Mobile robot navigation based on Q-learning technique

This paper shows how Q-learning approach can be used in a successful way to deal with the problem of mobile robot navigation. In real situations where a large number of obstacles are involved, normal Q-learning approach would encounter two major problems due to excessively large state space. First, learning the Q-values in tabular form may be infeasible because of the excessive amount of memory needed to store the table. Second, rewards in the state space may be so sparse that with random exploration they will only be discovered extremely slowly. In this paper, we propose a navigation approach for mobile robot, in which the prior knowledge is used within Q-learning. We address the issue of individual behavior design using fuzzy logic. The strategy of behaviors based navigation reduces the complexity of the navigation problem by dividing them in small actions easier for design and implementation. The Q-Learning algorithm is applied to coordinate between these behaviors, which make a great reduction in learning convergence times. Simulation and experimental results confirm the convergence to the desired results in terms of saved time and computational resources.