测试端到端自动驾驶模型在100个高危场景下的安全表现，发现其实际安全能力远低于宣称水平。
Safe2Drive: Evaluating Safe Driving Behaviors of E2E Autonomous Driving Models
- 构建100个真实高危场景，聚焦施工区、乱穿马路和视线遮挡的行人风险。
- 引入安全评分新指标，结果显示顶尖模型安全得分暴跌至40分以下。
- 揭示当前模型对红灯违规、避障延迟等问题缺乏可靠应对能力，适合安全评估研究者参考。
近期端到端（E2E）自动驾驶策略在封闭回路仿真中取得高驾驶分数，但其在常见安全关键场景中的表现尚不明确。本文提出Safe2Drive（S2D），一套基于Bench2Drive的场景扩展，聚焦施工区、行人乱穿马路和被遮挡的弱势道路使用者（VRUs）三类高频路障。S2D新增100个常见但具挑战性的场景，并引入安全驾驶评分（SDS），该指标在原有评估基础上增加碰撞前制动、施工区物体接触、车道居中和行驶平顺性检查。在S2D上评估两种先进模型（LEAD与SimLingo），发现其驾驶分数显著下降：LEAD从Bench2Drive的94.70分降至39.95分，SimLingo从85.07分降至41.00分；且其SDS得分极低（分别为11.85与15.27）。结果表明这些模型存在脆弱的安全行为，如对施工区理解差、闯红灯、对行人反应迟缓或无反应。本研究揭示即使在训练集包含的CARLA城镇中，当前E2E模型仍缺乏可靠的安全部署推理能力。代码与所有100个场景视频将公开发布。
原文摘要 · Abstract (English)
Recent end-to-end (E2E) autonomous driving policies achieve high driving scores in closed-loop simulations. Yet it remains unclear whether these policies handle common safety-critical scenarios. We present Safe2Drive (S2D), a set of Bench2Drive-aligned scenario extensions focused on three frequent families of road hazards: work zones, pedestrian jaywalking, and occluded vulnerable road users (VRUs). Safe2Drive adds 100 common but challenging scenarios and introduces SafeDriving Score (SDS), a safety-centric metric that augments prior evaluators with pre-crash braking, work zone-object contact, lane centering, and smoothness checks. Evaluating two state-of-the-art policies (LEAD and SimLingo) on S2D, we find that their driving scores drop sharply relative to their reported Bench2Drive baselines (LEAD: from 94.70 DS on Bench2Drive to 39.95 DS on S2D; SimLingo: from 85.07 DS on Bench2Drive to 41.00 DS on S2D) and that SDS on S2D is low (11.85 for LEAD and 15.27 for Sim-Lingo). These results are consistent with brittle safe-driving behaviors such as poor work-zone understanding, red-light violations, and late or absent braking for pedestrians. This study highlights a lack of safe behavioral reasoning in E2E models even when tested on CARLA towns that are part of the training set. We plan to release the code and videos for all 100 S2D scenarios.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。