A Design of Intelligent Controllers Using CMACs and Multilayered Neural Networks

Abstract

Neural-net based controllers have been proposed for nonlinear systems. The reason why the neural network is employed for such systems is that the neural network has the capability of highly approximating nonlinear properties. However, the problem is pointed out which much time is needed until the good control performance is obtained. On the other hand, a CMAC has been proposed by Albus. According to the CMAC, there is little time for training it, although it has an disadvantage that the accuracy of nonlinear approximation is not good.
In this paper, a new design scheme of intelligent controllers fused the multilayered neural network (NN) with the CMAC, in which the NN effectively works in the initial stage of training, and it automatically changes from the NN to the CMAC if the learning progresses. According to the newly proposed scheme, the disadvantages in the NN and the CMAC are supplemented each other, and a good control performance can be obtained with a few training. Finally, the effectiveness of the newly proposed control scheme is numerically evaluated on some simulation examples.