A Self-tuning Fuzzy Controller for Networked Control System

This paper mainly deals with the controller design for a class of networked control system (NCS) with the presence of networkinduced delay and data packet dropout. Aiming at the stochastic and time-varying characteristics of delay and dropout, a NCS model with uncertain linear parameters was established by converting the uncertainties of delay and dropout into the uncertainties of the coefficient matrixes of system state equation. Based on the mathematical model, a self-tuning fuzzy controller composed with two basic fuzzy controllers for NCS was proposed. The first-level fuzzy controller was used to control the controlled plant directly. In addition, in order to solve the influences of the uncertainties on the system stability effectively, a second-level fuzzy controller was designed to adjust the quantification and the scaling factors of the first-level fuzzy controller in an on-line and adaptive way according to the system error and the change of the error. Simulation results based on Truetime simulation platform verify the validity of the selftuning fuzzy controller for NCS.