Title :
Robots playing to win: evolutionary soccer strategies
Author :
Agah, Arvin ; Tanie, Kazuo
Author_Institution :
Biorobotics Div., AIST-MITI, Tsukuba, Japan
Abstract :
Automatic development and learning of robot soccer strategies are presented in this paper. It is shown that using a novel control system, it is possible to allow teams of robots to acquire strategies for playing a better game of soccer through successive generations, utilizing simulated evolution. A number of soccer techniques, as developed through robot games, are discussed. The mechanism presented in the paper is suitable for other tasks requiring multiple robots to interact and cooperate in teams
Keywords :
cooperative systems; learning (artificial intelligence); mobile robots; path planning; evolutionary soccer strategies; multiple robots; robot games; simulated evolution; successive generations; Automatic generation control; Control system synthesis; Energy efficiency; Laboratories; Mechanical engineering; Multirobot systems; Robot kinematics; Robot sensing systems; Robotics and automation; Strategic planning;
Conference_Titel :
Robotics and Automation, 1997. Proceedings., 1997 IEEE International Conference on
Conference_Location :
Albuquerque, NM
Print_ISBN :
0-7803-3612-7
DOI :
10.1109/ROBOT.1997.620107
