Sugeno-type fuzzy time-optimal controller for nonlinear systems

Abstract

Solving classical optimal control problems in nonlinear cases are an arduous task. Determination of time-optimal solutions by solving a two-point boundary-value problem (TPBVPs) is generally difficult. In the present study, an alternative technique to achieve a near-optimal controller for nonlinear systems without solving the TPBVP problem is investigated. The proposed controller uses the phase response trajectories of sliding-mode and fuzzy bang-bang controllers to achieve a near-optimal solution. The input membership function is tuned on-line to improve the time-optimal output. The controller rules are selected to satisfy stability and optimality conditions for the new fuzzy time-optimal controller. The paper describes a systematic procedure for design of the controller to achieve the desired result. To benchmark the controller performance, a sliding-mode controller is used for comparison. Simulation of three nonlinear examples shows promising results © 2012 Elsevier B.V.