Adaptive fault-tolerant trajectory tracking control of twin-propeller non-rudder unmanned surface vehicles

This paper studies the fault-tolerant trajectory tracking control problem of twin-propeller non-rudder unmanned surface vehicles (TPNR USVs) subject to propeller faults. Firstly, a propeller model of TPNR USVs is constructed by decomposing propeller thrusts on the body-fixed reference frame. A propeller fault model is also established by taking into account floating and loss-of-effectiveness faults. Secondly, to ensure tracking errors stay in reasonable ranges, a novel guaranteed transient performance method is proposed. Meanwhile, corresponding error transformation functions are constructed. Thirdly, by utilizing the excellent nonlinearity approximation performance of neural networks (NNs), an adaptive fault-tolerant trajectory tracking control scheme, which can guarantee TPNR USVs track the desired trajectory quickly and accurately even in the event of propeller faults, is proposed. Finally, the fault-tolerant trajectory tracking performance analysis demonstrates the efficiency of the proposed control scheme.