Research on Safety and Security Co-Analysis Methods Based on the AEB System of Intelligent Connected Vehicles

Recently, with the advent of advanced driver assistance systems (ADAS), potential risks of safety-related hazards and security related threats to intelligent connected vehicles (ICVs) have been on the rise. Since safety and security can negatively influence each other, it's necessary to analyze their interplay to achieve a safe and secure system. Unfortunately, neither hazard analysis and risk assessment (HARA) nor threat analysis and risk assessment (TARA) provide specific guidance on the new risks associated with the interaction of hazards and threats. Driven by safety and security requirements, we focus on their interaction especially the conflicts of safety and security countermeasures. This research initially scrutinizes recent literature on safety and security (S&S) co-analysis methods for cyber-physical systems (CPSs), discussing their limitations to address the gaps. Subsequently, we reviewed S&S co-analysis methods, categorizing them into safety-based, security-based, and integrated S&S co-analysis methods. To validate the comprehensiveness, effectiveness, efficiency, and applicability to other domains of the methods, we conducted case studies on the safety-critical autonomous emergency braking (AEB) system. Furthermore, the three types of co analysis methods were compared using three dimensions: completeness, expressiveness, and analysis effort. The results indicated that integrated co-analysis method exhibited a robust capacity for representing interactive relationships, albeit necessitating a more considerable investment of effort from the analysts. In response to the complexities of manual analysis and reliance on expert experience, we have proposed to design an ontology-based automated S&S co-analysis and risk assessment tool in the future to bridge the gap of the inefficiency of S&S co-analysis methods.