IJCAS

This article investigates the adaptive tracking control problem for the marine surface vessels (MSVs) with unknown uncertainties and multiple constraints. Firstly, a novel prescribed performance-based time-varying asystematic barrier Lyapunov function (PP-TABLF) is proposed to control the MSVs to navigate in a variable narrow waterway and to improve the transient performance of MSVs. Secondly, the predictor-based radial basis function neural networks (P-RBFNNs) are developed to approximate the system uncertainties and external disturbances. Specifically, the tracking errors of the RBFNNs are estimated in advance, and the prediction errors are utilized to update the RBFNNs and improve the estimation precision. The command filter and the idea of the recursive sliding mode are integrated with the control law to limit the amplitude of the virtual control signals and to reduce the computational burden. With the proposed control scheme, tracking errors will not override the prescribed performance ranges, and the control force will not be violated in the presence of total unknown uncertainties. Finally, the semi-global uniformly ultimate boundedness of the system is guaranteed by the proposed control scheme, and the simulation results are given to further demonstrate the effectiveness of the proposed approach.

Keywords: Barrier Lyapunov function, command filter, marine surface vessels (MSVs), predictor-based neural networks, prescribed performance, trajectory tracking.