure 3), we get the distance Xw: Xw = X’ – 95 mm

We’re done with the lateral part! Now it’s time for the longitudinal part of the geometric model. As you can see in Figure 5, a right triangle is created within the lengths c and Xw, and the vertical distance between the base plane and the wrist (Z – p). Applying the Pythagorean Theorem to that triangle, the distance c can be easily obtained: c = √ Xw2 + (Z – p)2

A second triangle is created within two robot arm links (a and b) and the imaginary line c. The lengths a and b are known, and now c is also known, so with the Law of Cosines we can obtain the angle C:

(

a2 + b2 – c2

C = cos -1

2

•

a

•

b

)

Last but not least, the Law of Sines helps us get the angles A and B: A = sin -1 Figure 4. Lateral inverse kinematics (top view).

24

SERVO 07.2017

B = sin -1

a • sinC c

( (

b • sinC c

) )

Caja - Arduino arm par 2 - Jul 17_Blank Rough SV.qxd 5/30/2017 7:16 AM Page 25

Considering again the right triangle formed between Xw, c, and Z - p, getting τ and n is child’s play: τ = cos -1

n = sin -1

Figure 5. Longitudinal inverse kinematics (side view).

( ) ( ) Xw

c

Xw

c

As mentioned before, in our simplified geometric model, the end effector will always remain in the horizontal position, with an angle m with respect to the a link: m = B + n + 90° = sin -1

(

b • sinC c

)

+ sin -1

() Xw c

From a more practical point of view, you may wonder which angles out of all this mess will be sent to the servos as inputs ... we have you covered! The servos will rotate from their zero positions (again, marked in red in Figures 4 and 5) with the input angles S (shoulder), E (elbow), and W (wrist):

// 2) Define servo object

void setup() { servo1.attach(9);

// 3) Attach servo on digital // pin 9

} void loop() { servo1.write(45);

// 4) Move servo to desired // angle

}

S = 90° – A – τ E = 180° – C W = m – 180°

Arduino millis() Function

In this simplified geometric model, the longitudinal axis of the wrist joint (defined as Wrist2, as you’ll see later) is not used, as the end effector will always stay parallel to the horizontal plane. We therefore have a spare DOF (degree of freedom) that won’t be used. That’s it for math class! Let’s start with some programming.

The Arduino Servo Library The Arduino script that I wrote for controlling the robot arm has been getting more and more complicated as I have been adding new functionalities (servo control, serial communication, initialization function, etc.). Consequently, in this article I will go through some extracts and samples of the script. The whole script with comments/explanations is available at the article link. A very easy way of controlling servos with an Arduino is using the Servo library. It supports up to 12 servos on the ATmega328 chip (the one we are using, and the one the popular Arduino Uno uses), and up to 48 on the Arduino Mega. The following sample code shows the four steps needed to control a servo with this library. Moving the servo to our desired angle is as simple as inputting a value in Step 4: #include

Servo servo1;

// 1) Include Servo library

The first problem I faced when I started programming and testing the servos with the Arduino was the servo speed. The servo travelled from its current position to the angle given from my input with such a high speed that it would destabilize the whole arm structure and possibly break it in the long term. My first solution was to use the delay() function and apply incremental or decremental angle inputs with for loops, so that the servos would move slower. It worked, until I tried to move more than one servo at the same time. Only once the first servo had reached its final position, could the second servo start to move! I immediately went to the Arduino forums and found out that delay() is a kind of forbidden function, as it totally blocks all the processing capacity of the chip, not allowing other tasks to be performed in the meantime. I could breathe again when I came across the millis() function: a powerful and popular function of the Arduino library that makes multitasking simple; millis() runs constantly in the Arduino main loop and keeps asking “How much time has passed?” instead of just stopping and waiting (like the lazy delay() function). Here is an example of the code I used in my project, for the case when the final angle (pos2) is higher to or equal as the initial angle (pos1): if ((pos2 >= pos1)){ if((millis() - lastUpdate) > updateInterval) {

SERVO 07.2017

25

Caja - Arduino arm par 2 - Jul 17_Blank Rough SV.qxd 5/30/2017 7:16 AM Page 26

Figure 6. Robot arm controller 18-character input string. lastUpdate = millis(); pos += increment; servo.write(pos2); } }

Depending on the defined value for the increment variable, the servo will move faster or slower. So now, not only can my robot arm control several servos at the same time in a slow and smooth fashion, but it can also perform other tasks! These include reading inputs (USB, Bluetooth, potentiometers, etc.) and processing new servo positions, among others.

Control Box As discussed in the first article, the control box has 6x 10K potentiometers, which are connected through a Micro JST 2.0 PH eight-pin connector to the analog inputs of the Arduino chip (pins 23 to 28). Each potentiometer controls a specific servo (except for the end effector, where two servos are controlled simultaneously by a single potentiometer). The following code shows how the input given by a potentiometer (pos) is mapped down to a desired range (posm); in this case, from 0 to 180 degrees: pos = analogRead(pot1); // Read analog input from potentiometer pot1 posm = map(pos, 10, 1023, 0, 180); // Map down the potentiometer input from its // original range (0 to 1024) to 0 to 180 degrees

The mapped down value of the potentiometer is then used as input of the servo angle.

that there are no spacers to separate one value from another, all the angles must always be given with three digits (0 degrees will be 000, 12 degrees will be 012, and so on). Although a bit tricky in the beginning, getting an Arduino connected with other devices over the serial port is a piece of cake! So, don’t panic if you spend a couple of hours racking your brains. I also went through that! The number one rule is to use the same baud rate in all the devices connected through the serial port (most commonly 9600 bps). The following chunk of code shows how the serial communication is implemented in the Arduino to obtain the Base angle (servo1) from the input string. Note that the elements used to include the servo library and define the servo1 object have been omitted for the sake of simplicity (for the whole code, see the Arduino script available at the article link): void setup() { Serial.begin(9600); // Open serial port and set data // rate to 9600 bps } void loop() { while (Serial.available()>0) { // Proceed only when there is data received delay(10); // Let’s give Arduino some time to take // a breath char c = Serial.read(); // Get one byte from // the serial buffer readString += c; // Read characters from serial // buffer into a string } if (readString.length() >0) { // Proceed if the string has any character servo1 = readString.substring(0, 3); // Get first three characters of string } // Now that we have the servo1 angle in string format, // we need to convert it into an integer number: char carray1[18];

Serial Communication The robot arm will communicate with other devices (PC, smartphone, etc.) through the serial port (USB or Bluetooth). Although at first my plan was to compute the inverse kinematics with the Arduino, in the end I have decided to compute it externally (i.e., with the PC) in order to simplify the Arduino code and avoid pushing the ATmega328 chip to the limit. Therefore, the input to the robot arm controller (in both Modes 2 and 3) is an 18-character string with the six servo angles (Base, Shoulder, Elbow, Wrist1, Wrist2, and End_effector). An example of this string is shown in Figure 6. Given

26

SERVO 07.2017

// Create 18-digit array to // store values of all servos

// Convert servo1 angle string into an integer number // (pos1): servo1.toCharArray(carray1, sizeof(carray1)); pos1 = atoi(carray1); sweeper1.Update(); readString=”“;

// // // //

Update sweeper function for servo1 (see whole script) Clear string variable for new input

}

We can now control the robot arm through the serial port (hooray!). We just need a couple of finishing touches and we are done!

Caja - Arduino arm par 2 - Jul 17_Blank Rough SV.qxd 5/30/2017 7:16 AM Page 27

Initialization Script As explained previously, the robot arm controller waits for four seconds after being turned on and checks for devices connected over the serial port. If no serial device is connected, Mode 1 (control box) will start; if there’s a serial device, Mode 2 (manual control with GUI) will start. This is done with an initialization script on both sides of the serial port: The robot arm electronic controller “listens” for any serial device, and the device connected through the serial port must send any data message to the robot arm controller (it’s important for this to be within the first four seconds!). This is implemented in the Arduino code as follows: int goto;

// Declare variable goto as integer // number

void setup() { Serial.begin(9600); // Open serial port and set data // rate to 9600 bps while(Serial.available()==0 && millis()<4000); // Wait for up to 4 sec. if no data is received if(Serial.available()>0){ // Proceed if there’s data received: goto = 1; // Set goto variable to 1 } } void loop() { if(goto == 1){ // If goto was set to 1: // THE CODE SHOWN IN THE “SERIAL COMMUNICATION” SECTION // IS IMPLEMENTED HERE } else{ // If goto was not set to 1: // THE CODE SHOWN IN THE “CONTROL BOX” SECTION IS // IMPLEMENTED HERE } }

The initialization script on the other side of the serial port (PC) will be shown shortly.

Graphical User Interface For years, I had heard wonders about Processing, so I decided to give it a try for creating a GUI. I couldn’t be more pleased! It took me just a couple of hours to start defining the layout of the GUI with Processing. Before that, I played around with some examples and did some research through online tutorials. The online community is not as huge as it is for the Arduino, but the similarity between both programming languages makes it very easy to get started with Processing. For the GUI, I needed six sliders to move the robot arm servos; a toggle button to switch between Modes 2 and 3; three text boxes to input the end effector coordinates (X, Y, and Z); and a button to submit these values. I came across the ControlP5 library (see Resources),

Figure 7. Processing GUI.

which provided all of that with a simple and straightforward implementation. The final GUI layout is shown in Figure 7. The Processing script performs four different functions: • • • •

Creation of GUI Initialization script Computation of inverse kinematics Transmission over the serial port of 18-character strings with the six servo angles

The code that follows shows how the control elements of the GUI are implemented, and how the string with the servo angle Base is transmitted over the serial port. For simplicity, only one element of each type is shown (slider, text box, toggle button, Submit button), as well as only one servo angle (instead of the whole 18-character string with all the servo angles). Again, for the whole code, see the Processing script. import processing.serial.*; // Import serial library import controlP5.*; // Import controlP5 library ControlP5 cp5; // Call controlP5 object Serial myPort; // Create object from Serial // class int Base; // Declare Base integer boolean Inv_Kin = false; // Declare boolean for // toggle button String X_coord; // Declare X_coord string void setup() { myPort = new Serial(this, portName, 9600); // Declare serial port myPort.write(1); // INITIALIZATION FUNCTION: Write something to // serial port size(700,400); // Define size (in px) of GUI // background noStroke(); cp5 = new ControlP5(this); // Needed for creating controlP5 objects PFont font = createFont(“arial”,20); // Define font type and size cp5.addToggle(“Inv_Kin”,false,290,20,20,20); // Define toggle button, inactive by default

SERVO 07.2017

27

Caja - Arduino arm par 2 - Jul 17_Blank Rough SV.qxd 5/30/2017 7:16 AM Page 28

cp5.addSlider(“Base”)

// Define slider for // Base

.setPosition(100,240) .setSize(40,100) .setRange(0,180) ; cp5.addTextfield(“X_coord”)

// Define text box for // the X coordinate

.setPosition(200, 80) .setSize(40, 40) .setFont(font) .setAutoClear(false) ; cp5.addBang(“Submit”) // Define Submit button .setPosition(260, 170) .setSize(80, 20) .setTriggerEvent(Bang.RELEASE) ; } void draw() { background(100,100,100);

// Define color of GUI // background

if(Inv_Kin==false){ // Proceed if the INV_KIN toggle // button is inactive String Base_s = nf(Base, 3); // Convert numerical Base value into string with 3 // digits myPort.write(Base_s); // Write Base_s (string) value to serial port } }

myPort.write(1); // Write 1 to serial port

The inverse kinematics are computed in the Submit() function, when the Submit button is pressed by the user (taking as inputs the space coordinates X, Y, and Z from the text boxes). As this is just the implementation of the trigonometric functions shown in the previous “Inverse Kinematics — Geometric Solution” section, you can have a look at the Processing script.

Bluetooth Connectivity As you may recall from last month’s article, the robot arm controller has an HC-06 Bluetooth board, which I decided to include in my design for experimenting with wireless connectivity. The implementation of serial communication through Bluetooth with the Arduino is pretty similar to the regular serial communication via USB. Since the HC-06 board is connected to the digital pins 10 and 11 of the Arduino chip (normally, it would be connected to the RX/TX pins, but these were already taken by the FTDI232 USB to serial breakout board), we need to use the SoftwareSerial library: #include

void Submit() { // The Submit() function will be called // whenever the Submit button is // pressed. X_coord = cp5.get(Textfield.class,”X_coord”).getText(); // Get value (string) of X coordinate int X = int(X_coord); // Convert value of X coordinate into integer (to be // able to make calculations with it)

The serial communication is initialized in the setup() function with the following command:

// INVERSE KINEMATICS ARE CALCULATED HERE (SEE WHOLE // SCRIPT)

In the rest of the script, the old Serial shall be replaced by BT; for example:

myPort.write(Base); // Write Base (integer) value to // serial port

SoftwareSerial BT(10, 11);

BT.begin(9600);

char c = BT.read();

}

It looks quite similar to Arduino code, doesn’t it? In the setup() function, the different graphic elements (background, sliders, text boxes, toggle button, Submit button) are created. The draw() function is comparable to the loop() function for the Arduino, as it loops consecutively which allows the program to change and respond to the user inputs. Note that despite having defined the sliders in the setup() loop, every time the user moves them, they provide a new value! However, the slider (float) values are later converted into strings in the draw() function (only in Mode 2, when the INV_KIN toggle button is inactive). The initialization script “on the other side” of the serial port consists of just a data byte that is sent through the serial port, and is implemented within the draw() function of the Processing code:

28

SERVO 07.2017

I have also included a second Arduino script with serial communication through Bluetooth in the download package. A future improvement that I have in mind is to merge both scripts (USB/Bluetooth) into a single one, so the user can choose to control the robot arm either through USB or Bluetooth, with some sort of initialization function. I have created another GUI for my Android smartphone with MIT App Inventor — an amazing online tool for developing Android apps. Everything is done in a visual environment, which means that you don’t need to write a single line of code. Just drag and drop blocks! For now, the GUI is quite simple (Figure 8) and only includes the six sliders to control every robot arm joint.

Load Arduino Scripts to the ATmega328 For loading Arduino scripts to the ATmega328 chip, I use an ISP (In-System Programmer) shield that I built a long

Caja - Arduino arm par 2 - Jul 17_Blank Rough SV.qxd 5/30/2017 7:16 AM Page 29

Figure 9. Homemade Arduino ISP shield. Figure 8. Android GUI designed with MIT App Inventor.

time ago (Figure 9) when I started to create my own printed circuit board designs for projects. It’s very simple to build one of these, and they’re also quite cheap if you want to buy one. I recommend you get one!

