Design of an actuator fault-tolerant controller for an air vehicle with nonlinear dynamics

Abstract

In this paper, a novel method is presented to design an autopilot for an air vehicle with a polynomial nonlinear model. This method employs the nonlinear model directly in the control design process without the need for local linearization about an operating point. It is shown that the control design problem can be formulated as a sum-of-squares optimization problem. This method guarantees exponential stability of the closed-loop nonlinear system by introducing a polynomial Lyapunov function. The nonlinear dynamic model of air vehicles can usually be represented in the polynomial form. Therefore, the proposed method can widely be applied to design an air vehicle autopilot. Besides using the proposed method along with the projection based and online redesign methods, a fault-tolerant controller is designed for the air vehicle. Furthermore, a new approach is developed by combination of these methods to fault-tolerant control system design. The proposed method is applied to design a fault-tolerant controller for a nonlinear pitch-axis model of an air vehicle subject to loss of effectiveness actuator fault. The simulation results show the efficiency of the proposed method.