A Self-Tuning Method of Fuzzy Inference Rules by Descent Method

Abstract

In this paper, we propose a self-tuning method of fuzzy inference rules by a descent method. From input-output data, the inference rules expressing the input-output relation of the data, are obtained automatically using the proposed method. The membership functions in antecedent parts and the real numbers in consequent parts of inference rules are tuned by means of the descent method. The learning speed and the generalization capability of this method are higher than those of a conventional backpropagation type neural network. In order to demonstrate these advantages over the conventional neural network, some numerical examples, an application to a mobile robot that avoids a moving obstacle and its computer simulation are reported.