ODFa${}^{2}$: Overall Defense Framework Against Cyber-Attacks on Intelligent Connected Vehicles

Autonomous driving, as an emerging technology, is developing rapidly. However, a number of obstacles appear to be holding back the widespread adoption of autonomous driving. Cybersecurity issues, which have a significant impact both technically and psychologically on people's willingness to use autonomous driving, are one of the most significant barriers. Attackers have adopted increasingly diverse threat methods, increasing cyber-attacks. Despite these threats, global standards have been developed for road vehicle safety and cybersecurity. These standards are intended to guide manufacturers of automotive systems. For instance, the ISO and SAE have jointly released the international standard ISO/SAE 21434 (Road Vehicles—Cybersecurity Engineering) for road vehicle cybersecurity. However, focusing on the development of onboard systems alone when examining the application of these standards is not sufficient. Currently, we lack a methodology to directly apply the ISO/SAE 21434 standard to automotive system engineering projects. In particular, the overall framework of cybersecurity design for intelligent connected vehicle systems needs to be developed. This paper proposes an overall cybersecurity assurance framework (ODFa
${}^{2}$
) based on the ISO/SAE 21434 standard to assist system engineers in understanding security risks, evaluating the feasibility of attack implementation, and determining the trade-off methods for security countermeasures. We illustrate through a case study how this model-based framework may be applied to cybersecurity engineering in intelligent connected vehicles and mobility service platforms. Furthermore, the results of comparative experiments indicate that the proposed strategy optimization algorithm outperforms existing algorithms in terms of time efficiency under certain conditions.