Robust adaptive neural network control for a class of non-linear systems

Abstract

In this paper, a general framework for robust parallel adaptive neural network (NN) control design is presented for a class of non-linear systems motivated by the work in references (14) and (15). The controller is based on applying direct adaptive techniques to an additional parallel neural network to provide adaptive enhancements to a basic fixed controller and incorporating a sliding mode term for robustness. It is shown that if bounded basis function (BBF) networks are used for the additional parallel NN, uniformly stable adaptation is assured and asymptotic tracking of the reference signal is achieved. Because of the introduction of the GL (Ge-Lee) matrices and operator, the results presented here are more general than the existing results.