To address the severe safety challenges of AI-driven autonomous driving under the Autonomy 2.0 paradigm, a joint research team from Tongji University, SAMR Defective Product Recall Technical Center, and Tsinghua University, proposed a hierarchical risk-filtering safety assurance framework, establishing a four-category assessment metric system and layered safety strategies, with a dedicated safety monitor for system-level architecture design.
The team published their study in Communications in Transportation Research ( https://doi.org/10.26599/COMMTR.2026.9640018 ).
“We systematically classify state-of-the-art safety assessment metrics into four core categories: scenario criticality, situation complexity, scene consistency, and vehicle-self safety handling envelope, integrating both external environmental dynamics and internal vehicle physical constraints,” says Weiwei Zhang, a researcher at the School of Automotive Studies, Tongji University. “This unified metric framework resolves the fragmentation and inconsistency of existing evaluation standards, providing a common risk language for AI-driven AD safety research.”
Based on the four-category risk perception metrics, the research team designed a time-depth hierarchical safety assurance strategy covering three core measures aligned with the risk evolution timeline: proactive safety behaviors (10–3 s) for long-term strategic risk prevention, reactive safety responses (3–1 s) for real-time hazard mitigation, and emergency evasive maneuvers (0.1–1 s) for instant collision avoidance. These measures realize layered risk filtering by matching different risk levels of driving scenarios, effectively avoiding the misalignment of safety responses and actual risks.
A key innovation of the study is the design of a system-level safety assurance architecture embedded with a dedicated safety monitor unit. This unit captures and calculates the four-category assessment metrics in real time, and the architecture is fully compatible with mainstream end-to-end AD systems. Through hierarchical validation and risk inhibition, the architecture effectively expands the safety Operational Design Domain (ODD) boundary of AI-driven AD systems, a critical issue for the commercial deployment of SAE L3/L4 autonomous vehicles.
Lingyun Xiao, a professor-level senior engineer at SAMR Defective Product Recall Technical Center and corresponding author of the study, notes that the traditional scenario-based safety assurance framework lacks dynamic risk evolution tracking and hierarchical risk control. “Our hierarchical risk-filtering framework bridges the gap between fragmented research on safety metrics, strategies and architectures in the field. It establishes a logical closed loop from risk perception to safety response and architectural implementation, providing actionable technical guidance for accident prevention of AI-driven autonomous vehicles.”
The research team also conducted in-depth verification of the framework’s effectiveness through typical scenario validation and industrial case integration, including its application in highway driving and urban intersection scenarios. The results show that the framework can significantly improve the operation safety of end-to-end AD systems.
About Communications in Transportation Research
Communications in Transportation Research was launched in 2021, with academic support provided by Tsinghua University and China Intelligent Transportation Systems Association. The Editors-in-Chief are Professor Xiaobo Qu, a member of the Academia Europaea from Tsinghua University, and Professor Xiaopeng (Shaw) Li from University of Wisconsin–Madison. The journal mainly publishes high-quality, original research and review articles that are of significant importance to emerging transportation systems, aiming to serve as an international platform for showcasing and exchanging innovative achievements in transportation and related fields, fostering academic exchange and development between China and the global community.
It has been indexed in SCIE, SSCI, Ei Compendex, Scopus, CSTPCD, CSCD, OAJ, DOAJ, TRID and other databases. It was selected as Q1 Top Journal in the Engineering and Technology category of the Chinese Academy of Sciences (CAS) Journal Ranking List. In 2022, it was selected as a High-Starting-Point new journal project of the “China Science and Technology Journal Excellence Action Plan”. In 2024, it was selected as the Support the Development Project of “High-Level International Scientific and Technological Journals”. The same year, it was also chosen as an English Journal Tier Project of the “China Science and Technology Journal Excellence Action Plan PhaseⅡ”. In 2024, it received the first impact factor (2023 IF) of 12.5, ranking Top1 (1/58, Q1) among all journals in "TRANSPORTATION" category. In 2026, its 2025 IF was announced as 12.7, maintaining the Top1 position (1/66, Q1) in the same category.
From Volume 6 (2026), Communications in Transportation Research will be published by Tsinghua University Press on the SciOpen platform with the official journal website at https://www.sciopen.com/journal/2097-5023 . We kindly request that all new manuscript submissions be made through the journal’s submission system at https://mc03.manuscriptcentral.com/commtr . For any submission-related inquiries, please contact the Editorial Office at commtr_e@mail.tsinghua.edu.cn.
Communications in Transportation Research
Survey on Building Operation Safety Assurance Framework towards AI-driven Autonomous Driving: A Hierarchical Risk-Filtering Perspective
30-Sep-2026