Robust Autonomous Driving via Vulnerability-Aware Co-Optimization Framework Against Safety-Critical Scenarios
IEEE Open Journal of Intelligent Transportation Systems · 2026 · 7 · 1107-1122
DOI: 10.1109/ojits.2026.3680881 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Open Journal of Intelligent Transportation Systems, 2026. Verfasst von Guangyao Bai und 6 weiteren Autor:innen. Die Arbeit nennt 54 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Key Research and Development Program of Henan Province, Key Scientific and Technology Project in Henan Province of China.
Abstract (Originalsprache)
As autonomous driving technology matures, evaluating and improving the robustness of autonomous vehicles is vital. While realistic and challenging long-tail traffic scenarios are crucial for training, their inherent rarity leaves even well-trained systems vulnerable. Although large-scale simulated scenario generation has been explored, it often suffers from a simulation-to-reality distribution gap, and existing methods lack a unified strategy to leverage real-world data to enhance robustness and safety. Complementary to prior work handling safety-critical scenarios at the policy level, we introduce a novel vulnerability-aware co-optimization framework designed to improve autonomous systems in the presence of collision-prone driving scenarios. Our framework establishes a closed-loop process between scenario generation and policy optimization, alternately exposing robustness vulnerabilities and refining policies over time. Specifically, we decompose real-world scenario generation into three tractable sub-tasks: pairwise collision checking, autonomous vehicle on-policy rollout estimation, and surrounding vehicle motion prior sampling. Simultaneously, we iteratively optimize driving policies within these scenarios while enforcing kinematic feasibility and cost constraints. Extensive experiments demonstrate the feasibility and generalizability of our proposed framework in exploring robustness vulnerabilities of autonomous driving systems.
Förderung
- National Natural Science Foundation of China
- Key Research and Development Program of Henan Province
- Key Scientific and Technology Project in Henan Province of China
- Project of Joint Graduate-Student Education Base in Henan Province
- Key Project of Collaborative Innovation in Nanyang
Zitieren
APA
Bai, G., Li, J., Shi, Y., Shi, L., Gao, Y., Zhang, S., & Wang, W. (2026). Robust Autonomous Driving via Vulnerability-Aware Co-Optimization Framework Against Safety-Critical Scenarios. IEEE Open Journal of Intelligent Transportation Systems, 7, 1107-1122. https://doi.org/10.1109/ojits.2026.3680881
Vancouver
Bai G, Li J, Shi Y, Shi L, Gao Y, Zhang S, et al. Robust Autonomous Driving via Vulnerability-Aware Co-Optimization Framework Against Safety-Critical Scenarios. IEEE Open Journal of Intelligent Transportation Systems. 2026;7:1107-1122. doi:10.1109/ojits.2026.3680881
BibTeX
@article{Bai2026,
author = {Bai, Guangyao and Li, Jie and Shi, Yucheng and Shi, Lei and Gao, Yufei and Zhang, Shuqin and Wang, Wei},
title = {Robust Autonomous Driving via Vulnerability-Aware Co-Optimization Framework Against Safety-Critical Scenarios},
journal = {IEEE Open Journal of Intelligent Transportation Systems},
year = {2026},
volume = {7},
pages = {1107--1122},
doi = {10.1109/ojits.2026.3680881}
}
EUVIMED