Adaptive fuzzy actuator failure compensation for a class of nonlinear systems

This paper proposes a direct adaptive fuzzy control scheme for actuator failure compensation of a class of multi-input single-output (MISO) nonlinear systems. The proposed controller assumes a proportional actuation scheme which allows transforming the MISO system with uncertain actuator failures to an uncertain SISO system for which a fuzzy logic system is used for designing the control input signal to be distributed among redundant actuators. The parameter update laws for the fuzzy controller are derived using a gradient descent type algorithm that minimizes a quadratic cost function of the control error. Closed-loop system stability and tracking requirements are proved using Lyapunov stability theory and a piecewise analysis to deal with parameter jumps due to abrupt failures. A simulation study is carried out to show the effectiveness of the proposed control scheme.