A novel switching control for ship course-keeping autopilot with steering machine bias failure and fault alarm

This paper investigates the switching control of ship course-keeping autopilot (SCKA) with steering machine bias failure. Steering machine bias failure will cause a ship to deviate from the desired heading, especially in course -keeping situations. Consequently, a switching control for SCKA with steering machine bias failure and fault alarm is presented. First, the SCKA model is established and the feasibility of the model is analyzed. Second, sufficient conditions for the system to have passive and H infinity performances are analyzed and given, and the passive robust controller (PRC) and H infinity fault-tolerant controller (HFTC) are designed based on such conditions. Then, an augmented fault observer (AFO) is designed to estimate unknown steering machine bias failure, and the esti-mated value is utilized to compensate the system when bias failure occurs. Subsequently, a fault alarm (FA) is designed to detect bias failure in time, and the switching signal is provided for which controller (PRC or HFTC) should be invoked. Finally, the effectiveness of the method is verified by MATLAB simulation experiments.