Attack detection based on invariant state set for SDN-enabled vehicle platoon control system

Abstract

The characteristics of physical constraints and widespread communication of vehicle platoon system render it vulnerable to physical intrusion or cyber attacks. To enhance the fault-tolerant ability and detection accuracy of the software-defined networking enabled vehicle platoon system, we present a novel attack detection method based on invariant state set, which models both control strategies and physical properties of system. More specifically, we construct a distributed information-weighted set-membership filtering estimator to generate invariant state set that always contains actual state of the plant despite the presence of disturbances or uncertainties. Firstly, a convex optimisation issue is formed to determine local state estimation sets with credibility-based consensus weights. Furthermore, a union method using weighted Minkowski sum of local state set is proposed to produce global invariant state set. To verify the reliability of our detection method, we prove that the global invariant state set is contractively invariant set and converges to the domain of consensus attraction. Finally, evaluation results are provided to validate the secure state estimation effectiveness, and attack detection capability of the proposed detection scheme.