Encoder Front Page
SRS Home | Front Page | Monthly Issue | Index
Google
Search WWW Search seattlerobotics.org

Tom Dickens and his Robots

by Steven D. Kaehler (c) 2000

Introduction 

This article is the second of an ongoing series of articles that focuses on a specific member of the Seattle Robotics Society, highlighting them, their interests, and their accomplishments.  I hope these articles will be an entertaining and informative look at members of the SRS, and will provide some insight into what makes robot builders tick. This article spotlights Tom Dickens, an engineer, musician, programmer, and avid robot builder who has achieved notoriety in the SRS for simple but successful machines through his competition in robot contests.

Last month I interviewed Gary Teachout and showed you some of his projects.  This month, I interviewed Tom Dickens, a very active longtime member of the SRS.  Tom has built many different types of robots and says he's had interests in robotics that go back to his childhood.  As a kid, he built  a 5-foot tall Erector set robot with a walkie-talkie in it so that it could speak.  He used an AC-powered motor in it which meant it only went forward and so had to chase it and carry an extension cord.  He says it was pretty silly looking back on it from where he is today, but at the time he thought it was awesome.  Quite frankly, "silly" stuff is how most of us got started.  You never know where things will take you.

Tom has always liked building things and doesn't remember when this started.  He says it must just be the way he's wired.  Since he was a kid he's been interested in computers, programming, and electronics.  He was (and still is) very active in music, playing drums and keyboards.  After high school he had planned on going to WSU to major in math (computer science degrees were not available yet) but decided to pursue his music career.  He spent two years at Green River Community College as a music major (composition and theory), and for fun took a year of electronics and a year of calculus.  After experiencing what is was like being a starving musician for a number of years, he landed a programming job at Boeing. Going to school nights, he earned a BS degree from Henry Cogswell College in Electronic Engineering in '90 and started teaching college classes there in the evenings.  From '95 to '98 he worked on a Masters degree in Computer Engineering from the National Technical University.  With the completion of this formidable effort, he was promoted to professor status at Cogswell.  His teaching at their has been reduced since the college moved its main campus from Kirkland to Everett, but he has been teaching off-hours classes at Boeing and at the University of Washington for the past three years.

During this time he has worked primarily in software design and development for Boeing, but also spent a few years doing embedded system design around the 68HC11 and 68HC12 microcontrollers (which he had learned about from the SRS).  He is currently the software architect on the conversion of a major 3D geometry-based engineering application to 100% Java.

A recent development in Tom's career at Boeing was the attainment of Associate Technical Fellowship (ATF).  This is a Boeing program that recognizes experts in various fields of science, engineering, and business and makes them visible across the company to assist in problem solving and troubleshooting.  His "official" title as an ATF is "software architect for 3D geometry systems, in support of BCAG aerodynamics".  His systems are used for designing the external shape of an airplane.  Here is a snapshot of the Boeing-internal web page for his ATF promotion.  This is a considerable achievement for Tom and required a great deal of effort and persistence on his part.  Congratulations, Tom!

He continues to stay active in music through CD projects and playing at local theaters.  In fact, he's currently playing a Christmas show at the Renton Civic Theater.

Tom's Master's thesis was based on a field of Artificial Intelligence, specifically genetic algorithms.  This technique uses the Darwinian model to "grow" solutions to problems, and the related field of genetic programming actually grows computer algorithms.  He hopes to spend more time exploring these computer genetics techniques and applying them to both engineering problems and robotics systems.

The picture to the left is of "Leo", Tom's 3Kg sumo, winner of the 2000 NorthWest Robot Sumo competition.  It uses 2 electric-drill motors (from garage sales for $1 each), a BOTBoard ('E1), and the motor driver are 5V latching relays for the H-bridges with a single power MOSFET per motor for speed control.  This little guy may not be pretty but he's won a number of sumo contest for Tom without him making any changes.

Sensors include white-line sensors on the 4 corners, a front-looking DIRRS distance sensor, and bumper switches on the 4 sides.  The algorithm used for the last competition only used the 2 front white-line sensors and the front bumper switch (if something is there, push at full speed!).

