An LMI-based Risk Assessment of Leader-Follower Multi-Agent System under Stealthy Cyberattacks
Primary Investigator:
Inseok Hwang
Sounghwan Hwang, Minhyun Cho, Sungsoo Kim
Abstract
We present a method for quantifying the potential risk from cyberattacks in multi-agent systems (MASs). Since MASs inherently depend on the communication between agents, the security
vulnerabilities of the communication links make MASs more vulnerable to cyberattacks than single-agent systems. Therefore, the impact of cyberattacks could lead to the disruption of performance or the violation of safety. To handle these limitations, we aim to develop a risk assessment method for MASs by applying the reachability analysis which computes the reachable set of the MASs via a Lyapunov function and its corresponding linear matrix inequalities (LMIs). The proposed method can evaluate the potential risk against cyberattacks at the agent and entire system levels by deriving ellipsoidal over-approximated reachable sets. An illustrative example is provided to validate the potency of the proposed method, which shows the risk associated with the formation control of a leader-follower MAS in an environment with scattered obstacles. Finally, we suggest that our risk assessment method can help improve the safety of the MASs in various applications.