Tomorrow (May 1st) the University of Northern Iowa is hosting their 3rd annual “ship in” mini sumo event where they invite people to mail their robot off for competition against students and others.

I’ve sent them The Thin Man for competition this year, we’ll see how well it works (or doesn’t work). My robot is a tad bit complicated to operate and even though I’ve written very detailed instructions as well as made a small video of the operation, I worry. Rick Brooks (maker of ExSpurt, also competing in the event for the second year in a row) likens the experience to your children: you send them off on their own and hope things go well…

The event will have live video broadcast starting at 9am CET and can be viewed with RealPlayer. Visit this page to download the RealPlayer media file (check the upper right corner of the page). If you miss the live broadcast then be sure to look for the replay.

I’ve started work on a new minisumo robot. Unlike all of my other projects, this one is a collaboration between a Danish fellow and myself. He and I have never met in person but we found a common interest in building a new, faster design via the online forum Minisumo.Org.UK.

My first part (the brass wheels) is now complete and I’ll be sending them through the mail to my collaborator. He has already finished his first part (aluminum side panels) and sent them to me. It has been very interesting to build a robot this way, we’ll see how it comes out when we are done.

Since this is a collaborative project I decided to document my machining in greater detail than I’ve done in the past for The Thin Man. I wanted to create a “photo tutorial” that would be simple enough for anyone with a little machining experience could follow and create their own parts for this same minisumo design. Feedback about whether this captures enough detail for you to make your own parts is welcome. Any future parts I make will also be documented in a similar way, making it easy for people to “clone” my work if they would like to.

One of the really fun parts of Robothon is the opportunity to get really good, up-close views of the fantastic robots made by other people. In this case, I’m highlighting Foo made by Pete Miles (that is, Mr. Roboto, if you read Servo magazine).

Pete has spent a lot of time making Foo walk very smoothly, as you can see in this movie. Although it walks slowly, it is very stable. In the background, if you listen very carefully, you can hear Pete explaining his challenge in programming the walking gait.

Biped walkers always get a lot of attention because they look really cool. Watch the movie and tell me if you agree or not!

Robothon is over for another year and it was quite exciting. I only took two robots this year: The Thin Man and Velox, both were for the mini sumo event. I opted to not take any other robots simply because I haven’t spent any time improving them so it was unlikely they’d do well.

The Thin Man took first place in the mini sumo event this year (beating my friend and long-time nemesis Dave Hylands), while Velox suffered some sort of mental break that is likely a shorted wire or toasted component. Given the terrible state of its wiring, I’m not surprised. If I want to run it in a future event it’s going to require whole new sensor and processor boards. I already have a design for both though, so it shouldn’t be too hard.

The Thin Man got quite a bit of attention as it was never beaten by any other robot. One significant improvement for this year was to program four different start modes: start straight, start with a left turn, start with a right turn, or pause a little longer before going straight. I didn’t use the fourth mode but the first three worked really well. The rules of mini sumo dictate that the winner of the previous round places their robot first, which is typically a big disadvantage.

It appears other people didn’t know I could change my start program (perfectly legal and even recommended by Pete Miles in his book if I recall correctly) so they would place their robot where they though my robot wouldn’t go. But The Thin Man went right at them anyways, even when I had to turn first. It was like magic!

My micro sumo robot AL also got quite a bit of attention even though Robothon doesn’t run an event for that class. I take it every year anyways because it is a great demonstration of how even small sumo robots can be well made, very fast and very aggressive.

Once again I was very intrigued with the Robo-Magellan event and will start my own project very soon. I have some sketches done for some parts, and as soon as I have something worth showing I’ll blog about it.

I’ve uploaded quite a few pictures from the event here. Enjoy!

Today Google and the X Prize Foundation revealed their latest challenge: a potential $20 million (and up to $30 million including the bonus challenges) for landing a robot on the moon, roam around a little bit, and send back some video.

Hmmm. Sounds very interesting and is likely a very difficult problem. My first thoughts about the challenge:

• Roaming around on the surface of the moon will be very dusty, thus any exposed gears or moving parts will be gunked up in about 10 seconds
• I have no idea what vacuum does to gear motors, electronics, etc. but you gotta believe it will be different than roaming around in my living room
• Landing on the moon is very challenging; many of the early Russian and US probes crashed
• Soft-landing is hard because there isn’t any atmosphere for parachutes, but air bags might help (I don’t know if anyone has tried this though, I believe rockets are the conventional approach to control descent velocity)
• Sending back good quality video probably isn’t terribly difficult but having a camera that operates after a rough landing and operating in vacuum might be; I have no idea how difficult the antenna aiming would be but I’d expect you could rent time on some serious radio telescopes here on Earth to receive the stream once it was directed in the right direction
• I recall the Russians succeeded in landing a tele-operated rover Lunokhod 1 back in 1970 but likely cost way more than $30 million
• The Lunokhod probes were very successful and not a bad model for this challenge
• Lifting a robot to outer space and on to the moon is really difficult – just to get to orbit requires accelerating to greater than 8,000 meters per second – and I imagine efficient navigation to the moon is quite delicate too
• There is a lot of information available about potential landing sites but to really maximize the prize money you’d have to land near something else (like a NASA Lunar Module or Lunokhod 1)
• It is probably worthwhile to budget for multiple attempts, especially in terms of that soft landing

I’m am really excited about this contest. I doubt I have a chance to participate but you can believe I’m going to follow whomever does make the attempt.

What do you think about this contest? How would you solve some of these problems?