SoAD: Safety-Oriented Value Estimation for Enhanced Closed-Loop End-to-End Autonomous Driving
IEEE Transactions on Systems, Man, and Cybernetics: Systems · 2026-08 · 56 (8) · 4942-4955
DOI: 10.1109/tsmc.2026.3688954 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2026-08. Verfasst von Yinfeng Gao und 5 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 56 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Beijing Natural Science Foundation-Xiaomi Innovation Joint Fund, Beijing Natural Science Foundation.
Abstract (Originalsprache)
End-to-end (E2E) autonomous driving systems, which map sensory inputs directly to vehicle planning, have garnered attention for harnessing the potential of data-driven methodologies in motion planning. However, current methods face two limitations that undermine their safety performance in closed-loop driving tasks. First, the predominant imitation learning (IL) paradigm overlooks long-term safety beyond predefined planning horizons, potentially guiding the ego vehicle into hazardous states. Second, the lack of reliable online evaluation mechanisms limits real-time responses to safety risks. To overcome these challenges, we propose SoAD, a safety-oriented E2E framework that integrates long-term safety awareness into planning. SoAD is distinguished by a reinforcement learning (RL)-based value estimation module to quantify the safety of planned trajectories, and a vector world model (VWM) to generate interaction-aware future rollouts. During training, the system benefits from value-guided fine-tuning (VFT) that optimizes the planning distribution to favor safer trajectories. In closed-loop deployment, a planning rescoring (PRS) mechanism is designed to perform reliable online evaluation by combining ego-conditional predictions from the VWM with corresponding safety value estimates. Experimental results on the Bench2Drive closed-loop benchmark demonstrate the state-of-the-art (SoTA) performance of SoAD, achieving a 66.40% improvement in driving score (DS) compared to the vectorized scene representation for efficient autonomous driving (VAD) baseline, while zero-shot evaluation on the driving in occlusion simulation (DOS) benchmark further highlights its strong generalization ability.
Förderung
- National Natural Science Foundation of China
- Beijing Natural Science Foundation-Xiaomi Innovation Joint Fund
- Beijing Natural Science Foundation
Zitieren
APA
Gao, Y., Liu, D., Zheng, Y., Zhang, Q., Ding, D., & Zhao, D. (2026). SoAD: Safety-Oriented Value Estimation for Enhanced Closed-Loop End-to-End Autonomous Driving. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 56(8), 4942-4955. https://doi.org/10.1109/tsmc.2026.3688954
Vancouver
Gao Y, Liu D, Zheng Y, Zhang Q, Ding D, Zhao D. SoAD: Safety-Oriented Value Estimation for Enhanced Closed-Loop End-to-End Autonomous Driving. IEEE Transactions on Systems, Man, and Cybernetics: Systems. 2026;56(8):4942-4955. doi:10.1109/tsmc.2026.3688954
BibTeX
@article{Gao2026,
author = {Gao, Yinfeng and Liu, Deqing and Zheng, Yupeng and Zhang, Qichao and Ding, Da-Wei and Zhao, Dongbin},
title = {SoAD: Safety-Oriented Value Estimation for Enhanced Closed-Loop End-to-End Autonomous Driving},
journal = {IEEE Transactions on Systems, Man, and Cybernetics: Systems},
year = {2026},
volume = {56},
number = {8},
pages = {4942--4955},
doi = {10.1109/tsmc.2026.3688954}
}
EUVIMED