A control scheme based on RBF Neural Network for High-precision Servo System

In this paper, a novel control scheme based on RBF Neural Network is proposed for High-precision Servo System. The aim of this study is to reduce the influence which arises from modeling error, unknown model dynamics, parameter variation and disturbance acted on the practical system and to achieve high tracking precision. This scheme consists of a Neural Network controller (NNC), a Feedforward controller and a Feedback controller. All adaptive learning algorithms in this study are derived from the sense of Lyapunov stability analysis, so that system-tracking stability can be guaranteed in the closed-loop whether the uncertainties occur or not. In order to improve the tracking performance, a feedforward controller is added, whose parameters are obtained when nominal model´s parameters are fixed. With the selection of the poles of system, parameters of the Feedback controller could be determined. Experiment results on 3-axis flying simulator verify the proposed strategy can achieve high tracking precision for real-time position servo system.