Q-RAN: A Constructive Reinforcement Learning Approach for Robot Behavior Learning

Li Jun, Achim Lilienthal, Tomas Martinez-Marin, Tom Duckett
2006 2006 IEEE/RSJ International Conference on Intelligent Robots and Systems  
This paper presents a learning system that uses Qlearning with a resource allocating network (RAN) for behavior learning in mobile robotics. The RAN is used as a function approximator, and Q-learning is used to learn the control policy in 'off-policy' fashion that enables learning to be bootstrapped by a prior knowledge controller, thus speeding up the reinforcement learning. Our approach is verified on a PeopleBot robot executing a visual servoing based docking behavior in which the robot is
more » ... quired to reach a goal pose. Further experiments show that the RAN network can also be used for supervised learning prior to reinforcement learning in a layered architecture, thus further improving the performance of the docking behavior.
doi:10.1109/iros.2006.281986 dblp:conf/iros/LiLMD06 fatcat:iobskkcv4vgdrebprsar7sn7ga