Robust state tracking of nonlinear systems on sliding surface using neural network

In this paper, a new method for robust control of nonlinear systems using neural networks and SMC methodology is proposed. In this method, a radial basis function network is used as a controller whose parameters must be updated. A modified SMC methodology is used to adjust the parameter of controller such that a zero learning error level is reached in one-dimensional phase space defined on the system output. The performance of the proposed method is demonstrated via simulations.