Attack-Resilient Distributed Prescribed-Time Formation Control for Underactuated Unmanned Surface Vehicles Subject to Pose-Measurement False-Data Injection Attacks

This paper investigates an attack-resilient distributed prescribed-time formation control problem for underactuated unmanned surface vehicles (USVs) subject to pose-measurement false-data injection (FDI) attacks, input saturation, external disturbances, collision-avoidance constraints, and connectivity-maintenance requirements. A secure-pose reconstruction layer is first developed to estimate the injected pose attack and remove its influence from the received position and yaw measurements. Based on the reconstructed pose, the distributed formation variables, prescribed-time transformed errors, virtual control laws, dynamic filters, disturbance observers, saturation auxiliary signals, adaptive laws, and actual surge-yaw inputs are redesigned. The residual propagation from pose-estimation errors to distance and bearing errors is explicitly characterized, and tightened prescribed-time performance boundaries are introduced to preserve the safety constraints for the true USV positions. Lyapunov analysis proves that the closed-loop error signals are uniformly ultimately bounded and that the prescribed-time formation-performance envelope is maintained under pose-measurement FDI attacks. Numerical simulations demonstrate that the proposed secure-pose compensation scheme improves formation tracking and safety performance compared with the uncompensated attack case.