Tom discovered the Seattle Robotics Society in the late 80s, on a notice posted at Vetco, and has been hooked ever since.  The club met at the North Seattle Community College, quite a drive from Federal Way.  What impressed him the most was the wealth of technical information being shared and encouraged.  These were his kind of people!  He and his wife Shannon served as editors for the Encoder for a while.  He now lives in Issaquah, and finds the drive to Renton Technical College much easier though he'd drive a lot farther just to because of the valuable information exchange.  He currently serves as club secretary and is also on the Board of Directors for the SRS, primarily taking meeting notes and posting them.

Tom really likes the focus on the yearly Robothon (though it's tough to pull them off), and each year they should get better and better.  He pushed for events like this while the club was meeting at NSCC, but at the time, there wasn't someone like Doug Kelly to effectively organize it.  This year's Robothon will build on last year's success and will be viewed as one of the year's premiere Robotics events.  The club needs to keep this momentum going from year to year, but also needs to be careful to keep the focus on autonomous robots.  He doesn't want to see the SRS getting sucked into the current hype of the RC-controlled (tele-operated) robot contests like Battle-Bots or Robot-Wars.  These are fun events to watch, but they are very different from what the SRS is trying to do with Robothon; focusing on autonomous, intelligent, self-governing robots.  Besides, he works too hard building his robots to see them ripped to shreds in a "battle-to-the-death" contest.

What Tom likes most about building robots is the challenge of pulling together of multiple disciplines (electronics, programming, control systems, mechanical, sensors) to create something what is greater than the sum of its parts.  A big challenge he sees today is being able to see our robotic creations from the eyes of the non-robotics person, and to build something they view as interesting and useful.  This is tough since people unfamiliar with the challenges of building a robot usually don't appreciate what it takes to do it.

Tom's current technological push for his robots is incorporating the DIRRS sensor along with the hacked Gameboy camera into his HC11 and HC12 systems.  He has a half-a-dozen of each sitting on his workbench just waiting for the holiday break to start soldering them in, and then teaching his robots how to use them.  He plans to upgrade his sumo robots for the next Robothon as well as try to build a line maze/fire-fighting robot.

The picture to the right shows Tom demonstrating one of his mini sumos (Domey) on a home-made sumo ring at a recent SRS meeting.  He finds that having his own competition arenas is very helpful in fine tuning a robot for the real thing.  Check out the Encoder article he wrote on building your own from ordinary hardware store materials for under $20!

He sees building a robot as a great learning experience, and an opportunity to inspire others to join into the experience.  There is also the personal satisfaction of seeing your robotic creation in action, running the algorithms you designed into it, and trying to come up with better algorithms for it.

For those just starting out, Tom has the following advice:

Tom's somewhat messy 
but well equipped 
workshop

Many "works"
in progress

"FourHC11Bot", a Fire 
Fighter with CD wheels
& 3 DIRRS sensors

"Domey", a mini sumo

(click on the thumbnail images for bigger pictures)

As you might imagine, Tom has a fairly well equipped workshop for building and testing his creations.  One of his latest projects is a fire fighter ("FourHC11Bot") that uses four HC11 processors.  This one isn't quite ready to show at a meeting but will be soon.

You can visit Tom's HC11/12 website which contains a wealth of information about programming with the 68HC11 and 68HC12 processors including code examples.  He primarily uses assembly code for his HC11 and HC12 work.  If you can make the SRS meetings, you'll usually find him with a small crowd gathered around as he demos his mini sumos or shows some project he's been working on.  He's a fun guy to talk to and has great insights on building and programming robots.

Encoder articles Tom has written:

I want to offer special thanks to Tom for helping me do this article in spite of his wife currently being bed-ridden while awaiting the birth of their third child and Christmas only a couple weeks away.  His dedication to this hobby and the SRS is remarkable.  Thank you, Tom.  (Now, go get your wife her favorite beverage, fluff her pillow, and throw another load of laundry in the washing machine.)
 

I hope you've found this article interesting.  If you have suggestions, ideas or comments, please email me.  --SDK