Stable adaptive fuzzy PD plus PI controller for nonlinear uncertain systems

This paper presents an improved version of the stable fuzzy adaptive control structure, which comprises an approximation of the ideal controller and a supervisory controller (Wang, 1993 [2]). The proposed adaptive controller cancels the nonlinear function using a fuzzy system, whose inputs are the feedback error signals instead of the state variables. By using the feedback errors as the inputs to the fuzzy system, improved control and disturbance rejection performance may be achieved as the controller is made more sensitive to the presence of tracking errors. Another useful feature of this structure is that the supervisory controller can be designed without the need of prior knowledge about the plant. Furthermore, the computational burden of on-line tuning can be reduced by disabling on-line adaptation without incurring stability problems. The efficiency of the proposed adaptive fuzzy system is demonstrated through simulation studies.