Adaptive bounded neural network control for coordinated path-following of networked underactuated autonomous surface vehicles under time-varying state-dependent cyber-attack.

This paper is concerned with the problem of coordinated path-following for networked underactuated autonomous surface vehicles in the presence of time-varying state-dependent cyber-attack. An adaptive bounded neural network controller is proposed to mitigate the malicious effect of the cyber-attack. At first, an individual path-following control law is designed for each vehicle by fusing a back-stepping technique, a line-of-sight guidance principle and a predictor-based neural network method. Second, a path update law is developed based on a synchronization approach together with an adaptive control method. The salient features of the proposed controller are presented as follows. First, an adaptive corrective signal is incorporated into the path update law design such that a desired formation can be achieved regardless of the time-varying state-dependent cyber-attack. Second, by using a saturation function and a projection operator, the proposed controller is bounded and the bound is known as a priori. It is proven that the closed-loop system is input-to-state practical stable in the face of time-varying state-dependent cyber-attack. Simulation results show the effectiveness of the proposed adaptive bounded neural network controller for coordinated path-following of networked underactuated autonomous surface vehicles subject to the cyber-attack.