Non-linear control with neural networks

This paper is concerned with a non-linear self-tuning tracking problem using back-propagation (BP) neural learning and system identification techniques. Traditional self-tuning adaptive control techniques can only deal with linear systems or special nonlinear systems. BP neural networks have the capability to learn arbitrary non-linearities and show great potential for adaptive control applications. A scheme for combining BP neural networks with self-tuning adaptive control techniques is proposed. Two simple simulation studies are provided to illustrate the effectiveness of the control algorithm. Simulation results indicate that the indentification self-tuning scheme can deal with complex unknown non-linearities