![Author ORCID: We display the ORCID iD icon alongside authors names on our website to acknowledge that the ORCiD has been authenticated when entered by the user. To view the users ORCiD record click the icon. [opens in a new tab]](https://www.cambridge.org/engage/assets/public/coe/logo/orcid.png){kind=logo}
Abstract
As the development and deployment of autonomous vehicles (AVs) accelerate globally, the analysis of traffic accidents—particularly their causes, contexts, and outcomes-offers critical insights for improving AV safety, perception systems, and decision-making algorithms. This paper provides a comprehensive review and synthesis of recent research that investigates how learnings from traditional traffic accidents can inform the design, regulation, and behavior of autonomous driving systems. We categorize the implications into technological, behavioral, infrastructural, and regulatory domains. Furthermore, we integrate lessons from sensor failures, human factors, and intelligent transport systems (ITS) to propose a holistic framework for AV accident prevention. The findings underscore the necessity of incorporating crash reconstruction data, behavioral modeling, and multi-agent safety learning into AV development pipelines.
