Adaptive neural network state predictor and tracking control for nonlinear time-delay systems

A new adaptive nonlinear state predictor (ANSP) is presented for a class of unknown nonlinear systems with input time-delay. A dynamical identification with neural network (NN) is constructed to obtain NN weights and their derivatives. The future NN weights are deduced for the nonlinear state predictor design without iterative calculations. The time-delay and unknown nonlinearity are compensated by a feedback control using the predicted states. Rigorous stability analysis for the identification, predictor and feedback control are provided by means of Lyapunov criterion. Simulations and practical experiments of a temperature control system are included to verify the effectiveness of the proposed scheme.