Genetic Programming Produced Competitive Soccer Softbot Teams for RoboCup97

At RoboCup, teams of autonomous robots or software softbots compete in simulated soccer matches to demonstrate cooperative robotics techniques in a very difficult, real-time, noisy environment. At the IJCAI/RoboCup97 softbot competition, all entries but ours used human-crafted cooperative decision-making behaviors. We instead entered a softbot team whose high-level decision making behaviors had been entirely evolved using genetic programming. Our team won its first two games against human-crafted opponent teams, and received the RoboCup Scientific Challenge Award. This report discusses the issues we faced and the approach we took to use GP to evolve our robot soccer team for this difficult environment.