Model-free event-triggered resilient control for discrete-time nonlinear systems under sparse actuator attacks via GrHDP

Nonlinear modeling, multi-uncertainty analysis and robust control are the main challenges for networked control applications in industrial scenarios. To address these difficulties, this paper proposes a resilient control scheme for nonlinear cyber-physical systems with unknown disturbances and actuator attacks. First, a novel composite control framework under the strategy of goal representation heuristic dynamic programming (GrHDP) is constructed to mitigate adverse attacks, where a three-player zero-sum game (ZSG) is formulated to solve the optimal policy pair. Then, an adaptive event-triggered mechanism with performance guarantee is designed to save communication resources and improve system resilience, which does not rely on explicit dynamics and achieves both short- and long-term response performance. Under the ZSG issue, a Nash equilibrium-based GrHDP algorithm is developed to implement such a scheme, and the weight updating rules of each network are derived respectively. Furthermore, both the convergence of the performance index and the uniform ultimate boundedness of the closed-loop system are analyzed rigorously. Finally, the effectiveness of the proposed method is verified by an aero-engine networked system.