Neural network control systems incorporating parallel adaptive enhancements

Abstract

Neural networks can provide a suitable approach to control of certain classes of nonlinear dynamical systems, and adaptive techniques for neural network control can be incorporated. In this paper, we present neural network control strategies which incorporate an additional parallel neural network to provide adaptive enhancements to the basic fixed neural network-based controllers. These proposed adaptive neural network control systems are applicable to nonlinear dynamical systems of the type commonly encountered in many practical position control servomechanisms. The effectiveness of these controllers are demonstrated in real-time implementation experiments for position control in a servomechanism with asymmetrical loading and changes in the load.<>