Adaptive observer‐based attack estimation and event‐triggered security control for UAVs

Abstract

Summary With the increasingly close connection between unmanned aerial vehicles (UAV) and communication networks, it is facing an increasing number of cybersecurity threats. In this article, an event‐triggered security control problem for UAVs is investigated under the tampered input signal sent by the ground control station. First, a linear model of the lateral movement of UAV is established and the attack description is given. Second, an adaptive observer is developed to reconstruct the attack signal. The reconstructed signal is compensated as system disturbance, and then the safety control strategy with event‐triggered mechanism (ETM) is designed. Furthermore, through the time interval analysis approach, the ETM, state observer, controller and attack signal are unified into a new augment system. Then, the uniformly ultimately bounded analysis and controller/observer synthesis conditions are obtained by using piecewise Lyapunov–Krasovskii functional methods. Furthermore, a sufficient condition for the parameters involving controller gain, observer gain, and the ETM parameters is given via a linear matrix inequality. Finally, simulation results are conducted to verify the effectiveness of the proposed method.