Hybrid Intelligent Control for Path Planning Based on Fuzzy Modelization and Reinforcement Learning

摘要

In this paper, the fuzzy modelization and planning is combined with reinforcement learning to perform robust and optimal control for robot navigation. Fuzzy modelization and planning is efficient, but the control is not always optimal. Learning control can obtain the optimal control policy by on-line learning, but it takes long time for the strategy to converge. Hybrid method of the two obtains the advantages of both and has better control performance than either of them. In the computer simulation experiment, the hybrid intelligent method can perform efficient and optimal control for robot navigation in dynamic environments.