Gain scheduling based fuzzy controller design

This paper presents a gain scheduling based approach to the design of fuzzy controllers. We demonstrate that the Sugeno fuzzy controller is in fact a special case of gain scheduling where the feedback gains are adjusted ("scheduled") via membership functions in fuzzy if-then rules. When the desired gains at various operating points are available, we can apply the ANFIS learning algorithm proposed earlier to construct a Sugeno fuzzy controller that can generate interpolated control actions. The validity of the proposed method is confirmed through simulation of two well-known control tasks: the cart-pole system and the ball-beam system.<
>