In an earlier notebook entry, I talked briefly about my own involvement in designing and organizing robot competitions. I also provided a pointer to retrospective article describing the early years of the AAAI Robot Competitions. The competitions that I had a hand in were inspired by earlier competitions that my students and I participated in. Several of us had the opportunity to participate in a competition at MIT run by Rod Brooks, Anita Flynn and other members of the MIT AI Lab and this, in part, led to my working with Pete Bonasso to organize the first of the robot competitions that were held at the National Conferences on Artificial Intelligence (AI) which are run by AAAI. My early contributions to encouraging innovation in robotics through robotic competitions / celebrations were very quickly eclipsed by much more ambitious and better thought-out competitions. For an example of a more recent competition, check out the Hors d'oeuvres Anyone? event held at NCAI 2000 (AAAI'00). You can find lots of additional information on robots and robotics competitions at the AAAI robotics page.
Another interesting and ambitious competition is the international RoboCup competitions. "RoboCup is an international joint project to promote AI, robotics, and related field. It is an attempt to foster AI and intelligent robotics research by providing a standard problem where wide range of technologies can be integrated and examined. RoboCup chose to use soccer game as a central topic of research, aiming at innovations to be applied for socially significant problems and industries. The ultimate goal of the RoboCup project is: By 2050, develop a team of fully autonomous humaoid robot that can win against the human world champion team in soccer." from the RoboCup Homepage. The last competition was held in Melbourne with previous competitions in such locations as Paris and Stockholm. RoboCup-2001 will be held in Seattle at the same time as the International Joint Conference on AI (IJCAI'01). Featured in RoboCup-2002 in Japan will be the first appearance of the RoboCup humanoid league; there is already a Sony Legged Robot League.
Robot competitions are not by any means restricted to researchers or college students. The KISS Institute for Practical Robotics helps to run a series of competitions involving a game called Botball designed to encourage high school students to learn about robotics by actually building robots. Botball is played on 4' by 8' game board and robots score points by placing black or white Ping-Pong balls into designated scoring positions. The exact rules for the game change from year to year. The robots are Lego hybrids programmed using NQC and Interactive C.
At about the same time that Rodney Brooks began touting the advantages of his subsumption architecture, Mark Tilden came up with his first designs for extremely simple but elegant robots that didn't require conventional computing equipment for their control. Tilden developed an approach that he christened BEAM Robotics which stands for Biology, Electronics, Aesthetics and Mechanics (or, alternatively, Building, Evolution, Anarchy and Modularity).
"The idea is to improve robo-genetic stock through stratified competition and have an interesting time in the process. The science behind the idea stems from current concepts in artificial intelligence (AI), artificial life (ALife), evolutionary biology, and genetic algorithms. It seems that building large complex robots hasn't worked well, so why not try to evolve them from a lesser to a greater ability as mother nature has done with biologics? The problem is that such a concept requires self-reproducing robots which won't be possible to build (if at all) for years to come. A solution, however, is to view a human being as a robot's way of making another robot, to have an annual venue where experimenters can let their creations interact in real situations, and then watch as machine evolution occurs." - from the philosophy of BEAM robotics. You can learn more about BEAM robotics competitions at the BEAM website. Here is a BEAM design for a "photovore" called the SunEater II (click on the image to visit a web page describing the design); one feature of this robot is that it gets all of its power for locomotion from the sun (or other light source) by using a capacitive circuit to store energy until there is enough to power the motors for a short spurt.
You might find it interesting to looks at the rules for Robocup. In the following we look at some of the rules for the F-180 League one of several leagues (there is also a league in which the robots and environment are simulated) that compete regularly. A team consists of no less than one and no more than five robots; there are detailed restrictions on size (e.g., each robot must fit inside of a 18cm diameter cylinder), shape and color. The Minnow robot from CMU is an example of a team player for RoboCup.
A match consists of two halves of 10 minutes each and each team has up to 15 minutes of timeouts (two one-minute timeouts followed by one thirteen-minute timeout) - timeouts can only be called when play is already stopped. The field is approximately 152cm by 274cm which together with the stipulation that players must fit in a 18cm diameter cylinder gives you some idea of how crowded the field can be when both teams are of maximum size.
Wireless communication to computers or networks off the field is permitted but there are restrictions on power and frequency. Mark Sibenac has an interesting survey of wireless technologies for robotic applications. The rules for the game are debated in an open forum. As there are often new innovations in hardware, it can become necessary to revise the rules; obviously, robot players are evolving much faster than human players. Imagine if some humans (and therefore potential soccer players) suddenly evolved the ability to fly or developed legs that were substantially stronger than most other humans. It will be particularly interesting to see the rules that are developed for the first games involving robots and humans.
These are just a few of the robot competitions that inventors, researchers, and avid hobbyists can participate in. Having been involved in several such competitions over the years, both as organizer and as participant, I can only say they are both lots of fun and intellectually stimulating. By the way, I probably would have ignored the work of Mark Tilden if I hadn't picked up the popular science book "Virtual Organisms: The Startling World of Artificial Life" by Mark Ward (St. Martin's Press, 1999) at the library last week. Most of this book is on cellular automata and genetic algorithms but it has some interesting chapters on robotics including material on Tilden's BEAM robotics. I was fascinated with how much Tilden could do with a few components and an elegant design and the descriptions of the robots that he has scuttling and scavenging around his home are particularly interesting.
|
|
|
