A MITRE ATT&CK-Driven Risk Assessment Methodology for Connected and Autonomous Vehicles

Connected and Autonomous Vehicles (CAVs) operate as complex cyber-physical systems, exposing critical vehicle assets to sophisticated cyber threats that can propagate into unsafe physical behaviors. In this paper, we present a MITRE ATT&CK-driven risk assessment methodology for CAVs, integrating systematic identification of 8 critical assets, adversary-centric threat mapping, and quantitative risk evaluation. We analyze assets across perception, communication, computing and planning, and motion control layers, and identify 22 potential threats mapped to ATT&CK tactics, techniques, and sub-techniques. Threat exploitability is assessed using the Common Vulnerability Scoring System (CVSS) v3.1 and combined with 5 defined impact levels. Risk prioritization is then performed via a 5×5 probability-impact matrix to support actionable cybersecurity decisions, aligned with ISO/SAE 21434. Based on the results, we analyze ITS-specific countermeasures, including secure authentication, integrity verification, encrypted communications, and intrusion detection. The findings underline that integrity and availability represent the most critical systemic risks, especially in motion control and communication subsystems, while also highlighting the value of a behavior-driven knowledge-based approach for guiding cybersecurity decisions in CAVs.