Robust Cyber-Attack Detection in Networked MPC Systems: Application to Vehicle Platoon

This paper proposes a novel robust cyber-attack detection algorithm for networked control systems equipped with model predictive control (MPC). The growing threats of cyber-attacks on control systems necessitate methods that can quickly and robustly identify and isolate such attacks. Existing detection mechanisms often lack sensitivity to weak attacks and robustness against unknown disturbances and noises and need improved solutions. Our method addresses these challenges by combining a moving observation window with [Formula: see text] performance criteria to design a residual generation mechanism that ensures sensitivity to attack signals. The detector parameters are obtained by solving an optimization problem and an adaptive detection threshold with a compensation term is introduced to enhance accuracy. To evaluate the efficiency and robustness of the proposed method, the connected autonomous vehicle platoon problem is studied. In addition to analyzing the complexity of the proposed method, we derive a sufficient condition for the asymptotic stability of the closed-loop vehicular platoon system under cyber-attacks. Simulation results confirm the robustness and effectiveness of our method and demonstrate its ability to detect cyber-attacks even when they are weaker than unwanted signals. This approach not only improves detection accuracy but also ensures robustness, marking a significant advancement in cyber-attack detection for networked MPC systems.