Incorporating Fuzzy Logic to Reinforcement Learning

This paper proposes a sensor-based navigation method that utilizes fuzzy logic into reinforcement learning algorithms for navigation of mobile robot in uncertain environment. The sonar readings are codified in distance notions by fuzzy sets and a modification in the R-learning algorithm by incorporating fuzzy logic is proposed. Fuzzy logic is used for weighting the immediate reward value, that is a variable presents in the most of the reinforcement learning algorithms. The effectiveness of the modified algorithm, R'-learning, is verified in several tests and compared to the performance of the R-learning algorithm.