Trajectory tracking control of unmanned marine vehicles with thruster faults based on broad learning system.

The presence of unknown nonlinearities in the model of unmanned marine vehicles is inevitable. This paper proposes a new trajectory tracking control algorithm for nonlinear unmanned marine vehicles with model uncertainty, ocean disturbances and thruster faults. Broad learning system is first employed to approximate the nonlinear functions of the model, and an adaptive mechanism is used to estimate the unknown bounds of disturbances and thruster faults. And sliding mode controller is designed for unmanned marine vehicles with unknown nonlinearity and thruster faults to ensure the asymptotic stability of the tracking trajectory error system. It should be mentioned that the proposed approximator based on broad learning system can reduce the calculation time and decrease the estimation error arbitrarily. Finally, simulation studies have verified the effectiveness of the proposed control scheme. [ABSTRACT FROM AUTHOR]