Boreas Autonomous Driving Dataset

Description

This autonomous driving dataset includes data from a 128-beam Velodyne Alpha-Prime lidar, a 5MP Blackfly camera, a 360-degree Navtech radar, and post-processed Applanix POS LV GNSS data. This dataset was collect in various weather conditions (sun, rain, snow) over the course of a year. The intended purpose of this dataset is to enable benchmarking of long-term all-weather odometry and metric localization across various sensor types. In the future, we hope to also support an object detection benchmark.

Update Frequency

New driving sequences will be added as they are collected.

License

CC BY 4.0

Documentation

https://github.com/utiasASRL/pyboreas/blob/master/DATA_REFERENCE.md

Managed By

See all datasets managed by ASRL.

Contact

boreas@robotics.utias.utoronto.ca

How to Cite

Boreas Autonomous Driving Dataset was accessed on DATE from https://registry.opendata.aws/boreas.

Usage Examples

Tutorials

• Introduction to Visualizing Sensor Types (Jupyter notebook) by Keenan Burnett
  SageMaker
• Project Lidar onto Camera Frames (Jupyter notebook) by Keenan Burnett
  SageMaker

Publications

• Are We Ready for Radar to Replace Lidar in All-Weather Mapping and Localization? by K Burnett, Y Wu, D J Yoon, A P Schoellig, T D Barfoot
  
• Boreas: A multi-season autonomous driving dataset by K Burnett, D J Yoon, Y Wu, A Z Li, H Zhang, S Lu, J Qian, W Tseng, A Lambert, K YK Leung, A P Schoellig, Timothy D Barfoot
  
• Do we need to compensate for motion distortion and doppler effects in spinning radar navigation? by K Burnett, A P Schoellig, T D Barfoot
  
• Need for Speed: Fast Correspondence-Free Lidar Odometry Using Doppler Velocity by D J Yoon, K Burnett, J Laconte, Y Chen, H Vhavle, S Kammel, J Reuther, T D Barfoot
  
• Picking up speed: Continuous-time Lidar-only odometry using doppler velocity measurements by Y Wu, D J Yoon, K Burnett, S Kammel, Y Chen, H Vhavle, T D Barfoot
  
• Radar odometry combining probabilistic estimation and unsupervised feature learning by K Burnett, D J Yoon, A P Schoellig, T D Barfoot