Adaptive Security Tracking Control of Constrained Unmanned Surface Vehicles

This paper studies the issue of trajectory tracking control for constrained unmanned surface vehicles with uncertainties and cyber attack. To handle the uncertainties and output constraints, adaptive neural networks (NNs) and barrier Lyapunov functions are respectively employed. The input signals are modeled under the consideration of cyber attack. A control law based on adaptive NNs is proposed to solve the trajectory tracking problem. It is proved that with the proposed control law, the output constraints can be guaranteed, and the expected tracking errors can converge to zero asymptotically when cyber attack occurs. Furthermore, the analysis is extended to situations without output constraints. Finally, the performance of the new control laws are demonstrated through simulation results.