The scope of this paper is to design an implicit self-tuning fuzzy controller suitable for a class of nonlinear systems. The approach of parametric fuzzy modeling is used to decompose the nonlinear input-output mappings of a physical system in a universe of discourse constituted by linear models. The control objective, given in terms of an optimal control policy, is achieved through linear control techniques applied to each one of these model components. A fuzzy controller infers a composite control action based on the control laws designated for the individual linear models. A direct adaptive algorithm is introduced to identify online the control parameters. The performance of the self-tuner along with its adaptability against parametric variations is tested in simulation studies for an open-loop unstable nonlinear system.
展开▼