Final Thoughts and Considerations All in all, my project so far has required the practical application of a wide variety of engineering/scientific fields: mechanics, electronics, chemistry, programming,

Resources

Arduino official website: www.arduino.cc Processing official website: www.processing.org controlP5 website: www.sojamo.de/libraries/controlP5 MIT App Inventor: appinventor.mit.edu My blog: funwithcables.wordpress.com

mathematics, and — why not — project planning. Hey, this is robotics! In general, the whole project has been a sort of obstacle course with a lot of issues to overcome. The programming part has been especially challenging, and I had to spend a lot (a lot!) of hours developing the Arduino and Processing scripts. It’s very important to start testing simple code with individual elements, and then continue putting it all together like a jigsaw puzzle. For example, in this particular project, we use seven servos. So, it’s best to first make one or two servos work properly, and then apply that code to the whole set of servos. Also start writing the code and testing specific functionalities (e.g., servo control, serial communication). Once they work fine separately, put them together.

Next Steps As I mentioned before, the possibilities of adding functionalities to the robot arm are almost infinite thanks to serial communication. The only limit is your imagination! I still have plenty of ideas to implement in the future, such as robot vision with the addition of a webcam mounted on the end effector. This is just the beginning of a long adventure! SV SERVO 07.2017

29

Koci - Animatronics - Jul 17_Steve Koci Animatronics #1 Parrot.qxd 5/30/2017 7:20 AM Page 30

DIY Animatronics Shine a Light on Your Creations

By Steve Koci

What good is a fabulous prop if you can’t see it? Serious attention should be devoted to properly lighting your display. The effort you apply here can make or break your scene!

Planning Your Lighting Lighting should be a critical component in your planning. Allow sufficient funds in your budget to incorporate the necessary lighting. Developing a proper lighting scheme can be time-consuming as well, so set aside enough time to get it right! Having the ability to control your lighting will allow you to bring attention to the details you want to highlight. Just as importantly, it allows you to hide objects that detract from the story you are trying to portray. I recently had the opportunity to visit Disneyland with my kids and grandkids. For me, no trip to Disneyland would be complete without a visit to the Haunted Mansion. I always pay particular attention to how the characters fit into each scene. The séance room that includes Madame Leota who appears in the crystal ball has always fascinated me — so much so that I have included the illusion in my own display. In recent years, the scene was upgraded and the crystal ball no longer sits on a table. It now floats around the room with the image of Madame Leota perfectly aligned inside. My interest on this visit was centered on how the effect was pulled off. With great difficulty, I was able to barely make out the rod that supported and moved the crystal ball. To aid in the disguise, the rod was painted black. The real secret to pulling it off, though, was the meticulous attention that was paid to keep any of the numerous light sources from falling upon that rod. An understanding of how to properly integrate lighting into your scene will — in effect — allow you to “paint with light.” It is not always necessary to repaint your scene when looking to alter the feel. The ability to adjust and change your lighting scheme can be a powerful tool. The tone of your scene can be completely changed by altering the colors.

30

SERVO 07.2017

Let’s take a look at some of the essential components of a complete lighting system. You should take these into consideration when you plan the lighting to best suit your needs.

LED vs. Conventional Incandescent Bulbs The current trend is moving from incandescent bulbs to LEDs as their benefits outweigh the initial cost. This cost can be quickly recouped when you consider the significant savings in electricity used. The fact that LEDs have much lower power requirements allows you to maximize the number of lights on a circuit. The long life provided by LEDs is another significant advantage. I like not having to replace burned out bulbs! That is not to say that I do not use incandescent bulbs. I incorporate a variety of them mounted in clamp lights. They still Figure 1. My DIY RGBW floods. serve a valuable purpose in my display, and the use of the clamp lights allows for easy mounting. I am, however, transitioning into

Koci - Animatronics - Jul 17_Steve Koci Animatronics #1 Parrot.qxd 5/30/2017 7:20 AM Page 31

DIY Animatronics To post comments on this article and find any associated files and/or downloads, go to www.servomagazine.com/index.php/magazine/issue/2017/07.

more and more LED fixtures. If you will be incorporating LED light strings into your lighting scheme, you will want to pay attention to whether they are half wave rectified or fully rectified. Fully rectified LED strings are considerably brighter and do not flicker like the cheaper half wave rectified LED strings usually available at your local big box store. Some people can see the flicker that half wave rectified lights produce. If this is a concern for you, pay the price premium and go with the fully rectified sets. For those of you interested in wiring up your own LEDs, a handy calculator can be found at http://bit.ly/LEDwzrd. This wizard will help you design your setup and determine the necessary resistors required.

Just the Right Amount There is a fine line that must be walked when planning your lighting scheme. Too much light can be just as bad as too little. It can distract from your primary scene. You want to provide sufficient light to illuminate your display without overpowering it.

Static Lighting Certain situations lend themselves well to static or fixed lighting schemes. Such things as illuminating an information sign or pathway safety lights are a couple of examples where you will likely want a consistent light source. These lights will be turned on prior to your audience’s arrival and will remain on the entire time your show is active. These lights set the starting point for any additional lighting. I like to take care of this element first and layer my other lighting effects on top of it.

Flood vs. Spotlights One of the biggest decisions you will have to make in developing your lighting plan is the type and number of lights you will need. Most displays will benefit from a combination of flood, spot, and pin spotlights. This choice offers its own set of challenges. Space, budget, venue, mechanism type, and how it fits within the rest of the scene are all important considerations. You may need to consider in your planning the impact of any fixed light sources such as street lights. The difficulties these pose can be significant and take some creativity to overcome. Rendering street lights inoperative is not an acceptable solution! For my display, I decided to include 16 DMX flood lights for my primary light source (Figure 1). They each include a total of 60 LEDs: 15 each in red, green, blue, and white (Figure 2). Used in conjunction with my DMX controllers, these allow me

Figure 2. Yep, there really are 60 LEDs inside!

to create virtually any color I desire! I have yet to settle on a favorite spotlight. I primarily rely on battery powered units that are easy to place where needed (Figure 3). This frees me from the restrictions posed by running electrical lines. However, it does mean I need to constantly purchase and install fresh batteries. The placement of my pin spots is another area where running wires is not an option. I have modified small AAA flashlights for this purpose (Figure 4). By using lithium batteries, I am able to get a full night of operation out of a single battery. That is a small price to pay for highlighting the many small details that make up a display. The inclusion of auxiliary lighting sources such as lanterns and flickering candles can add interest as well. These are not primary light sources, but they can be that Figure 4. Single AAA pin added detail that spots. compliments your design.

Figure 3. 9V battery powered spots.

SERVO 07.2017

31

Koci - Animatronics - Jul 17_Steve Koci Animatronics #1 Parrot.qxd 5/30/2017 7:20 AM Page 32

DIY Animatronics RESOURCES LED Calculator — http://bit.ly/LEDwzrd Voltage Drop Calculator — http://bit.ly/Voltdrop Monoprice — http://bit.ly/mnpric Kill-A-Watt — http://store.nutsvolts.com/projectkits/sku16502 DIY LED Tutorial — http://bit.ly/ledsys Lights Alive — http://bit.ly/litealv Fright Light Illuminations — http://bit.ly/frtlte Mini Spotlight — http://bit.ly/minspt Gantom — http://bit.ly/gntmlt DIY Christmas — http://bit.ly/Clight My YouTube channel — http://bit.ly/Halstaff My Website — http://bit.ly/Hauntechdiy DIY Animatronics Forum — http://tinyurl.com/qjeehjs

Light Placement I generally prefer to place my flood lights down low as it creates wonderful shadows on the scene backdrops. The use of shadows can provide additional layers of interest and depth to your display. I then place the spotlights to highlight individual components and draw my guest’s attention right where I want it. These lights work in concert with the floods and create an interesting blend of lights. If I am including a thunder and lightning effect, the required lights are placed up high; preferably in a tree or shining through netting to create additional shadows. Pay particular attention to how all the lights work together. The object is to blend the lights and create a balanced and cohesive light scheme.

with more information on the subject can be found in the Resources sidebar. If in doubt of which size to use, go bigger!

Power You can choose to power your lighting with either batteries or a dedicated AC service. While batteries allow you to “sever the cord” and place your lights wherever you please, they do have some serious drawbacks and limitations. It can be costly and time-consuming to continually replace discharged batteries. If your program runs for an extended period, you may find your lights dimming and eventually going out before your show is finished. If you are looking for batteries that will provide the longest continuous light supply, I have found that lithium batteries fit the bill. AC power seems to fit the requirements the best, however, as it offers a stable, consistent, and economical supply of power. My primary lights which require the most power are all plugged into AC circuits. Another important consideration is the adequacy of the circuits that are providing the power. An extremely useful tool for monitoring the loads placed on individual circuits is the Kill-A-Watt electricity usage monitor (see Resources). This is an integral tool for me when I’m laying out my power distribution lines.

Emergency Lighting When planning your lighting requirements, do not forget to include some fixtures that will be used as a separate emergency lighting system. I recommend that these be connected to a separate dedicated circuit and be controlled by a master switch. In an emergency situation, you want to be able to quickly enable all of these lights at once to provide a safe and well lit exit route.

Colored Gels It is not always necessary to change out the color of a fixture when modifying your lighting. Colored gels are available that can be affixed to the front of your unit to allow you to change colors is a flash! I have found that photographer’s sample packs are a handy and inexpensive source of lighting gels for my mini spotlights. It is as simple as cutting one to size and attaching it with a rubber band. Use caution when applying them to lights that create a lot of heat as this could cause a safety issue.

DIY Lights and Controllers

Wiring It is important that you choose the right wire and size for your application. Making an improper choice can be hazardous and lead to a fire or shock. A link to a good voltage drop calculator

32

SERVO 07.2017

Figure 5. DMX central control.

Controlled lighting using a DMX system that is synchronized with your prop activations can take your project to the next level. I appreciate the ability to script the lighting for each independent area of my show. I’m still not using my system to its full capacity, but I work on improving it each year. Whenever this subject comes up, it always ignites the debate on the merits

Koci - Animatronics - Jul 17_Steve Koci Animatronics #1 Parrot.qxd 5/30/2017 7:20 AM Page 33

DIY Animatronics of having a central lighting system. Having a centralized system does allow you to quickly get your show up and running, but if it fails your entire show goes dark. My solution is to use a combination of DMX controlled scenes as well as independently activated and lit characters (Figure 5). In the unfortunate event that my computer controlled system crashes, enough other characters are still available to entertain my guests until repairs can be made.

each spotlight. Each LED spotlight is wired in series with Figure 6. the light controller. This allows Stephen’s DIY individual control of each lighting setup. spotlight, which includes six levels of dimming and three modes (strobe, flicker, and steady). Settings are maintained by the controller when the power is turned off, so there is no need to reprogram it every time. The light fixture is made from a 3” piece of 1/2” PEX water pipe which has one end flattened using a heat gun (Figure 8). The light mounts using hardware store L-brackets and the position adjusts via a screw and wing nut. The A cool Do-It-Yourself connecting wire is made system has been developed using cannibalized RCA by my good friend, Steven cables. Colberg. He was kind enough RCA cables are an to share the details of his inexpensive source of twoeasy-to-construct design conductor wire with (Figure 6). This project has connectors. My favorite found its way to the top of source for these as well as my “to do” list! any other cable needs is This DIY LED lighting Monoprice (see Resources). system is based around one In order to help you Figure 7. LED strip light controller. watt 10 mm LEDs in standard construct one for yourself, dome packages and LED strip Steven has provided a light controllers (Figure 7). Each spotlight produces about comprehensive step-by-step tutorial on his website (see 250,000 to 400,000 mcd. In comparison, a standard 5 mm Resources). Check it out and build your own! ultra bright LED typically produces less than 10,000 mcd. Each assembled light costs about $3 to make (not including the required RCA cable). Attending conventions provides me with one of my The system runs at six volts to reduce the amount of favorite opportunities for developing new ideas. While heat dissipated by a 10 ohm two watt resistor located in

A DIY LED Spotlight System

New and Exciting!

Figure 8. A super simple yet effective design for this spotlight.

Figure 9. Transworld brings all the big players to town!

SERVO 07.2017

33

Koci - Animatronics - Jul 17_Steve Koci Animatronics #1 Parrot.qxd 5/30/2017 7:20 AM Page 34

DIY Animatronics Figure 10. Lights Alive showcasing their products.

Figure 11. FrightLight bringing black lights to life.

(Figure 11). The reasonable use of backlights can add a truly unique look to your scene. They have extensive experience in the theme park industry and specialize in UV/black light solutions. Their budget-friendly lights will surely draw plenty of attention (see Resources). A company that specializes in LED lighting products is Minispotlight.com (Figure 12). Not only do they offer a wide variety of LED devices, but they also can create specialty custom lighting creations (see Resources). One of the most interesting and potentially useful products has been developed by the folks at Gantom Lighting & Controls (Figure 13). The recently released Precision DMX Flood is a microprocessor based intelligent Figure 12. Minispotlight and some of their cool toys. fixture and is designed for indoor or outdoor use (Figure attending the Transworld Halloween & Attractions show in 14). I am especially interested in the exciting feature which St. Louis this year, I had the opportunity to visit with several allows it to be configured with the free DMX Programmer lighting companies (Figure 9). They carry a wide variety of app and then operated in stand-alone mode. The address lights and lighting effects that will be useful when and programming is retained onboard the fixture and will displaying our characters. operate independently of the programmer. These include the folks at Lights Alive (Figure 10). They An added benefit is that it is also triggerable, which carry a complete line-up of lighting products, including allows for increased flexibility. The easy-to-use app includes some fantastic lightning simulators, lanterns, lasers, and a selection of built-in functions including flicker, strobe, many other special effects (see Resources). blink, fade, color change, dimming, and color macros Another interesting option is provided by FrightLight (Figure 15). This technology has a huge potential for our community. It greatly simplifies the process of setting up a scripted lighting plan. Using these fixtures frees you from having to purchase and program a separate DMX system. The cost and time savings is a game changer! These fixtures are so new that the mounting bracket I have is still a prototype. Hopefully, the production models will be available by the time this hits the streets. If not, I’m sure they will be ready soon afterwards. The folks at Gantom have created an outstanding series of Figure 13. Always plenty going on in the Gantom booth!

