【强基固本】自动驾驶场景中的毫米波雷达数据集

“强基固本，行稳致远”，科学研究离不开理论基础，人工智能学科更是需要数学、物理、神经科学等基础学科提供有力支撑，为了紧扣时代脉搏，我们推出“强基固本”专栏，讲解AI领域的基础知识，为你的科研学习提供助力，夯实理论基础，提升原始创新能力，敬请关注。

来源：知乎—罗驳思

地址：https://zhuanlan.zhihu.com/p/368040719

01

数据集简介

1. Oxford Radar RobotCar 数据集

项目地址：

https://oxford-robotics-institute.github.io/radar-robotcar-dataset/

论文：The Oxford Radar RobotCar Dataset: A Radar Extension to the Oxford RobotCar Dataset

https://arxiv.org/pdf/1909.01300.pdf

处理工具：

https://github.com/oxford-robotics-institute/radar-robotcar-dataset-sdk

传感器配置：

传感器布局：

数据规模：

2. ColoRadar 数据集

项目地址：

https://arpg.github.io/coloradar/

论文：ColoRadar: The Direct 3D Millimeter Wave Radar Dataset

https://arxiv.org/pdf/2103.04510.pdf

传感器配置：

• Cascaded Imaging Radar Sensor: Texas Instruments MMWCAS-RF-EVM
• Single Chip Radar Sensor: Texas Instruments AWR1843BOOST-EVM paired with a DCA1000-EVM for raw data capture
• Lidar Sensor: Ouster OS1; 10Hz; 64 beams; 1-degree angular accuracy; angular resolution 0.35-degree horizontal, 0.7-degree vertical; 3cm range accuracy; field of view 360-degree horizontal, 45-degree vertical; max range 120m; 65,536 points per scan
• IMU: Lord Microstrain 3DM-GX5-25; 300Hz

传感器布局：

数据规模：

3. RADIATE 数据集

项目地址：

http://pro.hw.ac.uk/radiate/

论文：RADIATE: A Radar Dataset for Automotive Perception in Bad Weather

https://arxiv.org/pdf/2010.09076.pdf

处理工具：

https://github.com/marcelsheeny/radiate_sdk

传感器配置：

传感器布局：

数据规模：

4. MulRan 数据集

项目地址：

https://sites.google.com/view/mulran-pr

论文：MulRan: Multimodal Range Dataset for Urban Place Recognition

https://irap.kaist.ac.kr/publications/gskim-2020-icra.pdf

处理工具：

https://github.com/irapkaist/file_player_mulran

传感器配置：

传感器布局：

数据规模：

5. CRUW 数据集

项目地址：

https://www.cruwdataset.org/

论文：RODNet: A Real-Time Radar Object Detection Network Cross-Supervised by Camera-Radar Fused Object 3D Localization

https://arxiv.org/pdf/2102.05150.pdf

处理工具：

https://github.com/yizhou-wang/cruw-devkit

传感器配置：

传感器布局：

数据规模：

6. CARRADA 数据集

项目地址：

https://arthurouaknine.github.io/codeanddata/carrada

论文：CARRADA Dataset: Camera and Automotive Radar with Range-Angle-Doppler Annotations

https://arxiv.org/pdf/2005.01456.pdf

处理工具：

https://github.com/valeoai/carrada_dataset

传感器配置：

数据规模：

7. RaDICaL 数据集

项目地址：

https://publish.illinois.edu/radicaldata/

论文：RaDICaL: A Synchronized FMCW Radar, Depth, IMU and RGB Camera Data Dataset with Low-Level FMCW Radar Signals

https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9361086

处理工具：

https://github.com/moodoki/radical_sdk

传感器配置：

• Radar: Texas Instuments IWR1443BOOST
• RGB-D Camera: Intel RealSense D435i
• Real-Time Data Capture System: Texas Instruments DCA1000EVM

传感器布局：

数据规模：

8. nuScenes数据集

项目地址：

https://www.nuscenes.org/

论文：nuScenes: A multimodal dataset for autonomous driving

https://arxiv.org/pdf/1903.11027.pdf

处理工具：

https://github.com/nutonomy/nuscenes-devkit

传感器配置：

传感器布局：

数据规模：

9. RadarScenes数据集

项目地址：

https://radar-scenes.com/

论文：RadarScenes: A Real-World Radar Point Cloud Data Set for Automotive Applications

https://arxiv.org/pdf/2104.02493.pdf

处理工具：

https://github.com/oleschum/radar_scenes

传感器配置：4个77GHz的自动驾驶雷达传感器，包含了两种模式：近距离模式（near-range）的检测范围是100米，远距离（far-range）模式的视角很小，一般不考虑用作自动驾驶。近距离模式下的FOV是±60度。此外还配置了一台纪实摄影机 (documentary camera)。

