Research On Fault-tolerant Control Technology For Hypersonic Vehicles Underactuator Failures

This paper addresses the fault tolerant control techniques for longitudinal dynamics and attitude model of hypersonic vehicles in the presence of actuator failures.Firstly, the knowledge about hypersonic vehicles and the features of reentry mode are introduced. And the background and the significance of the research are summarized. Furthermore, we overview the state-of-the-art of fault-tolerant control for hypersonic vehicles subject to actuator failures.Secondly, an adaptive fault-tolerant controller based on feedback linearization is developed for the longitudinal dynamics of hypersonic flight vehicles in the presence of actuator failures and parameter uncertainties. First, the relative degree of the system is obtained by calculating the derivatives of the output repeatedly. Then, the nominal controller is developed based on feedback linearization to make the altitude track the reference signal. When failures happen in the actuators, a fault-tolerant control scheme is designed based on the nominal controller, and the matching conditions for parameters in the fault-tolerant controller are given. Considering that the parameters in the model and the failure information are unknown, adaptive laws are designed based on Lyapunov stability theorem to update the parameters in the fault-tolerant controller.Thirdly, in an actual hypersonic flight, it is practical to measure the states V(velocity), h(altitude) and q(pitch rate). However, it is difficult to measure the angle of attack a and the flight-path angle g since they are generally very small. Hence, the high-gain observer technique is applied to develop an output feedback controller on the basis of the state feedback controller.Finally, a fault-tolerant attitude control scheme for hypersonic reentry mode is developed. According to the features of the reentry mode, Reaction control systems(RCS) are usually activated to assist aerodynamic surfaces in attitude control. Firstly, the Backstepping technique is applied to achieve the desired torque command. Then, as the primary actuators, aerodynamic surfaces are driven to achieve the desired moment. When the desired moment is unattainable using aerodynamic surfaces alone, the reaction control systems are activated to achieve the residual desired moment. When failures occur, observers are designed for the actuator dynamics to estimate the failure information. According to the estimation of the failure information, a fault-tolerant control scheme based on control allocation is proposed to ensure the system stability and recover tracking performance.