论文arXiv 2611.1327111 天前
RaMOD框架实现原始毫米波雷达RAD立方体实时动目标检测
本文由 AI 根据公开内容整理,可能有误,请以原文为准:RaMOD: Moving-Object Detection from Raw mmWave Radar Cubes under Ego Motion(arXiv 2611.13271)
核心内容
针对车载毫米波雷达感知中原始Range-Azimuth-Doppler(RAD)立方体利用率低、自运动诱导静态杂波及多普勒卷绕导致动目标检测困难的问题,研究提出RaMOD实时检测框架。该方法摒弃传统稀疏点云预处理,直接在原始RAD数据上进行端到端处理。在BRIDGE-Liv数据集测试中,RaMOD在RAD-cube、RA及RD三种设置下均超越现有基线,RAD-cube检测mAP@0.30达80.8%,较最强基线提升15.7个百分点,验证了原始信号处理在具身感知中的有效性。
技术架构上,RaMOD显式估计帧级自运动并利用雷达平移速度预测世界静态结构的多普勒区域以抑制回波,再将抑制后的立方体时序编码为多尺度特征预测带类别标签的RAD空间框。针对每个检测目标,系统在原始多普勒与解卷绕假设下联合预测径向及前向速度,并输出多普勒卷绕概率。实测速度MAE为1.128 m/s,多普勒解卷绕准确率达99.87%,有效解决了高速场景下的速度歧义问题,为机器人全天候感知提供了高鲁棒性解决方案。
关键看点
- 检测精度突破:在BRIDGE-Liv数据集RAD-cube设置下mAP@0.30达80.8%,较现有最强原始雷达基线提升15.7个百分点,全维度领先。
- 速度估计性能:动目标速度MAE低至1.128 m/s,多普勒卷绕解算准确率达99.87%,有效消除高速移动场景下的径向速度歧义。
- 自运动补偿机制:通过显式估计帧级雷达平移速度,精准预测并抑制世界静态结构的多普勒响应,从源头解决自运动引起的虚警干扰。
- 原始数据处理:跳过稀疏点云或2D投影等中间表征,直接对原始RAD立方体进行时序多尺度特征编码,保留了被传统管线丢弃的相位与幅度信息。
工程与产业观察
该工作证明了在具身智能感知中回归原始传感器信号处理的工程价值。相比依赖后处理点云的方案,直接挖掘RAD立方体不仅提升了15%以上的检测上限,更以99.87%的解卷绕精度解决了毫米波雷达测速痛点。这对机器人在雨雾、暗光等非结构化环境下的安全导航具有关键参考意义,提示算法研发应重视底层信号完整性而非仅优化高层语义网络。
英文摘要 (Abstract)
Millimeter-wave (mmWave) radar provides weather-robust sensing and direct Doppler measurements, yet automotive perception pipelines often reduce its output to sparse point clouds or two-dimensional projections, leaving raw range-azimuth-Doppler (RAD) cubes underutilized. Ego motion induces Doppler responses from world-static structures, while Doppler wrapping aliases radial velocities, complicating moving-object detection and velocity estimation. We present RaMOD (Radar-based Moving-Object Detection), a real-time framework for moving-object detection and velocity estimation from raw RAD cubes acquired by a vehicle-mounted single-chip radar. RaMOD explicitly estimates frame-wise ego motion and uses the estimated radar translational velocity to predict the Doppler region of world-static structures and suppress returns within it. It temporally encodes the suppressed cubes into multi-scale features to predict RAD-space boxes with class labels. For each detected object, it predicts radial and forward velocities under raw-Doppler and Doppler-unwrapped hypotheses, together with Doppler-wrapping probabilities. On our BRIDGE-Liv dataset, RaMOD outperforms all compared raw-radar methods across RAD-cube, RA, and RD detection settings, achieving 80.8% cube mAP at IoU 0.30, 15.7 percentage points above the strongest RAD-cube baseline. Velocity MAE is 1.128 m/s, with 99.87% Doppler-wrapping accuracy.
作者:Yuan Fang, Long Zhuang, Lingzi Xing, Xijia Wei, Jingqi Jiang, Fangzhan Shi, Simon Julier, Kevin Chetty
外部来源
本页事实对应的公开页面,以原文为准。(点击将离开本站)