传感器布局：

数据集规模：

10. Zendar数据集

项目地址：

https://www.zendar.io/dataset.html

论文：High Resolution Radar Dataset for Semi-Supervised Learning of Dynamic Objects

https://openaccess.thecvf.com/content_CVPRW_2020/papers/w6/Mostajabi_High-Resolution_Radar_Dataset_for_Semi-Supervised_Learning_of_Dynamic_Objects_CVPRW_2020_paper.pdf

处理工具：

https://githubmemory.com/repo/ZendarInc/ZendarSDK

传感器配置：

• High resolution SAR images
• Range doppler processed point cloud
• Raw radar data cube
• Lidar stream
• Camera stream
• Tracklog

数据集规模：

在 27 个不同场景中标记超过 11,000 辆移动汽车，自动生成超过 40,000 个移动汽车标签

11. UMA-SAR数据集

项目地址：

https://www.uma.es/robotics-and-mechatronics/info/124594/sar-datasets/?set_language=en

论文：The UMA-SAR Dataset: Multimodal Data Collection from a Ground Vehicle during Outdoor Disaster Response Training Exercises

https://www.uma.es/media/files/The_UMA_SAR_Dataset__Multimodal_Data_Collection_from_a_Ground_Vehicle_during_Outdoor_Disaster_Response_Training_Exercises_ACCEPTED_VERSIONpdf.pdf

处理工具：

https://github.com/Robotics-Mechatronics-UMA

传感器配置：

传感器布局：

数据集规模：

02

参考文献

1. Barnes, D., Gadd, M., Murcutt, P., Newman, P., & Posner, I. (2020, May). The oxford radar robotcar dataset: A radar extension to the oxford robotcar dataset. In2020 IEEE International Conference on Robotics and Automation (ICRA)(pp. 6433-6438). IEEE.

2. Kramer, A., Harlow, K., Williams, C., & Heckman, C. (2021). ColoRadar: The Direct 3D Millimeter Wave Radar Dataset.arXiv preprint arXiv:2103.04510.

3. Sheeny, M., De Pellegrin, E., Mukherjee, S., Ahrabian, A., Wang, S., & Wallace, A. (2020). Radiate: A radar dataset for automotive perception.arXiv preprint arXiv:2010.09076,3(4), 7.

4. Kim, G., Park, Y. S., Cho, Y., Jeong, J., & Kim, A. (2020, May). Mulran: Multimodal range dataset for urban place recognition. In2020 IEEE International Conference on Robotics and Automation (ICRA)(pp. 6246-6253). IEEE.

5. Wang, Y., Jiang, Z., Li, Y., Hwang, J. N., Xing, G., & Liu, H. (2021). RODNet: A Real-Time Radar Object Detection Network Cross-Supervised by Camera-Radar Fused Object 3D Localization.IEEE Journal of Selected Topics in Signal Processing.

6. Ouaknine, A., Newson, A., Rebut, J., Tupin, F., & Pérez, P. (2021, January). CARRADA dataset: camera and automotive radar with range-angle-Doppler annotations. In2020 25th International Conference on Pattern Recognition (ICPR)(pp. 5068-5075). IEEE.

7. Lim, T. Y., Markowitz, S., & Do, M. N. (2021). RaDICaL: A Synchronized FMCW Radar, Depth, IMU and RGB Camera Data Dataset with Low-Level FMCW Radar Signals.IEEE Journal of Selected Topics in Signal Processing.

8. Caesar, H., Bankiti, V., Lang, A. H., Vora, S., Liong, V. E., Xu, Q., ... & Beijbom, O. (2020). nuscenes: A multimodal dataset for autonomous driving. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition(pp. 11621-11631).

9. Mostajabi, Mohammadreza, et al. "High-Resolution Radar Dataset for Semi-Supervised Learning of Dynamic Objects."Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. 2020.

10. Schumann, Ole, et al. "RadarScenes: A Real-World Radar Point Cloud Data Set for Automotive Applications."arXiv preprint arXiv:2104.02493(2021).

11. Morales, Jesús, et al. "The UMA-SAR Dataset: Multimodal data collection from a ground vehicle during outdoor disaster response training exercises."The International Journal of Robotics Research40.6-7 (2021): 835-847.

本文目的在于学术交流，并不代表本公众号赞同其观点或对其内容真实性负责，版权归原作者所有，如有侵权请告知删除。

“强基固本”历史文章

更多强基固本专栏文章，

请点击文章底部“阅读原文”查看

分享、点赞、在看，给个三连击呗！