We had a chance to catch up with Rick and learn a little more about his Propeller-controlled robot. Here’s what he had to say:
How long have you been competing in RoboMagellan events?
No Name was started in October 2007 with the goal of a completed robot for the ChiBots Spring 2008 contest. Actually, there was no autonomous movement for the Spring 2008 contest. The Fall 2008 contest wasn't much better. Everything stopped working the day of the event. No useful GPS, marginal compass, encoder that worked up to the first run, etc. It was one of those days that you would rather forget. The next few years worked out better.
What are the biggest challenges you face?
The first challenge is that I'm not a programmer. I had done Basic Stamps for six or seven years, but the only previous experience was a Fortran course in the late 1960's. So the Propeller had a steep learning curve for me.
The second challenge (which hasn't been completely resolved) is working with sensors that I hadn't used before. Sensors don't seem to work the way that I think they should. So, each sensor is a separate project to determine how to best use them to get some useful information. In most cases the original choice of sensor was completely inadequate, so a lot of testing was required. That testing was very time consuming.
The third challenge was slowing down and stabilizing the chassis. It required mechanical, electrical, and programming changes to get the desired results.
Why did you choose the Propeller as the brains?
Actually, I had a Propeller Demo Board and couldn't do much with it. One of the major goals of No Name was to force me to learn to use the Propeller. So the Propeller chose the project, not the other way around.
All of the code for No Name came from the Propeller Object Exchange or the Propeller Forum. I just chose the right objects and shuffled them all together to make it work. Without those resources, this project would have never run.
Why is your robot named No Name?
All of my robots have done something very good, very bad, or unexpected that lead to a name. This robot never did any of that sort of thing. When it was entered into its first contest, it didn't have a name, so I wrote down "No Name". That seems to have stuck. I can't change it now.
Tell us more about your robot.
The chassis started out as a Traxxas E-Maxx 16.8V truck. The suspension has been modified to produce a very stable sensor platform up to about 15 mph over grassy surfaces. Rear steer was added to tighten turning at low speed. An E-Maxx can be easily rolled or flipped. A great deal of effort was put into the programing to limit the chances of rolling or flipping. Now the chassis can do some rather violent maneuvers while staying upright.
No Name does not use a video camera to find the orange cones. GPS results can be used to navigate to within a meter or so of the pylon and then the sonar will detect the pylon and drive up to and touch the cone. Someday I'll be motivated to add a camera, but there are several other areas that required extensive improvements first.
What is your advice to new competitors?
- A new competitor should allow an enormous amount of time to prepare for the first contest. This is not a simple, weekend project.
- Do a great deal of testing in a variety of locations. Just because you haven't seen a problem where you are testing, doesn't mean that the problem doesn't exist.
- When you get into trig functions, make sure that you test the results in all four quadrants.
- Keep watching the Propeller Object Exchange and Forums for a better way of solving a problem.
- Keep the R/C as an option. It sometimes helps to introduce situations with radio control to see how the robot reacts.
You can learn more about Rick’s projects by visiting his website: http://www.brooksbots.com/index.html
Here’s a video of his run with NoName at the 2012 competition. http://vimeo.com/47142980