Design and experimental validation of UDE based controller--observer structure for robust input--output linearisation

Abstract

In this work, uncertainty and disturbance estimation (UDE) technique is employed to robustify an input--output linearisation (IOL) controller. An IOL controller designed for a nominal system is augmented by the UDE estimated uncertainties to achieve robustness. In doing so, state dependent nonlinearities of the system are treated as a part of the uncertainties and thus, the controller does not require system states for its implementation. The resulting controller, however, needs derivatives of the output. To address the issue, a design of an observer that employs the UDE estimated uncertainties for robustness is proposed giving rise to the UDE based controller--observer structure. Closed loop stability of the overall system is established. The notable feature of the proposed design is that it neither requires accurate plant model nor system states or derivatives of output. Also the approach does not need any information about the uncertainty. To demonstrate the effectiveness, numerical simulation results of the proposed approach as applied to the wing-rock motion control problem are presented. Lastly, hardware implementation of the UDE based controller-observer structure for motion control of Quanser's DC servo motion control platform is carried out and it is shown that the proposed strategy offers a viable approach for designing implementable robust IOL controllers.