SaFeR: Safety-Critical Scenario Generation for Autonomous Driving Test via Feasibility-Constrained Token Resampling
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917313981186048 |
|---|---|
| author | Cui, Jinlong Liang, Fenghua Yang, Guo Tang, Chengcheng Cui, Jianxun |
| author_facet | Cui, Jinlong Liang, Fenghua Yang, Guo Tang, Chengcheng Cui, Jianxun |
| contents | Safety-critical scenario generation is crucial for evaluating autonomous driving systems. However, existing approaches often struggle to balance three conflicting objectives: adversarial criticality, physical feasibility, and behavioral realism. To bridge this gap, we propose SaFeR: safety-critical scenario generation for autonomous driving test via feasibility-constrained token resampling. We first formulate traffic generation as a discrete next token prediction problem, employing a Transformer-based model as a realism prior to capture naturalistic driving distributions. To capture complex interactions while effectively mitigating attention noise, we propose a novel differential attention mechanism within the realism prior. Building on this prior, SaFeR implements a novel resampling strategy that induces adversarial behaviors within a high-probability trust region to maintain naturalism, while enforcing a feasibility constraint derived from the Largest Feasible Region (LFR). By approximating the LFR via offline reinforcement learning, SaFeR effectively prevents the generation of theoretically inevitable collisions. Closed-loop experiments on the Waymo Open Motion Dataset and nuPlan demonstrate that SaFeR significantly outperforms state-of-the-art baselines, achieving a higher solution rate and superior kinematic realism while maintaining strong adversarial effectiveness. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_04071 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | SaFeR: Safety-Critical Scenario Generation for Autonomous Driving Test via Feasibility-Constrained Token Resampling Cui, Jinlong Liang, Fenghua Yang, Guo Tang, Chengcheng Cui, Jianxun Robotics Artificial Intelligence Safety-critical scenario generation is crucial for evaluating autonomous driving systems. However, existing approaches often struggle to balance three conflicting objectives: adversarial criticality, physical feasibility, and behavioral realism. To bridge this gap, we propose SaFeR: safety-critical scenario generation for autonomous driving test via feasibility-constrained token resampling. We first formulate traffic generation as a discrete next token prediction problem, employing a Transformer-based model as a realism prior to capture naturalistic driving distributions. To capture complex interactions while effectively mitigating attention noise, we propose a novel differential attention mechanism within the realism prior. Building on this prior, SaFeR implements a novel resampling strategy that induces adversarial behaviors within a high-probability trust region to maintain naturalism, while enforcing a feasibility constraint derived from the Largest Feasible Region (LFR). By approximating the LFR via offline reinforcement learning, SaFeR effectively prevents the generation of theoretically inevitable collisions. Closed-loop experiments on the Waymo Open Motion Dataset and nuPlan demonstrate that SaFeR significantly outperforms state-of-the-art baselines, achieving a higher solution rate and superior kinematic realism while maintaining strong adversarial effectiveness. |
| title | SaFeR: Safety-Critical Scenario Generation for Autonomous Driving Test via Feasibility-Constrained Token Resampling |
| topic | Robotics Artificial Intelligence |
| url | https://arxiv.org/abs/2603.04071 |