ADAPTIVE CONTROLLER FOR SISO LINEAR UNCERTAIN SYSTEM

ZENG, H.; LI, D.; JIANG, X.; NING, X.; LI, Y.

Abstract

In this paper, one adaptive controller is introduced to a class of SISO linear uncertain plant including physical parameter uncertainties. This controller, as alike as PID, is simple and easy to realize; includes integral actions for the compensation of system uncertainties; only needs to tune one coefficient besides the expected dynamical characteristic parameters. Based on the Kharitonov Theorem and Generalization of Kharitonov Theorem (GKT), a set of controller coefficients guaranteeing the robust stabilization to the plant is obtained by analyzing closed loop control system vertex polynomials. To seek an optimized controller for a given plant, within the set of controller coefficients accumulated, we define the objective weighted function between relative integral of the time weighted absolute error and control input. When the minimum objective weighted value, the optimized controller parameter comes forth. Three examples are simulated to test the validity of controller with tuned parameters. Simulations are, under normal system and parameter uncertainties and external disturbance, compared with the differential geometry control method. Results show that this adaptive controller with tuned coefficients has zero steady-state tracking, good adaptability and robustness to uncertainties.

Coresponding author e-mail: zenghh02[at]mails[dot]tsinghua[dot]edu[dot]cn

Session: Robust and Adaptive Control