Investigation of stability and convergence issues for an enhanced model reference neural adaptive control scheme

Abstract

An adaptive control procedure utilising neural networks is presented. The method is based on the model reference control technique and can be applied to discrete-time nonlinear systems of unknown structure. Multi-layered neural networks are used to approximate the plant Jacobian and synthesise the controller. A sufficient condition for the convergence of the tracking error between the desired output and controlled output is presented. Lyapunov theory is used to show that the overall system is stable. Simulation studies show that the proposed scheme performs well even in the presence of dynamic perturbations.<>