基于强化学习的机器人模糊控制系统设计

研究了基于强化学习(RL)的模糊逻辑控制器(FLC)设计方法,并将该控制器作为反应式自主移动机器人的控制系统。在缺乏专家知识的情况下,将模糊推理系统(FIS)和强化学习理论相结合构成模糊强化系统,通过强化学习算法获取FLC得模糊规则库,从而有效地解决了复杂未知环境的机器人导航问题。实验结果表明,由强化学习设计的模糊控制器的有效性,同时具有较强的适应能力,可以应用于不同的复杂环境。

Design of Robot Fuzzy Logic Controller Based on Reinforcement Learning

The design problem of the fuzzy logic controller(FLC)was focused on based on reinforcement learning(RL).And the fuzzy logic controller was applied to the reactive robot control system.It could be available without sufficient expert knowledge that the fuzzy inference system(FIS)and reinforcement learning were integrated.The consequence of fuzzy rules was refined through the SRASR reinforcement learning.This scheme could effectively solve the problem of navigation under complicated unknown environment.Experiment result indicates the efficiency and effectiveness of the proposed approach.Furthermore,the FLC learned by RL have robust and adaptability,and it can be apply to different environment.