Reconfigurable control of nonlinear Lipschitz systems after actuator faults

Abstract

In this paper control reconfiguration for nonlinear Lipschitz systems is done based on a new fault hiding approach in order to maintain the closed-loop stability after actuator fault. The virtual actuator block, placed between faulty plant and nominal controller, generates useful input signals for faulty plant using output signals of the nominal controller. The input-output behavior of the reconfigured plant is equivalent to the behavior of nominal plant and therefore, the effect of fault is hidden form the nominal controller point of view. Parameter design of the virtual actuator is accomplished using LMI, which guarantees both fault hiding and closed loop stability of reconfigured closed-loop plant. The proposed algorithm is simulated on a Single-Link Robot Arm and the results prove the efficiency of the algorithm.