34

SERVO 07.2017

Koci - Animatronics - Jul 17_Steve Koci Animatronics #1 Parrot.qxd 5/30/2017 7:20 AM Page 35

DIY Animatronics Cutting Acrylic Sheet In addition to writing this monthly column, I also travel to Halloween conventions to speak on the process of building animatronics. While I prefer to bring my large characters to demonstrate, the cost of transporting them is often prohibitive. Therefore, I have a head and torso that can be disassembled so that it will fit in my suitcase (Figure A). In order to allow all the mechanisms to be fully visible, I decided to construct the headpiece out of Plexiglas (Figure B). I have been successful cutting straight pieces with the score and snap method. However, this application required curved cuts which were something new for me. I decided to use my scroll saw and a fine blade. I thought that as long as I cut slowly, this would work. WRONG! The material heated up sufficiently that it melted together behind the cut. When you are no longer seeing waste being expelled from your cut, it is a good indication that the material is melting instead Figure A. Actobotics, Hitec, of cutting. It was time to go back to the drawing and Actuonix make it work! board. Figure B. Plexiglas face piece allows I had some success by applying WD40 to the everything to be seen. cut much like I would do when cutting steel. Although it did help, I found that I got even better results by adding a layer of blue painter’s tape to each side of the Plexiglas and then transferring my cutting lines onto it. This provided a clean cut and worked perfectly! I can now add this skill to my toolbox for the next project requiring curved cuts in Plexiglas!

useful YouTube videos detailing the setup and operation of their products. You can find them at http://bit.ly/tmgntm.

Want to Know More?

Figure 15. Using the Color Piano is a snap!

If you would like to explore this subject in much greater detail, a fantastic resource is available through the Christmas lighting community. My favorite is DIYChristmas.org, but there are plenty of others out there as well (see Resources). These various groups specialize in lighting effects and have revolutionized the home lighting craze. They offer an assortment of different DIY lighting solutions, as well as provide the option of purchasing complete packages if that is more your style.

Figure 14. The revolutionary new Precision DMX Flood.

Time to Turn Out the Lights! Properly illuminating your creations is a critical component in the creation of a top notch scene. It allows your guests the opportunity to fully appreciate all the hard work you have put into the design and construction of your masterpiece. We sometimes get caught up in the mechanical aspects of a build and overlook some of the other details. As in photography, proper lighting is often the secret to success! If you are looking to continue this discussion or have any questions regarding animatronics, please join us on the DIY Animatronics forum at http://bit.ly/DIYfrm. Until next month, MAY THE PASSION TO BUILD BE WITH YOU! SV SERVO 07.2017

35

The Wonderful World of Robot Mastermind, Tony Ellis

Prochnow - EZ1 robot - Jul 17_Blank Rough SV.qxd 5/30/2017 9:26 AM Page 36

36

SERVO 07.2017

Prochnow - EZ1 robot - Jul 17_Blank Rough SV.qxd 5/30/2017 9:27 AM Page 37

Roboticist, Inventor, and Entrepreneur, Tony Ellis does it again — he’s created a home-based autonomous robot that puts all others to shame.

P

By Dave Prochnow

To post comments on this article and find any associated files and/or downloads, go to www.servomagazine.com/index.php/magazine/issue/2017/07.

repare yourself for yet another British invasion. No, this time it isn’t an attempt to quell a rebellious tea party uprising nor is it a moptopped musical foursome trying to steal your teenage daughters’ hearts. Rather, this time it’s a technology invasion. Similar to the appearance of the Raspberry Pi and the BBC micro:bit, prepare yourself for a tidal wave of technology crashing onto our shores in the form of British robots. Huh? You thought that the only “good” robots come from the labs in the US and the factories in Asia? HA! Britain will soon be able to lay claim to being the undisputed champion for creating affordable robotic technology that is nothing short of incredible. This world of technological achievement rests squarely on the shoulders of one Atlas-like man: Tony Ellis.

That Was Then

Astute long-time readers of SERVO Magazine should recognize Ellis from his remarkable work on Vintage: a tone-programmable wood and metal robot thought to have been built in and around 1979. A description of Vintage (see Figure 1) was featured in the January 2004 issue of SERVO Magazine. Designed with an aesthetic resembling the Huey, Dewy, and Louie droids from the Bruce Dern classic sci-fi movie, Silent Running, Vintage housed an automobile stereo cassette deck in its chest along with a 16-channel RC system for programming, playing, decoding, and maneuvering itself via musical tones — mop-topped musical foursome, indeed! Through today’s eyes, Vintage looks whimsical, but it’s earned its place in the robot Hall of Fame. Furthermore, programming a robot via audio tones is not only laborious, it is also downright brilliant. In fact, these skills for “thinking way outside the box” enabled Ellis to push the robot design envelope to its limits with his most recent incarnation: a tour de force effort named ALTAIR.

This Is Now!

ALTAIR — or more specifically, Affordable Latest Technology Artificially-

Intelligent Robot — is actually an upgrade from another of Ellis’ successful designs: the PIC-based AIMEC robot series. You might recall the AIMEC robot; it was featured in several media outlets. The AIMEC:3 robot won the spotlight on Russian news, was interviewed on BBC TV, and even read the nightly sport scores on BBC radio. While Ellis gets most of the human media attention, he isn’t some intellectual lone wolf who does more in one year than most of us do in a lifetime. No, Ellis is actually a cog — albeit a massive vitally important cog — in a Sussex, England company called Applied Machine Intelligence (AMi; www.appliedmachineintelligence.co.uk). Supporting Ellis is an equally incredible team of talented contributors: Judie Ellis (co-director, inventor, wife); Mike Hodgson (programmer); Steve Carpenter (inventor, entrepreneur); Steve Cole (model maker); Bill Keller (computer scientist); and Ian Ozsvald (data scientist). The goal of AMi is to develop topnotch kickass robots for the fields of education/home use and developmental robotics. After just a cursory review of their Scientific, Technology, Engineering, Mathematics (STEM) ALTAIR EZ:2 and ALTAIR EZ:1 (and, AIMEC:4) robots, you can tell that Ellis and crew didn’t just meet their design goals, they shattered them!

Figure 1. Tony Ellis’ Vintage robot.

This Is Your Future

The ALTAIR EZ:1 is a four foot tall robot with advanced vision (including face and object recognition) and emotion/ gender/age detection. This robot is a sensory-rich platform with incredible dexterity via two 5DOF (degrees of freedom) arms (see Figure 2). Ellis calls ALTAIR a CMS type robot: Collaborative — Mobile — Social robot. In this context, he uses the terms: collaborative, due to the bot’s dexterity to achieve things while working with humans and other robots; mobile, because ALTAIR can move around freely; and social, based on the ability of ALTAIR to communicate in a human-like manner. Inside the ALTAIR robots, AMi uses the EZ-Robot EZ-B v4/2 controller with custom software, making a formidable “brain” for these bots. A P2 vision sensor gives high accuracy face recognition; emotion, gender, and age estimation; plus SERVO 07.2017

37

Prochnow - EZ1 robot - Jul 17_Blank Rough SV.qxd 5/30/2017 9:28 AM Page 38

Figure 2. Now this is the robot we all wish we could build! This EZ:1 model is bristling with sensors, actuators, servos, and good bot vibes. face and gesture tracking. Object, color, and glyph recognition are obtained via the EZ-Robot v2 camera. Other sensors include a thermal imager, long-range IR ranger, light level, right/left passive infrared (PIR), sound localization microphones, fourzone ultrasonic ranger, electromagnetic sensor with 360 degree sense scan, cliff/fall sensor, battery monitoring, and right/left claw object detector. All smart servos have (programmable) current sensing. (More about these smart servos in a bit.) Moving the ALTAIR robot is achieved via 2x high torque globe motors with high-res HEDS encoders; 10x worm driven smart servos and 2x linear actuators; and 2x worm driven smart servos for pan/tilt. Supporting the main EZ-B controller are 16 custom PIC microcontrollers. These PICs control: emotion animation; head master

38

SERVO 07.2017

Figure 4. An AMi “smart” servo — a custom-designed high torque unit. control (head sensors buffer); shoulder servos, bicep servos, elbow servos, wrist servos, claw servos; locomotion (high-res) odometry and location mapping/memory; and locomotion based sensors (the interface). The PIC master microcontrollers (head and locomotion) gather raw data from all their respective sub-system microcontrollers and filter out repetitive and non-relevant data, only sending on to the EZ-B valid and fresh data. This intelligent filtering greatly reduces any communication bottleneck and eases processing Figure 3. The ruggedized teach pendant.

Prochnow - EZ1 robot - Jul 17_Blank Rough SV.qxd 5/30/2017 9:28 AM Page 39

Figure 5. ALTAIR meet Pepper, Pepper meet ALTAIR. (Okay, Pepper, who’s your Daddy now?)

Figure 6. A promotional banner for ALTAIR with a moving homage to its “vintage” Silent Running roots.

overhead on the EZ-B controller. The locomotion master control is guided by a teach pendant RF interface (see Figure 3). This is a ruggedized handheld device that simplifies “driving” ALTAIR around via either fixed-definition movement angles or a joystick. In case you’re wondering, the ALTAIR EZ:2 is the proposed educational/commercial version of the EZ:1 development robot. Should you expect to see either of these ALTAIR robots begging for money on a crowdfunding website? According to Ellis, “I am not sure crowdfunding will work for me. To properly commercialize the ALTAIR robots, it’s going to take an awful lot of money, and I believe that investors are the best option. I have self-funded my previous robot projects and will continue until the ALTAIR robots are developed enough to attract funding interest.” Like most brilliant people, Ellis makes robot design look easy. However, getting from point A to point Z hasn’t always been a piece of cake. In fact, Ellis claims, that “after years of experimentation, the design has evolved and I learned how to do things better. For instance, regular high torque servos were very troublesome in a number of ways, so I decided to design my own servo. I developed worm driven ‘smart’ servos that have high torque and run virtually silent with zero noise and holding current.” These smart servos (like the one shown in Figure 4) are used in the arm and head pan/tilt mechanisms. Furthermore, Ellis is a little bit like you and me. He says, “To build a useful robot needs so much knowledge that

takes years to obtain for an individual — electrical systems, electronics, mechanics, software and firmware; also 3D design and general engineering. I have had to learn all these skills over the last 40 years. In this time, I have been able to learn enough skill in each category to now (hopefully) make truly unique and useful robots.” That kind of modesty should be instilled into more of our robot makers and robots (see Figure 5). So, where is Ellis and company going from here? Ellis summarizes his future plans, as his “next steps are to continue development; to further improve functionality to a point where the robots have true commercial value (become actually useful to their users). I am also looking into the educational sectors where a low cost (feature rich) modular robot platform like the EZ:1 could be of value to STEM curriculum and universities.” Figure 6 shows a banner advertisement for promoting ALTAIR. We wish you luck, Tony. SV

SERVO 07.2017

39

Hewitt - Hacking a power wheelchair - Jul 17_Blank Rough SV.qxd 5/30/2017 7:23 AM Page 40

Hardware Hacking a Power Wheelchair Control Module On the surface, using a power wheelchair as a robotic base seems like a no-brainer. The control modules for these chairs contain all of the necessary electronics to drive the motors, including DC/DC converters, PWM generation, and a microcontroller to govern acceleration, deceleration, and braking. Before I knew better, I purchased a power wheelchair thinking that it would be a simple task to wield the functionality of the electronics already built into the control module. However, as anyone who has attempted this knows, available documentation for wheelchair control modules is virtually non-existent. As a result, all of the previously-reported wheelchair robots I read about online bypassed the original control module altogether, used separate PWM motor drivers, and typically disabled the electric brakes on the motors. These solutions work well, but I couldn’t force myself to abandon the well-designed control module that I had at my disposal just because the make-up of its guts was elusive. Thus, began my experience of hacking a wheelchair control module. 40

SERVO 07.2017

Hewitt - Hacking a power wheelchair - Jul 17_Blank Rough SV.qxd 5/30/2017 7:23 AM Page 41

By Darby Hewitt

To post comments on this article and find any associated files and/or downloads, go to www.servomagazine.com/index.php/magazine/issue/2017/07.

Figure 1. The Pride Jet 3 power chair I found on eBay.

T

his project began when one of my students approached me at the beginning of summer 2016 with the desire to create a robot that could roam around our department, talk to students, take pictures to post to a Twitter account, and ultimately provide a scalable robotic platform for students to build upon. Because this robot could be subjected to some wear-and-tear as a result of its frequent interaction with students and faculty, I immediately thought of using a power wheelchair as its base. Relatively inexpensive power chairs are easy to find on eBay (if you are willing to go pick them up), and my student and I found one for $130 that was within driving distance: a Pride Jet 3, with a Penny+Giles Pride control module (see Figure 1). Once we purchased new batteries (for another $120), we thought we were in business, but the conversion from chair to robot was a bit more challenging than we first anticipated.

Figure 2. Inside view of the Penny+Giles Pride control module that came with my used power chair.

