A Neuro-Fuzzy Reasoning System for Mobile Robot Navigation

An Autonomous Mobile Robot is an artificially intelligent vehicle capable of traveling in unknown and unstructured environments independently. Among the proposed approaches in the literature to handle the navigation problem of a mobile robot is the simple fuzzy reactive approach. This approach, however, occasionally suffers from two major problems, i.e., escaping from trap situations and the combinatorial explosion of the if-then rules in the inference engine. This paper presents a neuro-fuzzy reasoning approach for mobile robot navigation. The proposed approach has the advantage of greatly reducing the number of if-then rules by introducing weighting factors for the sensor inputs, thus inferring the reflexive conclusions from each input to the system rather than putting all the possible states of all the inputs to infer a single conclusion. Four simple neural networks are used to determine the weighting factors. Each neural network is responsible for determining the weighting factor for one sensor input. Simulation results are presented to demonstrate the merits of the proposed system.