外观
论文arXiv 2609.180921 个月前
ReRadar：通过旋转等变描述子学习实现鲁棒雷达全局定位
原文链接：ReRadar: Robust Radar Global Localization via Rotation-Equivariant Descriptor Learning（arXiv 2609.18092 · 发布日期：2026-09-16）
ReRadar: Robust Radar Global Localization via Rotation-Equivariant Descriptor Learning
ReRadar针对扫描毫米波雷达全局定位，学习旋转等变特征与不变描述子，并在多个数据库-查询评测上验证。
核心要点
- 动机：现有地点识别描述子常丢弃空间结构，影响精确位姿检索。
- 方法：用可转向CNN提取旋转等变中间特征，经群池化和NetVLAD形成不变描述子。
- 定位：将描述子检索与基于地标的匹配结合，估计机器人3-DoF位姿。
- 结果：目标数据集适配在OORD、Mulran和降雪Boreas上取得高Recall@1和F1。
验证范围与局限：验证限于固定数据库-查询评测，覆盖OORD Bellmouth、Mulran DCC01和降雪Boreas序列。
英文摘要 (Abstract)
Global localization with scanning millimeter-wave radar remains challenging because place-recognition descriptors often discard spatial structure needed for accurate pose retrieval. We present ReRadar, a radar global localization pipeline that extracts rotation-equivariant intermediate features using steerable convolutional neural networks, forms rotation-invariant descriptors through group pooling and NetVLAD aggregation, and combines descriptor retrieval with landmark-based matching to estimate the robot's three-degree-of-freedom (3-DoF) pose. Across fixed database-query evaluations, ReRadar with target-dataset adaptation achieves 99.37% Recall@1 on OORD Bellmouth, 91.44% Recall@1 with 80.99% F1_max on Mulran DCC01, and 99.38% Recall@1 on falling-snow Boreas sequence. Without target-dataset data, the cross-dataset model reaches 98.07% Recall@1 on OORD, performing comparably to the evaluated state-of-the-art methods.
作者：Duc Manh Nguyen, Truong Giang Dao, Gia Nghiem Luong, Hoang, Viet Trung, Anh Quang Nguyen
来源
本页事实对应的公开页面，以原文为准。