RaLL: End-to-End Radar Localization on Lidar Map Using Differentiable Measurement Model
RaLL: End-to-End Radar Localization on Lidar Map Using Differentiable Measurement Model
复制标题
RaLL：使用可微测量模型在激光雷达地图上进行端到端雷达定位
DOI：
10.1109/tits.2021.3061165
复制
发表时间：
2020-09
影响因子：
8.5
通讯作者：
Rong Xiong
中科院分区：
文献类型：
--
作者：
Huan Yin;Runjian Chen;Yue Wang;Rong Xiong
与机载摄像头和激光扫描仪相比，雷达传感器提供照明和天气不变传感，自然适合在恶劣条件下进行长期定位。然而，雷达数据的稀疏性和噪声性给雷达制图带来了挑战。另一方面，目前最流行的地图是由激光雷达绘制的。本文提出了一种基于激光雷达地图的雷达定位（Radar Localization on Lidar Map, RaLL）的端到端深度学习框架来弥补这一空白，既实现了鲁棒雷达定位，又利用了成熟的激光雷达制图技术，从而降低了雷达制图成本。我们首先通过神经网络将两种传感器模态嵌入到公共特征空间中。然后在地图模态中加入多个偏移量，对当前雷达模态进行穷举相似度评估，得到当前姿态的回归。最后，我们将该可微测量模型应用于卡尔曼滤波器（KF），以端到端方式学习整个序列定位过程。整个学习系统是可微的，前端是基于网络的测量模型，后端是KF。为了验证可行性和有效性，我们使用了从现实世界中收集的多会话多场景数据集，结果表明，我们提出的系统在$90km$行驶中取得了卓越的性能，即使在模型训练在英国进行的泛化场景中，而在韩国进行测试。我们还公开发布源代码。
Compared to the onboard camera and laser scanner, radar sensor provides lighting and weather invariant sensing, which is naturally suitable for long-term localization under adverse conditions. However, radar data is sparse and noisy, resulting in challenges for radar mapping. On the other hand, the most popular available map currently is built by lidar. In this paper, we propose an end-to-end deep learning framework for Radar Localization on Lidar Map (RaLL) to bridge the gap, which not only achieves the robust radar localization but also exploits the mature lidar mapping technique, thus reducing the cost of radar mapping. We first embed both sensor modals into a common feature space by a neural network. Then multiple offsets are added to the map modal for exhaustive similarity evaluation against the current radar modal, yielding the regression of the current pose. Finally, we apply this differentiable measurement model to a Kalman Filter (KF) to learn the whole sequential localization process in an end-to-end manner. The whole learning system is differentiable with the network based measurement model at the front-end and KF at the back-end. To validate the feasibility and effectiveness, we employ multi-session multi-scene datasets collected from the real world, and the results demonstrate that our proposed system achieves superior performance over $90km$ driving, even in generalization scenarios where the model training is in UK, while testing in South Korea. We also release the source code publicly.
登录
查看更多内容
DOI：
--
发表时间：
2012
期刊：
Plant, Cell & Environment
影响因子：
--
作者：
Winson S. Churchill
通讯作者：
Winson S. Churchill
DOI：
--
发表时间：
2020-11
期刊：
ArXiv
影响因子：
--
作者：
Keenan Burnett;Angela P. Schoellig;T. Barfoot
通讯作者：
Keenan Burnett;Angela P. Schoellig;T. Barfoot
DOI：
10.1109/icra40945.2020.9197406
发表时间：
2020-05
期刊：
2020 IEEE International Conference on Robotics and Automation (ICRA)
影响因子：
--
作者：
Huaiyang Huang;Haoyang Ye;Yuxiang Sun;Ming Liu
通讯作者：
Huaiyang Huang;Haoyang Ye;Yuxiang Sun;Ming Liu
DOI：
10.1109/ccece47787.2020.9255714
发表时间：
2020-08
期刊：
2020 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE)
影响因子：
--
作者：
Thangarajah Akilan;E. Johnson;Gaurav Taluja;Japneet Sandhu;Ritika Chadha
通讯作者：
Thangarajah Akilan;E. Johnson;Gaurav Taluja;Japneet Sandhu;Ritika Chadha
DOI：
10.1136/bmj.323.7325.1375/a
发表时间：
2001-12
期刊：
BMJ : British Medical Journal
影响因子：
--
作者：
K. Barraclough
通讯作者：
K. Barraclough