Q learning behavior on autonomous navigation of physical robot

摘要

Behavior based architecture gives robot fast and reliable action. If there are many behaviors in robot, behavior coordination is needed. Subsumption architecture is behavior coordination method that give quick and robust response. Learning mechanism improve robot's performance in handling uncertainty. Q learning is popular reinforcement learning method that has been used in robot learning because it is simple, convergent and off policy. In this paper, Q learning will be used as learning mechanism for obstacle avoidance behavior in autonomous robot navigation. Learning rate of Q learning affect robot's performance in learning phase. As the result, Q learning algorithm is successfully implemented in a physical robot with its imperfect environment.