This paper deals with a local learning method of a multi-objective behavior coordination for a mobile robot. The multi-objective behavior coordination plays a role in integrating outputs of basic behavioral modules. To adapt the robot to dynamic environments with moving obstacles and other robot, we propose a local episode-based learning which is a learning method using self-reference of the relationship between previous perception and action in short-term memory. Simulation and experimental results show the effectiveness of the proposed method.
展开▼