Opening the Module and Fomulating a Game Plan We started by taking a look under the hood of the control module, which — again — amazingly houses all of the electronics for power, motor driving, and electric brake control. The photo in Figure 2 shows an inside view of the control module. It may be difficult to tell, but the module is tightly packed; there is hardly any unused space inside. The top of the box holds the bargraph power indicator, the power button, and the joystick, along with several PCBs (printed circuit boards) which serve to interface these components to the rest of the module via a ribbon cable and a bundle of two small wires. The dense packing of the electronics in the module was overwhelming at first. However, since the joystick is the most direct way of controlling the motion of the chair, that seemed like the most logical place to begin our SERVO 07.2017

41

Hewitt - Hacking a power wheelchair - Jul 17_Blank Rough SV.qxd 5/30/2017 7:23 AM Page 42

modulated (PWM) digital signals is a common technique for generating analog output signals from a digital device, and when analog output is needed from a microcontroller platform without a built-in DAC (digital-to-analog converter) peripheral (such as the ATmega328P at the heart of the Arduino Uno, for example), this is often the most logical choice. Things can get complicated, however, when the voltages involved don’t lie within the 0-5V range. Also, even after filtering PWM signals with a capacitor, there can be some residual high frequency noise that rides on top of the Figure 3. Schematic of a simple resulting analog output signal. In this case, voltage divider with two resistances. since I knew we were going to need to (at least) provide an offset voltage for a filtered investigation of the inner workings Figure 4. Schematic of a slightly more PWM ouput signal and I wasn’t sure how of the control module. The joystick complicated variable voltage divider module itself is housed in a metallic with three resistances, one of which is sensitive the control module would be to a potentiometer. The output voltage noisy control voltages, I scrapped the idea of cylinder (shown prominently in from this circuit is in the range of 5-7V. PWM in favor of a more elegant and Figure 2) which is mounted on the platform-agnostic solution. top of the control module. A bundle The easiest, quickest, and least reliable way to create a of five different colored (black, green, red, yellow, and blue) variable voltage is through the use of a voltage divider. A wires connects this module to the rest of the electronics in voltage divider is nothing more than a series of resistances the box. Each of these wires fortuitously connects to a across which a voltage is applied. Figure 3 shows a solder pad on the PCB that houses the power button and schematic of the simplest form of a voltage divider: two the LED indicator bargraph. Using these pads as contact resistors in series. In this circuit, the output voltage is points, we were able to determine the purpose of each of related to the input voltage and the resistors according to these wires through the use of a digital multimeter (DMM). the relationship: Our initial guess (which turned out to be correct) was that the black wire provided a circuit ground connection for R1 the joystick module. So, we connected the ground lead of Vout = x Vin R 1 + R2 our DMM to the black wire and probed the voltages of the other wires with respect to that node. We found that the This equation is known as the voltage divider rule. I red wire was held at a constant 12V, and that the green won’t bore you by proving that it works; if you have wire was held at a constant 6V with respect to the black doubts, Google it. wire. The voltages of the other two wires varied according My first thought when designing the circuit for these to the motion of the joystick. control lines was to use a voltage divider with two resistors When the stick was at rest, both the blue and yellow and a potentiometer (shown in Figure 4) to generate the 5wires were held at 6V. However, when the joystick was 7V range needed for each channel. Voltage dividers do not rocked from fully backward to fully forward, the voltage of provide stable voltages when they are loaded down (as the yellow wire varied smoothly from 5V to 7V. Similarly, would be the case if — for example — I was going to try to when the joystick was rocked from left to right, the voltage drive a DC motor between the output voltage from this of the blue wire varied smoothly from 5V to 7V. circuit and ground), but if they are used to provide a Discovering the behavior of the yellow and blue wires voltage to a high impedance input (which I correctly was a breakthrough in the wheelchair hacking process; all it assumed was the case for this control module), then they takes to harness the power of the chair is to emulate the are sufficiently stable in most cases. So far, so good. behavior of these two wires. If we could figure out a way At this point, I just needed to find a digitally to control these two analog voltages using a standard controllable potentiometer to place in my circuit of Figure 4 digital platform, we knew we would be in business. that could interface easily with a digital platform. There are many devices out there that seem like they would fit in this circuit just like a mechanical potentiometer would, but I learned the hard way that digital pots are not always as flexible as their physically tangible counterparts. Generating and low pass filtering pulse width

( )

Digital Generation of Analog Voltages: First Steps and Mishaps 42

SERVO 07.2017

Hewitt - Hacking a power wheelchair - Jul 17_Blank Rough SV.qxd 5/30/2017 7:24 AM Page 43

I chose to use a Microchip MCP4251 for my circuit (the datasheet is available at the article link). The MCP4251 is a dual-channel SPI-controllable digital potentiometer which comes in 5, 10, 50, and 100 kΩ varieties. I chose to use the 10 kΩ version, of course, in light of the resistance values I had used in Figure 4. Using my trusty protoboard, I wired up this simple voltage divider, and I wrote some quick-and-dirty Figure 5. Schematic of the variable Figure 6. Schematic of a simple summing amplifier voltage divider used in this project. Arduino code to communicate with circuit. The output voltage from this circuit is The output voltage from this circuit the potentiometer via SPI and vary proportional to the weighted sum of the input is in the range of 0-2V. voltages. the output voltage sinusoidally. Then, I connected the output voltage to my oscilloscope, turned on the power, and watched the screen. Nothing happened. At first, I assumed my problem was related to the SPI commands I was sending. (Did I misread the datasheet?) After double- and triple-checking, I decided that the commands should work. To check this conclusion in another way, I disconnected the voltage divider from the power supply, set my DMM to resistance mode, and started checking the resistive properties of the digital multimeter. The resistance across the entire device was measured at 10 kΩ. So far, so good. Next, I checked the resistance between the wiper and one of the ends of the pot. I saw a large oscillating variation from 0 to 10 kΩ. At this point, I Figure 7. Schematic of the voltage divider from Figure 5 was stumped. The potentiometer was behaving exactly as I incorporated with a summing amplifier like that in Figure 6. This is the circuit used to control a single degree of freedom for the expected, and yet the output voltage from my circuit was wheelchair robot. The digital potentiometer is boxed in gray. not varying when the power supply was connected. It didn’t take too much digging into the datasheet for in Figure 6. The output voltage from this amplifier is a the MCP4251 to discover my mistake. The voltage applied weighted sum of the two input voltages, V1 and V2, to the wipers of these digital potentiometers must be defined according to the relationship: between Vss (0V in my circuit) and Vdd + 0.3V (5.3V in my circuit). In contrast, I had attempted to apply 5V and 7V to Rf + Rg R2 V1 + R1 V2 the ends of the potentiometer. Vout = x Rg R1 + R2 Rats! It was time to modify my initial circuit.

( )(

The Solution: Another Simple Circuit to the Rescue Since I was able to vary the wiper position of my digital pot as desired, I knew that this device could still be useful in my final design. I could easily create a voltage divider to provide a variable 0-2V output voltage by applying 5V across a 15 kΩ resistor connected in series with my 10 kΩ pot, as shown schematically in Figure 5. Prototyping this circuit was a breeze, and it produced the desired results. Once this circuit was completed, I just needed to apply a 5V DC offset to its output, which I knew I could do with another simple circuit: a summing amplifier. The schematic of a basic summing amplifier is shown

)

If all resistances in this circuit are equal, Vout = V1 + V2. Putting the voltage divider from Figure 5 together with this amplifier produced the circuit shown in Figure 7, where R1, R2, Rf, and Rg from the circuit of Figure 6 have been replaced by 10 kΩ resistors. This circuit ultimately produces a range of 5-7.7V at the output of the op-amp. We could increase the 10 kΩ resistances to bring the upper limit of the output voltage range closer to 7V, but we can easily circumvent this upper limit when we write software to control the wiper position for the pot, thus controlling the output voltage. Given this solid design, we knew we just needed two of these circuits to take control of the translation and rotation of our power chair base. SERVO 07.2017

43

Hewitt - Hacking a power wheelchair - Jul 17_Blank Rough SV.qxd 5/30/2017 7:24 AM Page 44

Putting the Pieces Together and Warming Up the Soldering Iron

connection (just in case), which I found while poking around with my DMM. Finally, I placed IC sockets and resistors in place on my protoboard and soldered the circuit together, connecting the 12V, ground, and control lines in their appropriate places. While a number of digital devices could be used to control the wheelchair through this circuit, I chose to initially use a Raspberry Pi. Given the fairly substantial current requirements of the Pi, I decided to also incorporate an L7805 5V 1.5A DC/DC regulator on the board to tap 5V off of the 12V supply from the chair. This seemed like a safer bet for powering the Pi than using the 4.8V connection I found, since I wasn’t sure how much power I could eek out of this regulated voltage from the chair, and I knew that the Pi really requires 5V at a minimum anyway.

To realize this circuit in hardware, I used an LM358N IC (datasheet also available at the article link) which is a singlesource dual op-amp. Whereas op-amps typically require equalamplitude positive and negative voltage sources with respect to circuit ground in order to function correctly, single-source op-amps allow for a singleended power connection, as long as the output voltage remains between ground and +Vcc. I chose this device because I knew I would be using these opamps to produce a positive range of voltages, and I knew I would have easy access to the +12V power of the chair. Using my trusty breadboard, DC power supply, DMM, and some simple Figure 8. The prototype, soldered together on perfboard. It's not pretty, but it works! Arduino code (at the article link), I built both channels of the circuit and successfully put each through its paces. The ultimate result of all of this work is shown in With a working prototype in hand, it was almost time Figure 8. Admittedly, I was rushing to get a quick-and-dirty to get out some protoboard and solder it up. First, however, prototype together, and while the board looks ugly, it works some slight modding of the control module itself was in great. One of my not-too-distant goals is to rebuild this order. As I mentioned above, I needed to tap into the circuit on a Raspberry Pi prototyping HAT, so it will sit rigidly power of the chair in order to power my circuit. Also, I on the Pi rather than being tethered by a bundle of wires. needed access to the two control lines from the joystick. As for software to control the resulting robot, we have Furthermore, I wanted to maintain the ability to put the created a simple Python class to represent the robot and to chair back into manual control mode. encapsulate its motion capabilities (see Robot.py in the So, I mounted two SPDT switches in the housing of the Github repository at https://github.com/ACURobotics/ control module: one for each channel. I desoldered the darbot-code.git). My plan is to transition to using the control lines from their pads on the power button PCB, and Robot Operating System (ROS) on the Pi to provide a nice resoldered them to the switches. I then soldered wires from software platform for expanding this robot’s functionality. the center taps of each of the switches to the pads This project has resulted in a sturdy reconfigurable originally directly connected to the control lines. At this platform that can be unleashed on a crowd without point, I had the ability to disengage the control lines worrying about it getting broken. My hope is that it can be completely. used to conduct robotics research (in addition to taking The last step in modifying the control module was to pictures at department events). connect several leads to the power and control lines of interest so they could be accessed outside the box. I drilled Ultimately, we invested under $500 for the chair, a small hole in the side and passed five wires out of the batteries, electronic components, and Raspberry Pi, and module: the two joystick control lines (connected via the with some basic knowledge of electric circuits we were able open terminals of the two toggle switches), 12V and to turn that investment into something awesome. SV ground (0V) connections, and also a regulated 4.8V

It Works! What’s Next?

44

SERVO 07.2017

Peavy - RoboGames - Jul 17_Blank Rough SV.qxd 5/30/2017 7:26 AM Page 45

RoboGames: The Original Robot Amateur Hour This is not some slick trade show with suits, ties, and over-the-top corporate sales booths. It’s not a Hollywood production either, with predictable scripts and special effects where everything just works out. RoboGames (the Olympics for robots) is the biggest “open” robot event in the world. Of course, the big draw is and always has been the combat robots. However, in addition to loud bangs, crashes, smoke, sparks, fire, and mechanical mayhem, there are dozens of other events at RoboGames from Humanoid robots, ArtBots, Soccer robots, Fire Fighting robots, RoboMagellan, Best of Show, and more. It’s “Amateur Hour” and we love it ... the friends, the robots, the medals, the competition, and the chaos. By Camp Peavy

To post comments on this article and find any associated files and/or downloads, go to www.servomagazine.com/index.php/magazine/issue/2017/07.

arrived Friday afternoon anticipating the HomeBrew Robotics Club’s event, “Table Top Navigation.” This event is modeled after our own “TABLEBot Challenge.” In the final phase, robots move a block into a shoebox mounted at the end of a table. There were only four competitors this year, but we all knew each other and bragging rights were on the line. In the end, Chris Mayer (USA) with his artificially intelligent MotBot won Gold, mesmerizing the judges with its tabletop prowess. Marco Walther (Germany) took Silver with his high-tech TABLEBot Dexter, and Al Margolis’ (USA) VNAVS1 (which was originally designed for RoboMagellan) took Bronze. I guess the judges really like big bots! For me, I didn’t totally embarrass myself. I scored a few goals (with some coaxing) and didn’t fall off the table. Saturday was a beautiful sunny spring day as opposed to last year when it rained. My primary job today would be RoboMagellan. This is an event where a GPS coordinate is specified and an 18” OSHA orange “Goal Cone” is placed on the spot that is the final destination. Bonus Cones are also specified, which give competitors a fractional multiplier that decreases the robot’s timed score.

I

First thing was to create the map. Normal people would have created the map the previous week or even the day before, but no. This is amateur hour. Early in the morning before the event while pouring over Google maps on a tablet, I got an idea of where I wanted to put the cones. Once we arrived at the Alameda County Fairgrounds (where RoboGames was being held), it was just a matter of logging the GPS coordinates and making a map. By 11:30 am, I had the map drawn but could find neither a printer nor a copier, and it was now time to start the event. Eighteen teams registered for RoboMagellan; 14 showed up and were all now gathered around waiting for

HomeBrew Robotics Club members, Al Margolis, Camp Peavy, Marco Walther, and Chris Mayer compete in Table Top Navigation.

SERVO 07.2017

45

Peavy - RoboGames - Jul 17_Blank Rough SV.qxd 5/30/2017 7:26 AM Page 46

The robot Sugoi from the RoboMagellan team, Pure Big Mad Boat Men got Silver with a game time of 18.56 seconds (that's four minutes 57 seconds x two .25 Bonus cones).

This is Victoria from Team Victoria. She's made from a hacked Husqvarna robot lawnmower and touched the Goal Cone in six minutes 42 seconds. (Photo courtesy of Ralph Campbell.)

DROID from Brazil starts a run. This was the only robot that used water to extinquish the "fire." One time, it ran up the wall and spilled its entire reservoir down the hallway. (Photo courtesy of Marco Walther.)

Marco Walther launches his Gold winning EnDeckerL. Marco's RoboMagellan robot is interesting in that it uses the Pixhawk autopilot system normally used in drones.

A homebrewed tricycle named Hodor from Old Guy Robots was very competitive. It touched the Goal Cone on two different runs with the best time of 1:05 (straight shot). (Photo courtesy of Ralph Campbell.)

46

SERVO 07.2017

That's Fire Fighting Eagle 1 that took Silver, Blue Flame that took Gold, and DROID for Bronze. (Photo courtesy of Marco Walther.)

Peavy - RoboGames - Jul 17_Blank Rough SV.qxd 5/30/2017 7:26 AM Page 47

It's Taiwan vs. Taiwan in MiddleWeight Kung Fu. Georgybuster (nearside) earned Silver against Big Toe for Gold (farside). (Photo courtesy of Michael Leon Overstreet.)

