Cooperative behavior acquisition of multiple autonomous mobile robots by an objective-based reinforcement learning system

Abstract

The present paper proposes an objective-based reinforcement learning system for multiple autonomous mobile robots to acquire cooperative behavior. The proposed system employs profit sharing (PS) as a learning method. A major characteristic of the system is using two kinds of PS tables. One is to learn cooperative behavior using information on other agents' positions and the other is to learn how to control basic movements. Through computer simulation and real robot experiment using a garbage-collection problem, the performance of the proposed system is evaluated. As a result, it is verified that agents select the most available garbage for cooperative behavior using visual information in an unknown environment and move to the target avoiding obstacles.