Slashdot Mirror

First off, this is NOT a new robot. The robot pictured is a commercially available mobile robot called Khepera II. These robots are fairly stupid, but are easily tethered to more capable machines via a 19200 baud serial link. Mostly, they're used in research (usually undergraduate) because there are whole hosts of Matlab libraries available to interface to these buggers.

bot=kopen([0,19200,1]); % open a connection to tethered robot on /dev/ttyS0 at 19200 baud with one second timeout

And so on and so forth. The Khepera robots have been available for many years, along with the k-team matlab resources. That aside, what the robot in question seems to be doing is using the Matlab Neural Network Toolbox to recognize and classify behavior by observation. Sorry folks, but kids at underfunded state schools do this as undergraduate work in AI. This is nothing new.

In my experience, this is correct. I actually competed at the FIRA World Cup in Vienna in 2003.

It started as a simple, week-long assignment in a graduate robotics course, to have a half-popcan sized robot play soccer with a tennis ball. It ended up with the school partly sponsoring me to go to the competition.

Of the five entrants in my category, two were from Germany, one from Thailand, and one from Brazil. The two German competitors (based on the movements of the robots) had extremely similar, if not identical code. But then, of the two, one was a PhD, and the other was his student, writing a thesis on the AI.

I ended up in 4'th place of 5. I had spent a total of about a week and a half on the code, using nearly the same AI that I had developed for my class, with some modifications. The other teams had been competing for several years in this category, and had a more mature battle strategy, and more experience in competing with others. Also, the other teams had a certain advantage, since they were using a newer version of the khepera robot with proximity sensors that worked about twice as far.

In my case, the German, Thailand, and Brazilian team all put much more effort into the programming - but mostly because of differing backgrounds. For me it was a small assignment, for them it was a thesis, or a full semester's effort.

Strangely enough, I was the only person competing from the USA, even though there were a dozen or so categories, and about 20 other countries represented.

http://www.k-team.com/robots/khepera/base.html for more information on the robots.

http://www.fira.net/ for information on the competition.

Despite a persitent and erroneour rumor, Khepera II price is $1'800 -- K-Team S.A. Mobile Robots for Research, Education, and Hobby http://www.k-team.com http://www.hemisson.com

How long before K-Team sue KDE for use of the letter "K"? ;-)

Well, they didn't actually use an entire lamprey brain, just took some tissue and hooked it up to a Khepera robot and some light sensors. The robot then could be made to avoid or follow light sources. There's a was a brief article about it in the New Scientist.