Robot Path Planning Using Q- Learning Algorithm

In order to reduce the loss of resources and improve the efficiency of the system, efficient methods are required to solve mobile robot maze. In this paper, the Q-learning method is applied to enable the constantly update of the Q table for robot under the feedback of the maze. Specifically, it is essentially an iterative process, which adjusts the path during moving and uses the optimal path to find the correct exit eventually. By comparing the way in different environment we draw a conclusion that through Q-learning the robot can find the shortest way to solve the maze.