Design and Simulation of an Adaptive Networked Tracking Control System

Abstract

A novel Adaptive Networked Tracking Control System (ANTCS) has been proposed in this paper. The system first uses a data-based modeling technology named fuzzy cluster modeling to obtain the T-S fuzzy model of the process. Then, an equivalent fuzzy singleton model is built which satisfies the invertibiity conditions. Thus, the future control actions can be obtained by prediction, iteration and inversion of the fuzzy singleton model. To improve the control performance of the proposed method, an adaptation strategy is implemented by updating the consequent parameters of the fuzzy singleton model and the corresponding inverse model with least-squares algorithm. Radom time delay in the forward channel can be easily compensated by selecting appropriate control actions from the candidate ones which are generated from the networked controller and transmitted from the controller side to the plant side Simulative results demonstrate the effectiveness of this method.