Privacy-Preserving Event-Triggered Optimal Control for Helicopter UAV Landing on an Unmanned Surface Vessel Under FDI-Attacks

The autonomous landing of a helicopter UAV on a moving marine surface vessel (USV) presents significant challenges in rescue missions, particularly when cyber-attacks compromise the communication channel. This paper addresses the secure control problem of helicopter landing on a USV under false data injection (FDI) attacks affecting the one-way communication link from the USV to the helicopter. To mitigate the adverse effects of such attacks on transmission integrity, data privacy, and landing stability, a novel resilient masking approach is proposed. Specifically, virtual states are introduced and interconnected with the physical states of the USV such that the landing mission is guaranteed despite unknown FDI cyber-attacks. The dynamics of these virtual states act as a dynamical output mask, mapping the physical states of the USV to virtual states that are transmitted through the communication channel. To further enhance efficiency and reduce communication burden between the controller and the actuator, an event-triggered control (ETC) mechanism is incorporated, whereby control updates are transmitted only when a state-dependent triggering condition is satisfied. This strategy significantly reduces communication cost and control execution effort while preserving closed-loop stability and performance. The helicopter UAV decodes the masked signal via an estimator and is guided toward the USV landing pad using an optimal control strategy based on a reinforcement learning framework with an identifier-actor-critic architecture. Furthermore, position-state constraints are addressed by introducing a universal nonlinear transformation that unifies constrained and unconstrained cases into a single framework. Using Lyapunov stability theory, the boundedness of the actor-critic adaptive laws is rigorously verified. Numerical simulations are carried out to validate the effectiveness of the proposed scheme.