Marcio from Brazil takes his dog, "Fido" for a walk. Fido features glowing eyes, hard-wired remote control, and a viscious bark! His head is made with an old boot and his ears are made from socks.

the map. (Well, it is what it is ... amateur hour.) After apologizing for no map, I sheepishly handed out blank sheets of paper and we all walked the course with the competitors calling out the coordinates for each cone, taking care to make sure everyone copied the information accurately. After all that, the event was on! As mentioned, 18 robots registered, 14 showed up, 12 robots competed (that is, got off the starting line), and nine touched the Goal Cone! That’s way better than last year when no one touched it. I believe RoboMagellan is probably one of the best events at RoboGames because it involves full-scale robots and it does something potentially useful: finding its way to a specified GPS coordinate. There was an interesting assortment of RoboMagellan robots this year. While RC cars are the predominate base, a homebrewed tricycle named Hodor from Old Guy Robots was very competitive. There were four-wheel drive robots, six-wheel drive, a Power Wheels™ vehicle (also from the “Old Guys”) ... even a hacked robot lawnmower (Team Victoria). In the end, Marco Walther took Gold for Germany (yes, the same guy who won Silver in Table Top Nav). His big RoboMagallen robot, EnDeckerL (the “L” stands for large) earned a 12.06 second “game time” by touching the Goal Cone in three minutes/13 seconds plus two .25 Bonus scores (3:13 x .25 x .25). The robot “Sugoi” (USA) from Team Pure Big Mad Boat Men earned Silver with a game time of 18.56 seconds (4:57 x .25 x .25), and finally longtime competitor “Monty” (USA) from the Palos Verdes Institute of Technology won Bronze with a straight shot from the Start Cone to the Goal Cone in 43 seconds.

Middleweights (120 lb) Blender vs. Touro hitting hard! (Photo courtesy of Jon Bennett.)

Whiplash vs. Original Sin in the final match. Original Sin took Gold after winning all five of his matches. (Photo courtesy of Tony Woodward.)

Sunday was Fire Fighting day. My good friend, Bob Allen is the head fire chief and also a competitor. Since it’s SERVO 07.2017

47

Peavy - RoboGames - Jul 17_Blank Rough SV.qxd 5/30/2017 7:26 AM Page 48

Opening Speech https://www.twitch.tv/videos/ 137401440 Results from RoboGames 2017 http://robogames.net/2017.php RoboGames 2017 Full Results — Combat https://www.reddit.com/r/battlebots/ comments/67jhkr/robogames_2017_ full_results Twitch.TV RoboGames https://www.twitch.tv/robogames/ videos/all

Long-time competitor, Fuzzy Mauldin (center) with Team Toad and the Polar Vortex. (Photo courtesy of Jon Bennett.)

KungFu mid-weight RoboGames 2017 #2 https://www.youtube.com/watch?v= Nje7PwQpXzg&list=PLM7oigWuoEy n_a7XI-AdxFZR8AGPMpYpa& index=2&t=16s Soccer RoboGames 2017 https://www.youtube.com/watch? v=1Xm7xu9Qr5Y&list=PLM7oigWuo Eyn_a7XI-AdxFZR8AGPMp Ypa&index=4

Robogames 2017 Interview #1 https://www.youtube.com/watch?v= sPgj4Bg8q9w&list=PLM7oigWuoEyn _a7XI-AdxFZR8AGPMpYpa&index =6&t=139s

Last Rites vs. Original Sin https://www.twitch.tv/videos/ 136755776

2017 RoboGames RoboMagellan — Winning run of EnDeckerL https://www.youtube.com/watch?ti me_continue=1&v=j0d6qT2OeDk

Last Rites goes sailing after hitting either Crash or Burn. (Photo courtesy of Jon Bennett.)

SERVO 07.2017

KungFu mid-weight RoboGames 2017 https://www.youtube.com/watch?v= cmqB0Nb_efc&list=PLM7oig WuoEyn_a7XIAdxFZR8AGPMpYpa&index=3

2017 RoboGames — Various Highlights https://www.youtube.com/watch?v=f GNSmW7IuZQ

autonomous, that’s okay. After all, once you press the on switch, the robot is on its own. (Except perhaps in Bob’s case. He got extra points by activating his robot with a piezoelectric buzzer between 3 kHz and 4 kHz — that’s the sound commonly used in smoke detectors). Bob’s first run went flawlessly and he successfully

48

Resources

Touro Maximus vs. Original Sin https://www.twitch.tv/videos/ 136755839

extinguished the flame (a candle). However, on his second run, he blew a fuse and did not start. After replacing the fuse, Bob’s next run turned into a success too. He put out two fires, and so his entry, Blue Flame won the Gold medal. Second place “Eagle 1” (USA) also extinguished a candle for Silver. However, third place “DROID” from Brazil did not. Although DROID didn’t successfully put out a candle, he was given Bronze because he explored more rooms than any of the other robots. One funny story ... DROID was the only robot to use water to extinguish the flame as opposed to a fan or CO2. So, Bob told the team they would have to clean up the fluid after putting out the fire. When DROID entered the room in which the candle was lit, it drove up the wall and poured the entire water reservoir down the hallway. Instead of being amateurish, I guess you could say it was a real SERVPRO moment. I entered my robot, Loco Mo into the ArtBot category, but in retrospect should have entered him in Best of Show. “Mo” is more of a Best of Show type robot than Art. Regardless, after helping Bob run the Fire Fighting, is was time for my ArtBot to be judged. Mo is a ROS-based Neato vacuum cleaner with table legs and a shallow box for a table top. Just for show, I’d put Rodney (my machine intelligent robot; see SERVO February 2015) bouncing around like an amoeba learning the motion pattern that maximizes displacement. I billed it as “Rodney and Mo ... the Dog and Pony Show.” Meh, it sounded good at the time. Anyway, I had made a map earlier and navigated to the end of the room autonomously while Rodney bumped back

Peavy - RoboGames - Jul 17_Blank Rough SV.qxd 5/30/2017 7:26 AM Page 49

and forth in the shallow table top *learning* to go backwards and forwards. (Yes, we are easily amused.) Again, in retrospect, I think it had something going for it, but it wasn’t art. The judges didn’t think so, either (but hey, it was amateur hour). Competing legitimately in “Best of Show” was a friend from Brazil I met last year, Marcio Nehrebecki. This year, his entry was a homebrewed dog named Fido. Marcio is a true artist and a simple man. He made his controller with clothespins. Fido could walk (roll), wag his tail, turn his head, bark, and his eyes would light up. What I found really inspiring, however, were the objects with which Fido was built. The dog’s head was a shoe. The ears? Socks. The light-up eyes were automotive bulbs. Fido was really a joy to behold. Michael Chen (Taiwan) took Gold in Best of Show with his humanoid robot, Big Toe. A few robots of interest that I saw along Ragin' Scotsman and Cavalier playing with fire. (Photo courtesy of Jon Bennett.) the way included the new TurtleBot3 (http://turtlebot3.robotis.com/en/latest). honor to fight for Gold. So, that’s Original Sin (USA) Gold, There are supposed to be two models available from Whiplash (USA) Silver, and Swamp Thing (USA) Bronze. Robotis: one with a Raspberry Pi; and the other based on I always hunt down my big Facebook friend, Fuzzy the Intel Joule board and a RealSense camera priced Mauldin (owner of the “Lazy Toad Ranch” in Texas). He and between $500-$1,600. his sidekick, Tony Woodward have added a new member to There was also an Interbotix Turtlebot 2i the team: Tony’s son Pierce, who got to try his hand at (www.trossenrobotics.com/m/interbotix). My friend, robot-wrangling. Unfortunately, it wasn’t in the cards for Andrew Dresner is Principal Engineer at Interbotix Labs. We Team Toad’s Polar Vortex this year, as they placed 14th had recently spent time at the Silicon Valley Robot Block winning only one of three fights. Party making maps and navigating around. The Turtlebot 2i Another top contender, Ray Billing’s Last Rites (a.k.a., is unique with its arm which Interbotix has programmed Tombstone) also had a tough go at it, winning only two of and integrated into the ROS rviz environment. its four fights. TIKITRON, by Samuel M. Coniglio IV (USA), won Gold in the Art Bots: Bartending Division. I understand it now dispenses kid-friendly drinks. Farad by Becky Sherman (USA) took Silver and is always a winner in the static ArtBot category. In humanoid robots, Taiwan’s Georgybuster took Gold At long last, it is coming down to the robotics age. I’ve in FreeStyle: Original, where scratch-built humanoid robots been saying this for decades, but now it’s really true. demonstrate a free-form exhibition of motions judged by Robots are where the microcomputers were in the ‘70s. originality, agility, and variety of moves. India took Silver Over the next 10 years, robots will be everywhere doing all with SRMTH Ghost, and the US took Bronze for Michael sorts of work. Overstreet’s Boomer-3D. As with the microcomputer revolution, there will be a According to Overstreet, “There was a lot of help and lot of self-taught professionals because we are moving so sharing among the humanoid teams about fixing robots fast this cannot be taught in schools. We are all “amateurs” and how to improve their robots.” He hopes to have events figuring it out as we go along. especially for the DARWIN-Minis next year. RoboGames is robot life, robot DNA. Event creator and This year, RoboGames’ combat was broadcast live on organizer, Dave Calkins has been known to inspire. He’s “twitch.tv,” so there are videos at https://www.twitch.tv been known to change lives. Who knows ... maybe he can /robogames to see (not total amateurs). change the world. What’s happening is we’re creating a The big draw at RoboGames has always been the record of robot competitions; an annual deadline where we heavyweight combat robots. This year (since there was no stop and take measure ... a basis on which to build. BattleBots™), the 220 lb heavyweight division filled up fast, Sometimes when you’re so close to something, it’s and all the big names were there. Original Sin took Gold hard to recognize your world has changed, and the best is against Whiplash, who had defeated Swamp Thing for the yet to come. SV

That’s a Wrap

SERVO 07.2017

49

Leeman - Multirotor Hobbyist - Jul 17_Blank Rough SV.qxd 5/30/2017 7:27 AM Page 50

The Multi-Rotor Hobbyist

ESC Basics By John Leeman To post comments on this article and find any associated files and/or downloads, go to www.servomagazine.com/index.php/ magazine/issue/2017/07.

Just like an engine powers your car but is controlled by a computer, the brushless DC motors (BLDCs) we talked about last month power your quad but are also controlled by a computer. This computer — the electronic speed controller (ESC) — can make or break the performance of your quad. This month, we’ll go over the basics of ESCs and test one out with various firmware versions to see the difference between them.

How ESCs Work To make our rotor rotate, the ESC needs to switch current between the three different sets of windings we discussed last month. To do this, the state of the three wires going into our motor are controlled with a set of MOSFETS (metal–oxide–semiconductor field-effect transistors). The windings are set to either high (connected to the battery voltage), low (connected to ground), or left floating. Since there are three wires that can be in three states, we can calculate the number of unique states. Reaching back into our math classes, we know that the number of permutations of a system with no repeating values is the factorial of the number of choices. In our case: 3! = 3 x 2 x 1 = 6

that make up one complete revolution of the motor (Figure 1). If the sequence is played in reverse, the motor rotates in the opposite direction. This explains why you can switch any two wires on a BLDC motor and get it to rotate in the opposite way! So, how does an ESC accomplish this switching? It needs some high current switches that can be controlled by a microcontroller that takes care of the timing and orientation sensing. A MOSFET voltage source inverter will do the job. Before we look at the entire three-phase switching array, let’s examine a single MOSFET voltage source switch (Figure 2). The N-channel MOSFET can be thought of as a single-pole single-throw (SPST) switch of sorts. When the gate voltage is at the ground potential, the MOSFET acts as an open switch. When the gate voltage increases above a certain threshold voltage, the MOSFET begins to conduct. Using the configuration shown, we can connect the

If we number the inputs to our motor 1, 2, and 3, we can chart out the state of the windings over the six states Figure 2: A single MOSFET source inverter circuit allows the output to be connected to either the voltage source or ground.

Figure 1: Six unique states compose a complete rotation of the BLDC motor. Each phase can be connected to the voltage source (red), ground (green), or left floating (orange).

50

SERVO 07.2017

Leeman - Multirotor Hobbyist - Jul 17_Blank Rough SV.qxd 5/30/2017 7:27 AM Page 51

output to either V+ or GND. If we create three of these source switching circuits and connect them to the windings, we can create the high, low, and floating states (Figure 3). Unfortunately, being able to create the proper switching sequence isn’t enough. The sequence needs to be properly timed with the orientation of the motor’s rotor. Some larger motors have Hall effect sensors near each coil and can determine the position of the rotor by sensing which magnets are nearest each coil. This technique is very reliable and can determine the position of the rotor at rest. It does require extra sensors to be built into the motor, increasing its weight and cost. It also requires more connections between the motor and ESC. These types of ESCs have much simpler firmware and are generally more reliable on startup. Figure 3: A complete ESC power switching network consists of three The way that most ESCs work on quadcopters (and the MOSFET source inverters. By manipulating the six inputs, each output can be put into one of the three possible states. one we’re looking are here) is with sensorless orientation determination. Remember the floating third phase from our power switching network? It can help us out in the determination of orientation by measuring the back electromotive force (EMF) generated on that coil. The back EMF can be modeled as roughly trapezoidal over the course of one revolution of the motor (Figure 4). By measuring this, the ESC can determine the position of the rotor. The biggest catch is that this method requires the motor to Figure 4: The back EMF measured on each of the sets of coils represents be rotating. It also requires some signal processing. a trapezoidal shape (hence, BLDCs are sometimes referred to as trapezoidal motors). (Image courtesy of Freescale Semiconductor.) The simplest way to determine the orientation of the rotor involves integrating the signal and determining when there is a zero-crossing. This can be done with analog comparators on each of the windings, but is Choosing the wrong ESC can result in poor more commonly done with purpose-built chips or signal performance, excessive cost, large releases of magic smoke, processing on the microcontroller. For the zero-crossing detection technique to be or some combination of these. In selecting your ESC, you effective, the signal needs to be filtered to remove as much need to consider the voltage rating, current rating, a noise as possible; a task that becomes harder at higher battery eliminator circuit (BEC), and firmware. speeds. If you really want to dig into how the orientation Selecting an ESC with the incorrect voltage rating is determination works, the gory math is fully discussed in an probably the quickest way to literally watch your hobby app note from NXP (www.nxp.com/filesdollars go up in smoke. ESCs will have the battery voltages static/dsp/doc/app_note/AN1914.pdf). they are rated to work with in their description. Often, this The signal can also be processed using the relatively is specified in terms of the battery configuration such as 2modern technique of Kalman filtering. While these filters 6S instead of the voltage. For our LiPo batteries, each cell is could make up an entirely different column (or book for 3.7 VDC, so a 2S (two cells in series) is a 7.4 VDC battery, that matter), the basic idea is using our current noisy and a 6S is 22.2 VDC. Connecting a 6S battery to a 2S knowledge of the system and some model of how we rated ESC can really ruin your day and potentially be expect the system to behave to get a better estimate of the dangerous. real state of the system. These are again more complex to The second most important rating is the current implement, but can provide much better estimates of the capacity of the ESC. Again, this should be specified by the rotor’s position. manufacturer. If your motor is rated to draw up to 10A, No matter which method is used to calculate the then you need at minimum a 10A ESC. Generally, you will position of the rotor, it must be turning for some back EMF want to buy an ESC that is a little bigger than you need — to be generated before the ESC can get proper maybe a 15-20A ESC for the 10A motor. synchronization established. Generally, the ESC applies Using an over-rated ESC reduces the risk of an ESC some current to the coils to start rotation, and then gets a overheating and/or failing in the middle of a flight and phase determination and locks into the rotor’s position. causing a spectacular crash. Don’t go overboard though, or

Choosing ESCs

SERVO 07.2017

51

Leeman - Multirotor Hobbyist - Jul 17_Blank Rough SV.qxd 5/30/2017 7:27 AM Page 52

network, but I’ve also never heard of an issue caused by using an onboard BEC. Finally, it is worth considering what firmware is installed on the ESC. While we can — and will — change the firmware, you may not be comfortable doing so or just don’t want the hassle. Generally, manufacturers will note if they have some of the more popular firmwares (SimonK or BLHeli) pre-flashed.

