Adaptive sliding mode observer-based fault diagnosis for flight control system

This paper considers the problem of actuator fault diagnosis for a class of systems with two kinds of uncertainties- unstructured uncertainty and structured one. To detect the unknown fault signals, the observer is designed for a system augmented by a dynamical model of the fault signals. It is well known that sliding mode observers designed for systems with matched uncertainty require an upper bound to be known exactly. To remove this requirement, a novel sliding mode observer is proposed. The novelty lies in the mechanism introduced to update the sliding mode observer gain for counteracting uncertainty. As a result, the knowledge of its upper bound is not required. It is proved that the observer can guarantee the state estimation error convergent asymptotically. Finally, a simulation study of an aircraft system is presented to show the effectiveness of the proposed scheme.