A Comprehensive Survey on Driving Compliance Assessment Methodologies for Autonomous Vehicles

Shunran Zhang, Zhengmin Jiang, Ming Sang, Yunjie Han, Huiyun Li, Leong Hou U

Abstract: Autonomous vehicles (AVs) have emerged as a focal point in contemporary technological research, promising significant enhancements in safety and efficiency for transportation systems and heralding a fundamental transformation in future mobility. Despite notable advancements in autonomous driving technology, the full-scale mass production and integration of AVs into transportation networks face substantial challenges, with driving compliance assessment (DCA) of driving behavior being a critical barrier. Compliance assessment encompasses multiple dimensions: from meeting basic safety standards and adhering to traffic regulations to implementing advanced driving protocols. This article systematically elaborates and compares methodologies at various levels of DCA, highlighting the importance and complexity of compliance evaluations in the development of AVs. The survey reveals deficiencies in the existing compliance assessment frameworks for AVs, particularly highlighting their lack of systematic and adaptive capabilities to effectively respond to new scenarios, demands, and technologies. It proposes potential research directions to address these gaps. Through a comprehensive examination of these methodologies, the article aims to provide theoretical and evidence-based guidance for the safe integration of AVs into both current and future transportation systems.

External IDs:dblp:journals/itsm/ZhangJSHLU25