Testing Setup To demonstrate and test an ESC, I made a quick benchtop setup. For the test, I ordered the Turnigy D2830Figure 5: The Turnigy D2830/11 is a 11 Brushless Outrunner motor (1000 kv; Figure 5), Afro nice sized motor for quads in the 20A ESC (Figure 6), and Turnigy 9X 2.4 GHz eight-channel 12-16” range and a handy spare to keep on hand. (Image courtesy of receiver (V2). I found an old acrylic test fixture plate from a Amazon.) previous project and repurposed it to hold all the pieces in you’ll just be paying more to place. From past experience, I’ve learned that it’s never a carry around more weight on good idea to test electromechanical setups without your quad. I’ve found that the fixturing. 1.5-2x range works well. I attached the motor to its mounting bracket to provide Of less importance is clearance for the back of the shaft and affixed the bracket whether the ESC contains a with two strips of double-stick foam tape. I also taped the BEC. This is a very convenient receiver and ESC to the fixture plate (Figure 7). Foam tape Figure 6: A slightly different form factor than the Afro Slim really is an excellent prototyping tool and one I think is inclusion as the ESC will ESCs used in the previous regulate the battery voltage it is quad build, the Afro 20A ESC often underappreciated! is a reliable, inexpensive, and receiving down to 5 VDC to I hooked up the motor to the ESC with the pre-installed easily programmable power the RC receiver on your bullet connectors. In this case, the wire color happens to controller. quad. You’ll only need one ESC match up, but there is no reason to do so. Remember that with a BEC on your quad, or reversing any two wires results in the motor running the you can buy a separate BEC as we did with our scratch built opposite direction — there is no “proper” connection order. quad. I then connected the ESC control wire to the receiver. For this article, I chose an ESC with a built-in BEC to In this setup, we will use the BEC to power the receiver, so make the testing setup as simple as possible. Personally, I all three wires of the “servo” cable will remain in place. I don’t like the idea of my radio power coming off the same plugged the servo wire into channel 3 of the receiver since PCB (printed circuit board) as a high current switching that is the default throttle channel for a mode 2 setup. Make sure you pay attention to the proper orientation of the plug. On the Turnigy receiver, the signal wire (orange) is the top connection. Next, we need to power the test setup. If you have a bench power supply capable of providing the required current, I would advise using it so that you can set current limits and avoid the dangers of using batteries in test setups. A battery can be used as long as proper precautions are taken. I used a freshly charged flight battery for all tests in this article — and there were consequences. To connect the battery, I made a simple XT60 to bullet connector wiring harness and attached it to the main power input of the ESC. It bears emphasizing that before any power is actually connected, there should be no propellers Figure 7: A nice test stand can be fashioned from an old piece of scrap acrylic connected to the motor. Not only are they plate and foam tape.

52

SERVO 07.2017

Leeman - Multirotor Hobbyist - Jul 17_Blank Rough SV.qxd 5/30/2017 7:27 AM Page 53

the minimum stable rotational speed and unnecessary for this test, but the test fixture maximum rotational speed with a photo is not equipped to combat the torque and tachometer. lift of the propeller. The setup could wildly These can be found for less than $20 take flight and cause injury, slice wires, and on Amazon (http://amzn.to/2ohS9Og) result in property damage. and are a handy thing to have in your Before any motor will turn, we have to toolbox (Figure 8). The one I found complete two important steps: binding of included a little cloth carrying case and the receiver and calibration of the ESC some reflective tape. range. Binding is the same process we used The tachometer works by shining a light when building our homemade drone last on the rotating surface and counting the year, but the ESC calibration that was number of peaks in the reflected light over handled by the flight controller is now our a given amount of time. For example, a responsibility. motor rotating at 3,600 RPM is rotating at: Receiver binding is slightly different based on which radio you have, but they all 3600 rev 1 minute 6 rev have the same basic procedure. First, set = * the transmitter to an unused model profile. 1 minute 60 seconds second I used the third profile on my remote as the So, counting six light pulses per second first two are reserved for our homebuilt would correspond to 3,600 RPM averaged quad and the Parallax ELEV-8. Figure 8: For about $15, a decent handheld tachometer can be over that time span. To get a good reading, Next, plug the “bind” jumper wire onto added to your tool bag. These a high reflectivity contrast region is needed. the receiver’s bind port. This jumper come in handy more often than Generally, a white paint pen suffices. In this connects the signal to ground and puts the you’d think if you deal with motor driven equipment. case, though, that will not reliably perform. receiver in a binding state. Power-up the The outer casing of the motor is a receiver, then turn on the transmitter while mirror finish, so we actually need to decrease the reflectivity holding down the “bind” button (on the back of the of most of the casing, leaving a small strip of reflective transmitter on many models.) Finally, power-down the surface. I wrapped painter’s tape around the casing and system and remove the binding jumper. trimmed a section about 1 cm wide out with a razor knife. I ESC calibration also varies slightly, but for the ESC used tested the motor at its maximum and minimum throttle in this article the setup is powered and the transmitter settings several times, and reliably obtained numbers of turned on with the throttle in the maximum position. Then, 12,500 RPM and 3,000 RPM, respectively. Arranging after some beeps from the ESC, the throttle is lowered to everything on the bench with a helpful adjustable height the minimum. The ESC records the width of the pulse width modulation (PWM) signal at each of these extents and scales its throttle range to match. If this step is skipped, it is very likely that only a small portion of the control stick range will have an effect on the throttle — a large apparent deadband in the response. With the ESC calibration completed, the motor should respond to the throttle control. The motor’s speed shouldn’t top out until the control is all the way at its maximum, and the motor should cut out and stop at the very bottom of the range. If not, the calibration was unsuccessful and needs to be tried again. Play with the throttle control some and observe the motor’s response. Try gradual changes in throttle as well as rapid changes — especially from stopped to maximum throttle. I videoed the response of the system in its out-of-the-box configuration (https://youtu.be/QQ7qr76GXR0). Figure 9: The ESC and motor ready for testing. Pads of sticky notes make great I rounded out my test suite by measuring adjustable height stands for lining the tachometer up with the motor. SERVO 07.2017

53

Leeman - Multirotor Hobbyist - Jul 17_Blank Rough SV.qxd 5/30/2017 7:27 AM Page 54

Figure 12: The KKMulticopter Flash Tool is very easy to work with. All of the settings are on a single interface page. You should see the message “Flashing of firmware was successful” at the end of the programming cycle.

stand (a.k.a., post-it notes) made for a really nice test setup (Figure 9). Figure 10: A simple USB programmer from Afro works well for programming many different ESCs. Using a USB extension cable made working with this much easier.

Flashing New Firmware As I’ve mentioned, there are arguments on message boards all over the Internet about which ESCs and ESC firmware combinations yield the snappiest response. I wanted to test the stock version against the two most popular firmware sets out there: SimonK and BLHeli. Since the ESCs are microcontroller based, I was prepared to pull out a programmer and solder or clip onto little surfacemount packages. I was happy to find that there is indeed an easier way. To Flash the firmware, we can use a USB programming tool (Figure 10). These can be found for about $14 (http://amzn.to/2p5cnzG). I Figure 11: Be sure that the servo lead from the ESC is plugged into the programmer correctly — the signal wire (orange) on the right hand side.

54

SERVO 07.2017

stuck with the Afro brand, but all of these ESC programmers should be interchangeable as long as the Flashing software supports that tool. The first step is to install the drivers for the programmer. In the case of the Afro ESC programmer, Silab USB interface chips are used, so the Silab driver needs to be installed from www.silabs.com/products/developmenttools/software/usb-to-uart-bridge-vcp-drivers. Once this is done, the tool should show up as a virtual COM port for your system. Check the devices pane with and without the device plugged in on Windows, or check in /dev/ for the Linux/Mac folks. The first firmware we’ll try is SimonK. Download and install Java if it isn’t already on your system (https://java.com/en/download). Then, head over to the KKMulticopter Flash Tool page (http://lazyzero.de/en/modellbau/kkmulticopterflasht ool) and download it. Apparently, there have been some problems with the author’s webpage and at one time folks were downloading directly from a Dropbox folder. When dealing with unknown software, it is always wise to run it on either a virtual machine (using tools like VirtualBox or Parallels) or on a dedicated machine. I have a few old Windows laptops that serve as testing platforms for this kind of thing. Unzip, download the folder, and open up the program. Before connecting anything to power, first disconnect the motor from the ESC. (Also read the rest of the article for a cautionary tale before trying this yourself.) Next, disconnect the ESC from the receiver and plug it into the USB programmer. Pay careful attention to the orientation,

Leeman - Multirotor Hobbyist - Jul 17_Blank Rough SV.qxd 5/30/2017 7:27 AM Page 55

remembering that the signal wire is orange on this ESC (Figure 11). Now, connect the battery to the setup to power the ESC. Then, plug in the USB programmer to your machine. A USB extension cable makes this plugging and unplugging a lot easier. In the KKFlashTool, set the programmer to Afro ESC USB Programming Tool from the dropdown menu. Make sure that the COM port setting is correct and that the “use defaults” check box is checked. Our Figure 13: My first attempts at Flashing new firmware ended with ESC is based on an an arc jumping between the output ATmega connectors and completely destroying one of the NFETS, microcontroller, so blowing a hole in the casing! select the atmega 8based brushlessESC (8kb flash) option from the controller menu. Set the firmware repository to Afro NFET and select the most recent version of the firmware (Figure 12). Finally, click the green image of a running person and let the program Flash and verify. When completed, the status should read “Flashing of firmware was successful.” Remember earlier when I espoused the merits of working from a current-limited power supply? It was not without reason! On my first attempt at Flashing SimonK onto the ESC, I heard a bang, saw a flash, then watched error messages scroll across the KKFlashTool terminal. I immediately removed power, but the ESC was nonresponsive in later attempts to Flash it. After ordering another ESC, I did a post-mortem. It turns out that the motor connectors — while disconnected — were still close enough that an arc jumped between two of them. I cut away the plastic shrink tube enclosing the ESC and saw a small black piece of encapsulant fall out. That’s never a good sign. I flipped the board over and saw that there was a hole blown out of one of the NFETs (Figure 13). A currentlimited supply would likely have prevented this from happening. After this incident, I taped the motor connections down in a spread configuration (Figure 14) and never had another issue.

Figure 14: To prevent the arcing, I taped the output leads down in a spread configuration and placed the motor’s connectors under the acrylic plate.

Figure 15: Finding the ESC in the BLHeli PDF just takes a little searching. This is a nice reference to have as it allows you to identify cloned ESCs and Flash them with the proper firmware.

After Flashing the newest SimonK firmware, I repeated my earlier tests with the same setup and videoed it (https://youtu.be/Z_HyZ3CUZXw). The motor sounded roughly the same, and had a minimum/maximum RPM of 3,200 and 12,500 (very similar to the stock configuration). The biggest difference I noticed was on the step test from full stop to full throttle. The motor’s torque actually rotated my test plate on the bench (see 11 seconds into the video). This tells me that the accelerations achieved by this firmware are much greater than the stock configuration. For an aerial photography platform, this probably isn’t good, but for a racing setup it would be a decent performance gain. I didn’t see any issues with/during the calibration procedure; the biggest downfall of this install was the questionable SERVO 07.2017

55

Leeman - Multirotor Hobbyist - Jul 17_Blank Rough SV.qxd 5/30/2017 7:27 AM Page 56

Figure 16: Select the “ATMEL SK Bootloader (Afro/Turnigy USB Linker) from the interface menu after starting BLHeliSuite.

Figure 18: Scroll through the listed firmware versions to find one that matches the description given in the BLHeli PDF reference. We want the MULTI version since this is a quadcopter and not a helicopter with a main and tail rotor motor.

nature of the download. The second firmware I tried was BLHeli. This version was touted to have more knobs to tweak, so that immediately got my attention. To start off, go download the BLHeli suite from https://www.mediafire.com/folder/dx6kfaasyo24l/BLH eliSuite. Also head over to the project’s GitHub page at https://github.com/bitdump/BLHeli and download that repository. You can do this either by cloning the repository if you are familiar with Git and GitHub, or by clicking the download button and getting a zip file. Go ahead and unzip everything we just downloaded. In

56

SERVO 07.2017

Figure 17: After selecting the correct COM port and clicking connect, the button should turn to disconnect and the setup interface will be activated.

