Robust adaptive control of uncertain nonlinear systems via disturbance observer techniques

Abstract

This paper deals with robust control problems for nonlinear systems with matched uncertainties. In order to handle the uncertainties, the uncertainties are considered as disturbances. First, a disturbance observer is designed and the uncertainties can be rejected by subtracting these estimated values of disturbances from the plant input. Since the disturbance observer is a kind of low-pass filter, it has inevitably estimation errors. To get rid of the effects of these errors, the additional control should be applied. This additional control input is designed using adaptive control scheme. The unknown disturbance error bounds are estimated by adaptive parameter estimation methods and then the control input is calculated using these estimate values. But the control input needs to be switched on the sliding surface and provokes a chattering phenomenon. The control method suggested in this paper can, however, reduce the chattering phenomenon as small as possible by using large disturbance observer gain because the magnitude of the switching control input is dependent on the disturbance estimation errors which are decided by the accuracy of the designed disturbance observer.