Nonlinear systems identification and control using dynamic multi-time scales neural networks

An on-line identification algorithm via dynamic neural networks with different time-scales followed by controller design is proposed for the dynamic systems with nonlinearity and uncertainty in this paper. The main contribution of the paper is that the Lyapunov function analysis, singularly perturbed technique and sliding mode methodology are combined to develop the control laws for trajectory tracking with consideration of the modeling error and disturbance. Simulations are given to demonstrate the effectiveness of the theoretical results.