Robust adaptive control of time-varying linear plants using polynomial approximation

Abstract

The robust adaptive control of time-varying linear systems in the presence of bounded external disturbances is discussed. A new polynomial approximation scheme is proposed to approximate unknown time-varying parameters of the plant. A modified least-squares estimator with a relative dead zone is applied to identify the resulting unknown parameters. The control law is derived by minimising a quadratic cost function which penalises both model tracking errors and excessive control actions. It is shown theoretically that robust stability can be established in the presence of unmodelled dynamics due to approximate errors and external disturbances. Simulation results demonstrate the effectiveness of the proposed new controller.