Design of a fuzzy gain scheduler using sliding mode control principles (Q5947591)

Fuzzy gain scheduling along a slowly time-varying trajectory is presented. The analysis starts from the nonlinear system being linearized around a given trajectory. A special feature of the approach described is that the antecedent parts of the open-loop fuzzy system and the controller depend on the desired state instead of on the current one, which makes the closed-loop fuzzy system linear. The control design is based both on linear control theory and sliding mode control and can be carried out without knowing the desired trajectory in advance. Linearization of the nonlinear system at the center of each fuzzy region leads to the design of a set of linear control laws that locally stabilize the linearized system, and consequently the original nonlinear system at the corresponding operating point. The fulfillment of a sufficient condition checks whether a given hyperbox of errors is attractive or not. It has been shown that an increase of the proportional gain increases the attraction of the hyperbox. The method was tested by the example of a robot arm.