Adaptive sliding mode control for stochastic jump systems against sensor and actuator failures

Abstract

This study investigates the fault-tolerant control problem in the presence of simultaneous actuator degradation and additive sensor fault for a class of stochastic Markovian jump systems, where external disturbance and partially accessible mode information are considered. A new adaptive observer approach is developed to generate the asymptotic estimations of system state, unknown actuator efficiency factor and additive sensor fault. An adaptive sliding mode control law based on the state estimation is proposed to stabilise the resulting closed-loop fault system. Finally, an example of a linearised model of an F-404 aircraft engine system is presented to illustrate the effectiveness of the proposed control method.