Figure 19: BLHeli has a warning before Flashing asking you to double-check the firmware version you’ve selected. Take a moment to pause and verify the information before clicking yes.

the BLHeli folder from the GitHub repository, navigate to the ATMEL directory and search for the file “BLHeli supported Atmel ESCs.pdf.” Open this file up and have a look inside. You’ll see many ESCs with front/back PCB photos. Using this guide, you can identify what version of the firmware is available to use on a mystery or clone ESC. Paging through, we can find our ESC and see that we should use the “Afro_20A_..” firmware and see the parameters in BLHeli that it supports (Figure 15). Next, open the extracted folder with the BLHeli application — no installation is necessary. Again, disconnect the motor from the ESC and separate the output leads. Power the ESC and connect the USB programmer to the

Leeman - Multirotor Hobbyist - Jul 17_Blank Rough SV.qxd 5/30/2017 7:27 AM Page 57

servo lead. We need to tell the software which programmer we’re using, so go to the “Select ATMEL/SILABS Interface” menu and select the appropriate tool. In my case, it’s the ATMEL SK Bootloader (Afro/Turnigy USB Linker) option (Figure 16). At the bottom left of the application, select the appropriate COM port for the programmer; check in your system Device Manager if you’re not sure which port the programmer is connected to. Click on connect and the system should now be connected to the programmer and ESC (Figure 17). Click the “Flash BLHeli” button to begin the process. A dialog box will pop up asking which firmware should be Flashed. Scroll through and find the firmware version that we looked up in the PDF reference manual. I used the “[REV: 14.9] Afro 20A MULTI” firmware (Figure 18). Click the green OK button and confirm that you want to start Flashing the firmware (Figure 19). After a bit, you should get a message that the firmware Flash was successful (Figure 20). Figure 20: After a successful Flash of the firmware, we are ready to test, but You can now adjust the various parameters for take the warning seriously. Using a current-limited power supply is highly the ESC like the damping, beep level, temperature recommended in case something goes wrong during the programming. protection, etc. I used the default values for the and the different optimizations that can be done. The most bench test. Once you have all the settings correct, click the interesting thing to do would be to make a more robust “Write Setup” button to save the settings to the ESC. When and instrumented test stand. you’re all done, click the “Disconnect” button and powerMeasuring things like the torque, speed, input throttle down the setup. We’re all set to reconnect the motor and command, and even thrust with a propeller would be very receiver to test the firmware! interesting. It would then be possible to quantify the During the bench test, I observed a very fast throttle differences between the various firmwares and settings response. The motor sounded slightly different, especially at with numerical measures. Maybe such a test stand will be a the lowest speeds — likely a product of how the future project for us! synchronization and position calculations are done. During There are lots of interesting components including a rapid throttle change, it did slightly torque the test setup strain gauge load cell measurement and some mechanical around as well. engineering. Such a system could also be modified with Other than response speed, I was amazed by the minimum stable running speed. This firmware could run the accelerometers to create a dynamic propeller balancing motor all the way down to 1,100 RPM — less than half of system. the other’s minimum speeds. The maximum speed was Another interesting test would be looking at the power again 12,300 RPM — seemingly the limit of reliable position consumption. Will different firmware versions modify how determination and control for this setup. the control signals are sent enough that there are battery The video of the test can be found at longevity advantages to one over the other? A hefty fixed https://youtu.be/JP2vQVF3za4. power supply with a current shunt for measurement would So, what’s the best firmware? It’s largely up to your easily tackle this question. personal preferences. I could see the rapid throttle response of BLHeli being great for performance fliers. The low speed capability would also be great for fixed wing models. As an instrument person, I’m often more interested in stable flight, so a slower responding firmware or tweaked BLHeli Now that you know all about ESCs and the motors they may be better. The good news is that you can try them all drive, it’s time to go out and optimize the setup of your for very little cost. quad. Just remember to respect the amount of current that can be drawn from your flight battery! Until next month, fly safely. SV

Closing Thoughts

Going Further

We could spend a very long time talking about ESCs SERVO 07.2017

57

SV Webstore - Jul 17 working_SV Webstore May 16 working.qxd 5/30/2017 7:30 AM Page 58

The SERVO Webstore CD-ROM SPECIALS

07.2017 FEATURED

ROBOTICS Arduino Robot Bonanza by Gordon McComb Create high-tech walking, talking, and thinking robots. "McComb hasn’t missed a beat. It’s an absolute winner!" Breathe life into the robots of your dreams — without advanced electronics or programming skills. Arduino Robot Bonanza shows you how to build autonomous robots using ordinary tools and common parts. Learn how to wire things up, program your robot's brain, and add your own unique flair. $29.95

Robot Builder's Sourcebook Fascinated by the world of robotics but don’t know how to tap into the incredible amount of information available on the subject? Want the names, addresses, phone numbers, and web sites of companies that can supply the exact part, plan, kit, building material, programming language, operating system & computer system? Robot Buider’s Sourcebook is just what you’ve been looking for! $36.00

58

SERVO 07.2017

Making Things Move: DIY Mechanisms for Inventors, Hobbyists, and Artists by Dustyn Roberts

In Making Things Move: DIY Mechanisms for Inventors, Hobbyists, and Artists, you'll learn how to successfully build moving mechanisms through non-technical explanations, examples, and do-it-yourself projects — from kinetic art installations to creative toys to energy-harvesting devices. Photographs, illustrations, screenshots, and images of 3D models are included for each project. $29.95

Build Your Own Humanoid Robots by Karl Williams GREAT 'DROIDS, INDEED! This unique guide to sophisticated robotics projects brings humanoid robot construction home to the hobbyist. Written by a well-known figure in the robotics community, Build Your Own Humanoid Robots provides step-by-step directions for six exciting projects, each costing less than $300. Together, they form the essential ingredients for making your own humanoid robot. $24.95

Build Your Own Teams of Robots with LEGO® MINDSTORMS® NXT and Bluetooth® Plan, design, assemble, and program robot squads that communicate and cooperate with each other to accomplish together what they can’t do individually. Inside the authors show you how to construct a team capability matrix (TCM) and use the Bluetooth RoboticOriented Network (BRON) so your robot teams can share sensors, actuators, end effectors, motor power, and programs. $30

Robotics Demystified by Edwin Wise YOU DON'T NEED ARTIFICIAL INTELLIGENCE TO LEARN ROBOTICS! Now anyone with an interest in robotics can gain a deeper understanding — without formal training, unlimited time, or a genius IQ. In Robotics Demystified, expert robot builder and author Edwin Wise provides an effective and totally painless way to learn about the technologies used to build robots! $19.95

To Order Call 800 783-4624 or visit our webstore

SV Webstore - Jul 17 working_SV Webstore May 16 working.qxd 5/30/2017 7:30 AM Page 59

Order online www.servomagazine.com SPECIAL OFFERS

Build Your Own Quadcopter by Donald Norris Build and customize radio-controlled quadcopters that take off, land, hover, and soar. Build Your Own Quadcopter features step-by-step assembly plans and experiments that will have you launching fully functioning quadcopters in no time. This fun, do-it-yourself guide fuels your creativity with ideas for radical enhancements, including return-to-home functionality, formation flying, and even artificial intelligence! Reg $50.00 Sale Price $39.95

The SERVO Buddy Kit

PROJECTS 3D LED Cube Kit

PS2 Servomotor Controller Kit

From the article “Build the 3D LED Matrix Cube” as seen in the August 2011 issue of Nuts & Volts Magazine. An inexpensive circuit you can build to control a servo without a microcontroller.

For more information, please check out the May 2008 issue or go to the SERVO webstore.

Includes an article reprint.

$39.55

This kit shows you how to build a really cool 3D cube with a 4 x 4 x 4 monochromatic LED matrix which has a total of 64 LEDs. The preprogrammed microcontroller that includes 29 patterns that will automatically play with a runtime of approximately 6-1/2 minutes. Colors available: Green, Red, Yellow & Blue. Jig and plastic cases also available.

This kit accompanied with your own PlayStation controller will allow you to control up to six servomotors. Includes all components and instruction manual. For more information, please see the February 2011 edition of SERVO Magazine. Assembled units available! $79.95

$57.95

SERVO 07.2017

59

Carroll - Then & Now - Jul 17_Then & Now - Sep15.qxd 5/30/2017 7:35 AM Page 60

a n d

g{xÇ Now

by Tom Carroll [email protected]

What is the Future of Humans Working Among Robots? For a while now, there have been many articles written about how robots have improved working conditions in modern factories. I recently read one commentary with the exact opposite view. In the New York Times' online newsletter, The UpShot, Claire Cain Miller recently wrote an interesting piece entitled "Evidence that Robots are Winning the Race for American Jobs." This was one of the first of many articles that I've read recently that have expressed the opinion (with factual data to back it up) that robots are beginning to displace humans in their manufacturing jobs. f you think about that statement and the progress that has occurred in robotics in the past five decades, the implementation of automatons in factories was supposed to remove workers from dangerous and dull jobs, allowing them to move to a better and safer job. That is not what this article was really about. Here, I’m going to focus on industrial robots and manufacturing to start, as that is what economists use as a point of measure.

I

The Future of Robots from the First Robot We really have made some great strides in the design and implementation of robots since the

first industrial version made the scene in a Trenton, NJ General Motors die casting plant in 1961. Named for UNIversal autoMATION, the Unimation Unimate serial number #001 (shown in Figure 1) now sits in a Ford museum. The ‘programmed article transfer’ device (later called a ‘robot’) had decades of growing pains as die casting seemed to be its only application, with 450 Unimates sold for that specific task. Joe Engelberger, Unimation CEO is shown examining an early Unimate in Figure 2 that used complex and leaky hydraulics as actuators. These were later replaced with clean and efficient electric motors. It was not until spot-welding of cars in assembly lines that robot

Figure 1. Joe Engelberger (in the bow tie) showing off an early Unimate.

60

SERVO 07.2017

sales boomed, due in part to this newer technology.

Robots Displacing Human Workers Instead of replacing a semi-skilled worker with a robot and sending the displaced worker to classes to make him/her a robot programmer, the employee was instead given a pink slip and sent to the unemployment line. Each single robot within a factory of hundreds did not require its own repair person and programmer. Only a few selected workers could be ‘upgraded’ to the higher positions. The evil red-eyed robots in the cartoon in Figure 3 show a troubling direction that some economists feel

Figure 2. The force behind Unimation (Joe Engelberger) looking at the complex hydraulics of a Unimate robot.

Carroll - Then & Now - Jul 17_Then & Now - Sep15.qxd 5/30/2017 7:35 AM Page 61

Advances in robots and robotics over the years. To post comments on this article and find any associated files and/or downloads, go to www.servomagazine.com/index.php/magazine/issue/2017/07.

modern automation and robotics is headed. On the positive side, robots are creating some great jobs in the US manufacturing industry for humans, and companies pay well Figure 3. Robots are taking menial jobs now, but what for these jobs about the future? because of the stiff turns out that those forces competition for skilled workers. work together, and have In an article by James Carroll in been equally hurtful to Vision Systems Design, experts discuss Figure 4. A whimsical (but frightening) view of the future of robots. the impact of robots on jobs in manufacturing jobs. The car industry, for example, imports more America. At Automate 2017, Jeff beginning to manufacture robots as and more parts from abroad, while Burnstein, A3 President (Association the image in Figure 5 implies. automating their assembly in the US.” for Advancing Automation) kicked off The cartoon in Figure 4 is a bit the “Working in the Automation Age” more sobering than the first one as forum with some interesting statistics today’s college degrees might not be that were an early indicator of how Setting aside all the amazing sufficient for graduates to enter the collective group felt about the features of today’s robots, economic today’s workforce. question of jobs and robots. figures make me wonder just who is Correll continued, “In 2014, China In the United States from 2010really winning the race for jobs exported more products that were 2016, “there were more than 136,748 between robots and humans. Robots more valuable than did the US for the new robot shipments, 894,000 new and advanced computer systems are first time. Many of these were made jobs, and an unemployment rate that not just employed in manufacturing in by the low wage laborers China has went down 5.1%,” according to factories, they work among us in become famous for. Yet China has Burnstein. His comments differ widely various disciplines of healthcare in from Claire Cain Miller’s opinions. also emerged as the largest growth hospitals — not only just assisting market for robotics. China has surgeries but also performing understood that its competitive surgeries, delivering goods around advantage of cheap labor will not last large hospitals, cleaning hospital forever. Chinese companies bought floors, and even assisting doctors in more than twice as many industrial I saw another article on diagnosing diseases with advanced AI robots (68,000) in 2015 than Robohub.org that was written by such as IBM’s Watson American companies did (27,000).” Nikolaus Correll where he stated: supercomputers. China is not only buying more “Since 2000, the US has lost five Private homes are benefiting from robots than the US, it is also million manufacturing jobs. small vacuum cleaner An estimated 2.4 million robots such as iRobot’s jobs went to low wage Roombas or Neato’s workers in China and Botvac, and even robot elsewhere between 1999 lawnmowers; however, and 2011. Though more these are not quite as than a million jobs have popular due to their higher returned since the 2008 cost. Besides toy robots, recession, the net loss has these are really the only devastated the lives of significant penetration of millions of people and their robots into the home families.” environment. “Some blame robotics, Was the job of a hired others globalization. It housekeeper eliminated Figure 5. Robots in China building products for America.

