EUVIMEDEUVIMED entdecken
Crossref

Adversarial Generation and Collaborative Evolution of Safety-Critical Scenarios for Autonomous Vehicles

Jiangfan Liu, Yongkang Guo, Fangzhi Zhong, Tianyuan Zhang, Zonglei Jing, Siyuan Liang, Jiakai Wang, Mingchuan Zhang, Aishan Liu, Xianglong Liu

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (45) · 38926-38934

DOI: 10.1609/aaai.v40i45.41238 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Jiangfan Liu und 9 weiteren Autor:innen. Die Arbeit wurde bisher 8-mal zitiert (Crossref).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The generation of safety-critical scenarios in simulation has become increasingly crucial for safety evaluation in autonomous vehicles (AV) prior to road deployment in society. However, current approaches largely rely on predefined threat patterns or rule-based strategies, which limit their ability to expose diverse and unforeseen failure modes. To overcome these, we propose ScenGE, a framework that can generate plentiful safety-critical scenarios by reasoning novel adversarial cases and then amplifying them with complex traffic flows. Given a simple prompt of a benign scene, it first performs Meta-Scenario Generation, where a large language model (LLM), grounded in structured driving knowledge (e.g., traffic regulations, real-world accident records), infers an adversarial agent whose behavior poses a threat that is both plausible and deliberately challenging. This meta-scenario is then specified in executable code for precise in-simulator control. Subsequently, Complex Scenario Evolution uses background vehicles to amplify the core threat introduced by Meta-Scenario. It builds an adversarial collaborator graph to identify key agent trajectories for optimization. These perturbations are designed to simultaneously reduce the ego vehicle's maneuvering space and create critical occlusions. Extensive experiments conducted on multiple reinforcement learning (RL) based AV models show that ScenGE uncovers more severe collision cases (+31.96%) on average than SoTA baselines. Additionally, our ScenGE can be applied to large model based AV systems and deployed on different simulators; we further observe that adversarial training on our scenarios improves the model robustness. We hope our paper can build up a critical step towards building public trust and ensuring their safe deployment.

Zitieren

APA

Liu, J., Guo, Y., Zhong, F., Zhang, T., Jing, Z., Liang, S., Wang, J., Zhang, M., Liu, A., & Liu, X. (2026). Adversarial Generation and Collaborative Evolution of Safety-Critical Scenarios for Autonomous Vehicles. Proceedings of the AAAI Conference on Artificial Intelligence, 40(45), 38926-38934. https://doi.org/10.1609/aaai.v40i45.41238

Vancouver

Liu J, Guo Y, Zhong F, Zhang T, Jing Z, Liang S, et al. Adversarial Generation and Collaborative Evolution of Safety-Critical Scenarios for Autonomous Vehicles. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(45):38926-38934. doi:10.1609/aaai.v40i45.41238

BibTeX
@article{Liu2026,
  author = {Liu, Jiangfan and Guo, Yongkang and Zhong, Fangzhi and Zhang, Tianyuan and Jing, Zonglei and Liang, Siyuan and Wang, Jiakai and Zhang, Mingchuan and Liu, Aishan and Liu, Xianglong},
  title = {Adversarial Generation and Collaborative Evolution of Safety-Critical Scenarios for Autonomous Vehicles},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {45},
  pages = {38926--38934},
  doi = {10.1609/aaai.v40i45.41238}
}
Kostenlos starten →