Actuator deception attack detection and estimation for a class of nonlinear systems

In this paper, an novel active safety monitoring system is constructed for a class of nonlinear discrete-time systems. The considered nonlinear system is subjected to unknown inputs, external disturbances, and possible unknown deception attacks, simultaneously. In order to secure the safety of control systems, an active attack estimator composed of state/output estimator, attack detector and attack/attacker action estimator is constructed to monitor the system running status. The analysis and synthesis of attack estimator is performed in the $H_{\infty}$ performance optimization manner. The off-line calculation and on-line application of active attack estimator are summarized simultaneously. The effectiveness of the proposed results is finally verified by an numerical example.