The Future of Robots with Humans

Chinese Competition Now Using Many Robots

SERVO 07.2017

61

Carroll - Then & Now - Jul 17_Then & Now - Sep15.qxd 5/30/2017 7:35 AM Page 62

when the homeowner purchase a Neato robot? No. There are many categories of robots now being manufactured, and new varieties are becoming a reality every day. For decades, industrial robots were relegated to factory floors with protective cages and humans to prevent accidental interactions. For the most part, the manufacturing of industrial robots has abandoned the shores of the US, but other new varieties of robots are appearing to assist us in ways we previously had not envisioned.

something an elementary school student might make. Yet, its three inventors — John Bardeen, Walter Brattain, and William Shockley — were given the 1956 Nobel Prize in physics for its development. I don’t think there was a technical achievement more important to the advancement of our society than this humble device. Figure 7. Jack Kilby's 'Integrated Circuit' developed at Later in 1958, newlyTexas Instruments. hired engineer, Jack Kilby developed the first multi-transistor and memory chips of today now integrated circuit shown in Figure 7 at contain billions of transistors. However, if we think about it, could Texas Instruments; he was also any human on earth have predicted awarded the Nobel Prize in Physics in 2000 for its development. To me, it the future of integrated circuit Less than a year ago, I wrote an looks even cruder than the first electronics back in 1958? No way! article entitled “Robots of the Future” transistor; it’s more like a in which I discussed various robot kindergartener’s project made with drones; some could assist humans in some glue and wires. This unassuming emergency situations, while others transistor joined its brethren in greater could assist humans in their homes and greater numbers on a single chip, and robot ground vehicles that deliver and changed the whole world as we goods. These were robots of the knew it! present. I didn’t cover industrial and In an article in Electronics manufacturing robot sectors. Let’s look at another prediction of Magazine written by Gordon Moore When you think about it, no the future that hasn’t turned out like of Fairchild Semiconductor in 1965, he so many people have predicted. Back author can accurately write about a predicted a two-fold increase in a year in the 1965 time period when Moore time that has not happened. They can of the number of transistors that made his prediction, others in the only guess what might be available by could be made on a single IC nuclear power field were looking at looking at today’s technology and substrate. Others nuclear fission power plants and how the world’s stated it was 18 submarines, and just knew that economic drivers are months, while others nuclear fusion was the future. directing society’s stated every two Proponents felt it would be cleaner, pathways. Can anyone years. Whichever way self-contained, and very efficient. really predict how any you might look at it, technology will evolve — However, decades later, it is still not the wealthy much less robotics? perfected. philanthropist Moore has given more to the world than just ICs. ‘Moore’s Law’ is “Who needs horseless carriages? still holding true to Horses do a fine job of getting us this day, though around.” “You want to fly through the some say that Look back at the air like a bird? You’re nuts.” “Why today’s 10 nm lowly transistor shown have a box looking thing in your living process is heading to in Figure 6 that first room to show moving pictures? We even smaller IC parts, made its appearance have theaters, magazines, and and will soon drop just before Christmas in newspapers.” What about there being below single rows of 1947. The whole crude “a computer in every home?” Back in atoms on chips. setup was larger than a 1943, Thomas Watson of IBM stated, Nonetheless, small vacuum tube of “I think there is a world market for Figure 6. First point contact processors, FPGAs, the day, and looked like maybe five computers.” transistor.

Robots of the Future

Predictions on the Future of Technology Don’t Always Come True

Looking at the Future of Technology from the Past

62

SERVO 07.2017

Technology Needs Dreamers

Carroll - Then & Now - Jul 17_Then & Now - Sep15.qxd 5/30/2017 7:35 AM Page 63

Bill Gates and Steve Jobs would have disagreed with that as these two dreamers performed technical miracles.

Cobots are a Great Start for HumanFriendly Robots Nowadays, we have collaborative robots (called cobots) that work alongside humans. Rethink Robotics’ Baxter and then later, Sawyer shown in Figure 8 were designed to work beside humans, assisting them in kitting, packaging, loading and unloading, machine tending, and material handling. These types of robots are becoming a mainstay in all sorts of industries — not just in the industrial and manufacturing sectors. The scene in Figure 9 shows a Baxter robot performing a simple pick-and-place operation next to a supervising human.

iRobot Roomba: the Most Popular Home Robot

Figure 8. Rethink Robotics’ Baxter and Sawyer.

Figure 9. Baxter robot from Rethink Robotics.

steps before the first products were fabricated. As of this year, 15 million+ Roombas have been sold around the world, so its success has been remarkable to say the least. iRobot’s first series of vacuums used what many people called the ‘sheep grazing in a pasture’ method of covering a floor area. There were early problems and

Let’s next discuss the iRobot Roomba: the first and still the most popular home vacuumcleaning robot. Originally envisioned by Joe Jones back in 1988, the first Roomba was introduced in 2002 after a long period of dreaming, incubation, trials, and errors. The ‘Dust Puppy’ prototype model shown in Figure Figure 11. Roomba's bushing and steel shaft 10 was one of the final bound up with hair.

returns, but the designs improved a lot. In 2008, I was given a Roomba by Chairperson of the board, Helen Greiner to test for an article for SERVO. I soon discovered a problem with one of the brushes binding up the gearbox. It frequently called out “Clean Roomba’s brushes” even though they had just been taken out and cleaned. I decided to place two multimeters on the Roomba: one to read current draw and the other battery voltage. The current began to rise and voltage drop a bit. Opening up the Roomba, I noticed that the brush had become very tight and was pressing against and distorting the plastic gearbox. Hmmm. I had easily pulled the hair out of the brush’s bushings as in Figure 11, but then I noticed a tiny hardened cone of hair burnished onto the steel shaft by friction of the bushing. Carving off this small amount of hair with an Xacto™ knife temporarily remedied the problem. I found a small internal ‘star’ lockwasher that fit over the steel

Figure 10. iRobot’s Dust Puppy Roomba-to-be mockup.

SERVO 07.2017

63

Carroll - Then & Now - Jul 17_Then & Now - Sep15.qxd 5/30/2017 7:35 AM Page 64

Figure 12. Roomba brush bearing covered with plastic sleeve — better than my washer quick-fix.

shaft, and cut away any hair trying to bind to the shaft. I forwarded the results of my tests to the iRobot engineering staff. Several years later, I noticed iRobot had changed the bushing design for the brushes with the new green plastic collar seen in Figure 12, closely surrounding the bushing assembly.

The Path to the Future of Robotics

into space. Starting with Zip2 and PayPal successes, Elon Musk went on to found SpaceX and Tesla, not because he was a ‘rocket scientist’ or car maker, but because he had a dream and the drive to make it come true. Helen Greiner, Colin Angle, and Rodney Brooks (the founder of Rethink Robotics mentioned earlier) created iRobot. The company struggled for years before the Roomba series became a phenomenal success. True entrepreneurs realize that success rarely comes early. They don’t just quit at the first sign of a failure.

How Do You Make a Robot of the Future? I’m not even going to try to lead you directly to a world-class robot design as the variations and design criteria are numerous. I would like to lay out some paths that I’ve observed, however, that seem to lead to success, and other directions that seem to stall. I’ll side-step computer-controlled or remotely-controlled mechanisms, and stick with what we all think of as robots: humanoid styles or at least automatons with manipulative arm(s) and gripper(s). Recent robots have included several telepresence versions that allow someone to be ‘present’ somewhere else, such as at a business meeting or a doctor speaking with a patient in another location. These robots have taken many various forms, such as two-wheeled self-balancing robots to a safer multi-wheeled base.

Let’s look at the path to viable robots for a modern society. Bill Gates, Steve Jobs, Jeff Bezos, Elon Musk, and other modern successful entrepreneurs did not always have the best ideas at the beginning of their careers. What they did have is drive and leadership abilities. Recall the saying: “Genius is 95% perspiration and 5% inspiration.” You can have a fantastic idea, but if you do nothing about it, you’ll see others go right past you with a similar idea and perfect it to a fantastic product. I believe it is a person’s drive rather than their technical expertise that is the core of success. Jeff Bezos started with a great idea of an online bookstore and not only ended up Figure 13. creating the wonder of Renishaw LIDAR Amazon, but is now drawing. building rockets to go

64

SERVO 07.2017

They have satisfied a need for someone to be virtually present in a distant location. Feedback from potential buyers of these types of bots has determined that they could use Skype and a webcam on a laptop at one end of a conference table to accomplish the same thing. Is having a mobile robot platform the ‘cool’ factor that overrides the simplicity of a webcam / laptop setup or even a smartphone on a stand? Building a mobile robot platform is always a more expensive option to using already-developed products. That is something that potential developers must take into consideration. Will their design stand out amongst the competition when they have the same basic features? Will a potential customer always go with a simpler and cheaper solution?

Robot Technology that Needs Improvement Though we have come a long way since the Unimate robot, there remains some very important gaps in needed technology. Sure, memory has advanced from the core and rotating drum variety in early industrial robots to gigabyte RAM and powerful processors, but advances are needed in sensors, usable AI for robots, and power systems.

Battery Technology Advances Needed for Mobile Robots Power systems for mobile robots are one of the first things that come to mind when I think of advancing robotics. We have slowly crawled out of the pit of liquid lead-acid batteries and NiCads to sealed lead-acid batteries (SLA). SLA batteries are still frightfully heavy, and NiCads have that

Carroll - Then & Now - Jul 17_Then & Now - Sep15.qxd 5/30/2017 7:35 AM Page 65

bothersome memory effect when not discharged completely before charging. So, we went to nickel-metal hydride technology that was lighter, but has a self-discharge rate that is still bothersome. Then, along came the lithium-ion series. These were quite light and the flat polymer versions could be placed into cell phones, tablets, and laptops, but they could catch fire and burn your robot to a smoldering mass. The cylindrical versions could still catch fire, and although the power density was great, it was just not enough for larger mobile robots — especially those with arms or bipedal legs. What’s a robot designer to do?

to the outside world, just as our eyes, ears, nose, and skin are our key senses. Prices have plummeted from upwards of $75,000 for a good LIDAR system (such as the SICK systems) to less than $100 units that are not only good enough for robot navigation but for autonomous automobiles. These ‘eyes’ join special cameras, sonar,

radar, and speech recognition systems to make robots very versatile. A drawing of a LIDAR system in Figure 13 shows how the collimated laser beam is projected up to a rotating 45° mirror to ‘paint’ distant objects as the head rotates. The reflection is then returned to the same mirror to be sent to a receiver. Figure 14 shows a typical

FREE

PCB Layout Software & PCB Schematic Software

Robot Mobility and Manipulation Requires Efficient Designs The use of more efficient motors using brushless technology and rare earth permanent magnets is a step in the right direction, but physical output power still requires a large amount of electrical power input. One example of a more efficient design would be larger wheel diameters for rolling robot platforms to cut down the resistance when going across rugs. As much as walking robots seem to be the ideal solution for certain applications (such as search and rescue scenarios), the DARPA DRC competitions showed just how difficult it is to approximate human balance and dexterity in robots. We “carbon units” can easily scramble off a pile of boards and stones, whereas milliondollar bipedal robots fall flat on their faces. We stuff several multi-axis MEMS gyros and accelerometers in our creations, but rapidly discover that these robots cannot react quickly enough in off-balance situations to prevent falls.

Sensor Technology for Future Robots Sensors are a robot’s connection

3 PCBs

41

$

3 PCBs

+ shipping

61

$

3 PCBs

81

$

+ shipping

+ shipping

2-layer MiniBoard

2-layer MiniBoardPro

4-layer MiniBoardPro

(Shipped in 1 business day)

(Shipped in 2 business days)

(Shipped in 3 business days)

1

DOWNLOAD our free CAD software

2

DESIGN your 2 or 4-layer PCB

3

CHECK your design with xCHECK DRC

4

SEND us your design with just a click

5

RECEIVE top quality boards in just days MADE IN THE USA

Now Offering

SMT STENCILS – Nano Coating available–

| www.expresspcb.com

SERVO 07.2017

65

Carroll - Then & Now - Jul 17_Then & Now - Sep15.qxd 5/30/2017 7:35 AM Page 66

scene as LIDAR sees it. Augmenting LIDAR with ‘intelligent’ cameras, GPS, and even radar and ultrasonic distance sensors can result in an amazing autonomous car, military robot, unmanned aerial vehicle, or a great robot for the home environment. However (and a big however), these systems sometimes fail with deadly results. A Tesla autonomous car failed a year ago when its cameras determined that the white side of a large truck was just white sky, and the car drove right under the truck’s bed at full speed.

must have gotten out of a chair and is walking to my right about two meters.” Seeing an object and knowing what it is might be a bit into the future. We can only hope that our creations can evolve to this level so that they can operate in the continual chaotic, random, and problem-strewn existence that we call life. To a robot, this might seem ridiculous. To us carbon units, this is normal. Figure 14. LIDAR image of street scene.

Final Thoughts

Autonomous cars might see an image that is about six feet tall and recognize it as a human just standing still. Suddenly, that image is now traveling at 15 MPH and the car now thinks that this is a human on a bicycle. Now the image is traveling at 60 MPH, so it must be a motorcycle instead. Each of these represent different traffic scenarios an autonomous vehicle will need to be able to deal with. Let’s consider a scene in a home as a personal assistant robot rolls into a room and ‘sees’ an object about five meters away. It thinks to itself that must be its elderly owner. Suddenly, the object breaks into two parts and the smaller object drops to the level of the floor and quickly moves out of the room. “That must be my owner’s cat,” it ‘thinks.’ Suddenly, the image seems to grow in size and separates into two images — one wider and lower and the other taller, and this one is moving across the room. “My owner

What is the future of robotics? No one can say exactly for sure. I just hope I’ve inspired you to realize the “sky’s the limit” in robot development. The needed advancements that I’ve mentioned are just some of the stumbling blocks that will have to be addressed as we build better and better robots. Autonomous cars will become used to the chaos that we humans create and will be able to smoothly drive around cities and on countryside roads. Robots will be able to take our place on the battlefield, in the air, on and under the sea, and in our homes. We’ll have bipedal humanoid robots taking place in and winning marathons and maybe even Iron Man competitions. Hey, in a few years, a robot could be on “Dancing with the Stars.” The only thing that I can absolutely guarantee is that we will have some fantastic robots in the future — the very near future! SV

Actuonix Motion Devices ..........................7

Hitec .............................................................2

SDP/SI ........................................................13

All Electronics Corp. ...........................13, 15

PanaVise .....................................................12

ServoCity .............................................15, 67

EarthLCD ....................................................13

Pololu ..........................................Back Cover

ExpressPCB ................................................65

Routakit ........................................................3

Artificial Intelligence Robots Nonetheless, that incident was not the only accident caused by an autonomous car as other manufacturers have had similar accidents, though not deadly. The call for AI to be integrated in autonomous cars as well as advanced robots is being heard. Some feel that AI will be the ruin of mankind, and that it will ‘autonomously’ develop superior capabilities in excess of its original programming. Other experts feel that AI will eventually be able to simulate that unique ‘humanness’ that a typical computer and all the sensors available cannot match. Without these capabilities, researchers do not feel that a robot can take the place of humans in all actions and tasks.

66

SERVO 07.2017

Tormach .....................................................11

Advertiser Index

LITTLE ACTUATORS

B I S I S L I O T P I E G S I B • UP TO 67 LB THRUST • UP TO 200 MM STROKE • AVAILABLE AS LINEAR ACTUATORS AND AS LINEAR SERVOS

79

PRICES $ STARTING AT

99 duct Pro ! Info AVAILABLE AT SERVOCITY.COM [email protected] • 620-221-0123

Full Page.indd 67

5/31/2017 12:11:07 PM

THIS SIDE UP

Introducing the Balboa 32U4 balancing robot kit from Pololu.

Find out more at www.pololu.com/balboa

5/1/2017 10:38:06 PM

Full Page.indd 68