Identification and control of nonlinear systems using dynamic neural networks

In this paper, the adaptive identification and control for continuous-time nonlinear systems are addressed using dynamic neural networks. The control scheme includes two parts: a dynamic neural network is employed to perform system identification, and then a controller based on dynamic neural network model is developed to track a reference trajectory. Stability analysis for the identification and tracking errors is performed. Finally, we illustrate the effectiveness of these methods by simulations of the Duffing chaotic system and the one-link rigid robot manipulator.