A new approach of fuzzy control by using neural network

A new approach to fuzzy control using a neural network is proposed in this paper, based on the saturation property of output of the neural networks and partial central symmetry of a simple fuzzy control rule table in some region, in which any smaller domain of a fuzzy relationship can be centrally mapped from the bigger domain in a linear or nonlinear way. The method has the advantages of concise knowledge representation, fast learning, good performance of interpolation, and convenience of hardware implementation. Computer simulations applying the method of reversing by a truck show the method to be feasible