Integrating learning with motor schema-based control for a Robot Soccer Team

Abstract

This paper describes a reinforcement learning-based strategy developed for Robocup simulator league competition. In overview: each agent is provided a common set of skills (motor schema-based behavioral assemblages) from which it builds a task-achieving strategy using reinforcement learning. The agents learn individually to activate particular behavioral assemblages given their current situation and a reward signal. The entire team is jointly rewarded or penalized when they score or are scored against (global reinforcement). This approach provides for diversity in individual behavior.