ISO/SAE 21434-Based Risk Assessment of Security Incidents in Automated Road Vehicles

Although numerous automotive security solutions have been presented in the last years, the question of how to properly react to security incidents during vehicle operation has not yet received much attention. In this work, we describe a context-aware scheme for automated road vehicles that assesses the risk of security incidents intending to automatically identify adequate countermeasures. We specifically focus on attack propagation, as related works proved how seemingly uncritical, but compromised vehicle components can cause dangerous situations. Our scheme is inspired by the risk assessment process of the novel ISO/SAE 21434 cybersecurity standard, which uses attack paths to model static threat scenarios. In contrast, our scheme dynamically queries an asset dependency graph once a security incident is reported, in order to identify attack paths leading to pre-assessed damage scenarios. Since the expected damage of a security incident also depends on the vehicle context, we include information such as speed, time, and traffic density into the risk computation. Based on the resulting risk value, the vehicle selects and realizes a compensating (safety) action. Finally, we discuss our scheme and conduct a case study on an automated prototype vehicle.