Dual-Channel Event-Triggered Prescribed Performance Control for USV Under False Data Injection Attack

To address the trajectory tracking problem of networked unmanned surface vessels (USVs) under false data injection (FDI) attacks, an adaptive neural network-based prescribed performance control scheme is proposed. First, considering the adverse effects of network attacks, system uncertainties, and time-varying disturbances, an adaptive neural network observer is designed to estimate and compensate for the lumped disturbances. Building on this, a dynamic event-triggered mechanism is separately developed for the sensor-to-controller and controller-to-actuator channels, forming a novel dual-channel dynamic event-triggered mechanism (DDETM). This mechanism reduces unnecessary communication overhead and actuator wear caused by frequent data exchanges while enabling thrust allocation for a quantitative analysis of actuator degradation. Furthermore, a control algorithm based on a second-order prescribed performance function (SOPPF) and dynamic surface control (DSC) is proposed to ensure transient and steady-state performance of the tracking error while mitigating the computational complexity associated with the traditional backstepping method. Using Lyapunov theory, it is demonstrated that all signals in the closed-loop system are uniformly ultimately bounded and that Zeno behavior is avoided. Simulation results further validate the effectiveness of the proposed control approach in solving the trajectory tracking problem of USV.