Fault-tolerant control against stuck actuator faults

Abstract

A fault-tolerant control system (FTCS) design technique against stuck actuators is investigated using an iterative learning observer (ILO). The principle of the proposed fault-tolerant control (FTC) is to design a reconfigurable controller using estimated system states, relying on control input adjustments on the redundant actuators to compensate for the effects of stuck actuators. The amount of adjustment is updated based on the transient of fault compensation. The ILO provides both the estimates of the system states and the information on such transients. Multiple faults can also be dealt with. The fault compensation can be carried out swiftly due to the rapid convergence of the ILO. It is shown that the proposed FTCS ensures that the system follows the reference model under both normal conditions and with some stuck actuators. The closed-loop stability of the system is established, and the performance is evaluated using the lateral dynamics of an F-8 aircraft model.