发表机构
School of Automation Engineering of the University of Electronic Science and Technology of China（电子科技大学自动化工程学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出物理感知雷达Transformer（PART），通过多普勒感知查询初始化等技术，实现仅用110万参数的类无关运动目标检测，在nuScenes数据集上表现优异，对稀有目标和恶劣场景鲁棒。
AI 中文摘要
在闭集标注上训练的检测器可能会遗漏训练分类体系之外的稀有运动目标。汽车雷达提供与类别无关的多普勒运动线索，且受恶劣光照和天气的影响较小，但稀疏、含噪的回波会阻碍类感知3D框检测。当难以恢复完整框几何时，表面位置和速度对运动推理和避障仍有用。本文提出了物理感知雷达Transformer（Physics-Aware Radar Transformer, PART），这是一种完全基于稀疏雷达的检测器，可为每个运动目标假设预测存在置信度、代表性表面点和2D地面平面速度。多普勒感知查询初始化（Doppler-Aware Query Initialization, DAQI）通过在位置和速度维度聚类雷达回波，将与场景无关的学习查询替换为依赖输入的提案，缓解了稀疏场景中的查询-目标分配问题。物理引导交叉注意力（Physics-Guided Cross-Attention, PGCA）将径向-多普勒一致性和雷达散射截面（RCS）纳入查询-点关联过程。不确定性感知监督会随机遮挡真实目标，并为模糊的雷达支持查询分配软存在目标，减少对详尽标注的依赖。仅用110万个参数，PART在nuScenes数据集上达到类无关平均精度（CA-AP）0.8827、平均表面平移误差（mASTE）0.3188米、平均速度误差（mAVE）0.8084米/秒；在标准评估排除的稀有及安全相关类别上达到0.9203的召回率，且在夜间、雨天和严重遮挡场景下仍保持有效。对明显假阳性的检查显示，部分预测对应nuScenes标注中未包含的运动目标。代码和预训练模型权重将在该网址公开提供。
英文摘要
Detectors trained on closed-set annotations can miss rare moving objects outside the training taxonomy. Automotive radar provides category-independent Doppler motion cues and is less affected by adverse illumination and weather, but sparse, noisy returns hinder class-aware 3D box detection. Surface location and velocity remain useful for motion reasoning and collision avoidance when full box geometry is difficult to recover. We present the Physics-Aware Radar Transformer (PART), a fully sparse radar-only detector that predicts existence confidence, a representative surface point, and 2D ground-plane velocity for each moving-object hypothesis. Doppler-Aware Query Initialization (DAQI) replaces scene-independent learned queries with input-dependent proposals by clustering radar returns in position and velocity, easing query-object assignment in sparse scenes. Physics-Guided Cross-Attention (PGCA) incorporates radial-Doppler consistency and radar cross section (RCS) into query-point association. Uncertainty-aware supervision randomly masks ground-truth objects and assigns soft existence targets to ambiguous radar-supported queries, reducing reliance on exhaustive annotations. With only 1.1 million parameters, PART achieves a class-agnostic average precision (CA-AP) of 0.8827, a mean average surface translation error (mASTE) of 0.3188 m, and a mean average velocity error (mAVE) of 0.8084 m/s on nuScenes. It attains 0.9203 recall on rare and safety-relevant categories excluded from the standard evaluation and remains effective at night, in rain, and under severe occlusion. Inspection of apparent false positives shows that some predictions correspond to moving objects absent from the nuScenes annotations. Code and pretrained model weights will be publicly available at https://github.com/sunyinghao-uestc/PART.
Comments8 pages, 5 figures, 4 tables, submitted to 2027 ICRA
