SWITCHING ROBUST ADAPTIVE CONTROL BASED ON RBF NEURAL NETWORKS

Abstract

The paper deals with robust adaptive control of a class of single-input single-output nonlinear system, in which robustness is guaranteed by switching control algorithm and adaptation law using smooth gradient projection. It is discussed the behavior of the control system, when the nonlinear part in the model of the controlled system is not known exactly. A modified control law is proposed that assures the boundedness of all the signals of the control system even if the nonlinear model contains unmodelled disturbance.