Safety validation for Level 4 autonomous vehicles (AVs) is currently bottlenecked by the inability to scale the detection of rare, high-risk long-tail scenarios using traditional rule-based heuristics. We present Deep-Flow, an unsupervised framework for safety-critical anomaly detection that utilizes Optimal Transport Conditional Flow Matching (OT-CFM) to characterize the continuous probability density of expert human driving behavior. Unlike standard generative approaches that operate in unstable, high-dimensional coordinate spaces, Deep-Flow constrains the generative process to a low-rank spectral manifold via a Principal Component Analysis (PCA) bottleneck. This ensures kinematic smoothness by design and enables the computation of the exact Jacobian trace for numerically stable, deterministic log-likelihood estimation. To resolve multi-modal ambiguity at complex junctions, we utilize an Early Fusion Transformer encoder with lane-aware goal conditioning, featuring a direct skip-connection to the flow head to maintain intent-integrity throughout the network. We introduce a kinematic complexity weighting scheme that prioritizes high-energy maneuvers (quantified via path tortuosity and jerk) during the simulation-free training process. Evaluated on the Waymo Open Motion Dataset (WOMD), our framework achieves an AUC-ROC of 0.766 against a heuristic golden set of safety-critical events. More significantly, our analysis reveals a fundamental distinction between kinematic danger and semantic non-compliance. Deep-Flow identifies a critical predictability gap by surfacing out-of-distribution behaviors, such as lane-boundary violations and non-normative junction maneuvers, that traditional safety filters overlook. This work provides a mathematically rigorous foundation for defining statistical safety gates, enabling objective, data-driven validation for the safe deployment of autonomous fleets.

翻译：当前，L4级自动驾驶汽车（AVs）的安全验证因无法通过传统基于规则的启发式方法规模化检测罕见、高风险的长尾场景而遭遇瓶颈。本文提出Deep-Flow，一种用于安全关键异常检测的无监督框架，该框架利用最优传输条件流匹配（OT-CFM）来刻画专家人类驾驶行为的连续概率密度。与在非稳定、高维坐标空间中运行的标准生成方法不同，Deep-Flow通过主成分分析（PCA）瓶颈将生成过程约束于一个低秩谱流形。这从设计上保证了运动学平滑性，并使得能够计算精确的雅可比迹，从而实现数值稳定的确定性对数似然估计。为解决复杂路口的多模态歧义，我们采用了一种具有车道感知目标条件的早期融合Transformer编码器，其特点是通过直接跳跃连接到流头部，以在整个网络中保持意图完整性。我们引入了一种运动学复杂度加权方案，该方案在无需模拟的训练过程中优先处理高能量机动（通过路径弯曲度和加加速度量化）。在Waymo开放运动数据集（WOMD）上的评估表明，我们的框架针对一组启发式安全关键事件黄金标准实现了0.766的AUC-ROC。更重要的是，我们的分析揭示了运动学危险与语义违规之间的根本区别。Deep-Flow通过揭示分布外行为（例如车道边界违规和非规范路口机动）识别出一个关键的预测性差距，这些行为是传统安全过滤器所忽视的。这项工作为定义统计安全门限提供了数学上严谨的基础，从而为自动驾驶车队的安全部署实现了客观的、数据驱动的验证。