Moving Target Defense based Cyberattack Mitigation for Connected Autonomous Vehicles

Abstract: Connected and Autonomous Vehicles (CAVs) are anticipated to offer substantial advantages in terms of safety, mobility, and sustainability. However, their susceptibility to cybersecurity vulnerabilities poses a significant challenge. CAVs are equipped with connectivity capabilities that enable them to exchange information with nearby CAVs within their communication range. Unfortunately, this information is vulnerable to cyberattacks, which can transform benign data into malicious content. When CAVs employ a fixed communication scheme, attackers can easily decipher the cooperative relationships between vehicles and manipulate their behavior by making minor modifications to transmitted information. To address this issue, this study introduces a proactive defense-based vehicle dynamics model that leverages moving target defense strategies to mitigate the impact of cyberattacks on the CAV system. The proposed approach employs active switching strategies to dynamically alter the information flow topology, enhancing the unpredictability of CAV behavior and thwarting attackers from achieving their objectives while preserving the benefits of vehicular cooperation. Numerical studies illustrate the robustness of the proposed strategy against attacks that generate malicious information.