SeDriCa campus drives
ROS 2 Humble recordings (MCAP) from the SeDriCa autonomous buggy at IIT Bombay, October 2026, for tuning the local/global EKF, OSM matching and lidar localization: front Ouster OS1-32, Xsens MTi-320, WTRTK-982 GNSS, CAN wheel speed, both EKFs, LIO-SAM odometry and maps. One folder per drive, each with the bag, the LIO-SAM map recorded with it (when there is one), the configs in use and notes. Later data: OneLegend/sedrica-controls-2026-11-01-part1/-part2 (controls debugging, with camera) and OneLegend/sedrica-mapping-2026-11-03 (mapping drives, with camera).
7 folder(s), 36.6 GB in total. Each folder has its own README.md with the recording notes, the sensors and every topic with its message count and rate.
| Folder | Start (IST) | Duration | Size | Data | Notes |
|---|---|---|---|---|---|
campus_2026-10-23_00-10 |
2026-10-23 00:10 | 4 min 35 s | 2.3 GB | Lidar point clouds, GNSS, IMU, Odometry / localization, Vehicle / DBW / CAN | Short drive (4.5 min) the night of 2026-10-23, before campus_2026-10-23_00-26: front Ouster, Xsens, GNSS, CAN wheel speed, EKFs, LIO-SAM. |
campus_2026-10-23_00-16 |
2026-10-23 00:16 | 6 min 34 s | 4.7 GB | Lidar point clouds, GNSS, IMU, Odometry / localization, Vehicle / DBW / CAN | Short drive (6.5 min) the night of 2026-10-23, before campus_2026-10-23_00-26: front Ouster, Xsens, GNSS, CAN wheel speed, EKFs, LIO-SAM. |
campus_2026-10-23_00-26 |
2026-10-23 00:26 | 16 min 10 s | 12.4 GB | Lidar point clouds, GNSS, IMU, Odometry / localization, Vehicle / DBW / CAN | 16 min 10 s (00:26:02 to 00:42:12 IST), 11.3 GiB MCAP, ROS 2 Humble. Recorded on the Jetson with |
sedrica-20261023T205524 |
2026-10-23 20:55 | 24 min 09 s | 454.2 MB | GNSS, IMU, Odometry / localization, Vehicle / DBW / CAN | GNSS/IMU/EKF-only recording (rosbag2 sqlite3, no lidar), 24 min from 20:55 on 2026-10-23, just before campus_2026-10-23_21-07. |
campus_2026-10-23_21-07 |
2026-10-23 21:07 | 9 min 58 s | 8.5 GB | Lidar point clouds, GNSS, IMU, Odometry / localization, Vehicle / DBW / CAN | 9 min 58 s (21:07:15 to 21:17:14 IST), 7.6 GiB MCAP, ROS 2 Humble, 1.34 km driven. First drive with the |
campus_2026-10-30_01-45 |
2026-10-30 01:45 | 16 min 45 s | 8.2 GB | Lidar point clouds, GNSS, IMU, Odometry / localization, Vehicle / DBW / CAN, Planning / control | 16 min 45 s (01:45:21 to 02:02:05 IST), 7.7 GiB MCAP, ROS 2 Humble, 2436050 messages. First real-vehicle |
sedrica_gnss_runs |
441.3 KB | 6 file(s) | GNSS receiver test runs (2026-10-22/23): logged fixes as CSV with a summary per run. |
Download
One folder (bags are large, so this is usually what you want):
hf download OneLegend/sedrica-campus-drives --repo-type dataset --include "<folder>/*" --local-dir .
Only the READMEs and bag metadata, to look around first:
hf download OneLegend/sedrica-campus-drives --repo-type dataset --include "*.md" --include "*/metadata.yaml" --local-dir .
Format
- Bags are ROS 2 (Humble) rosbag2 folders, mostly MCAP. Play with
ros2 bag play <folder>/bag --clockor open the.mcapfiles in Foxglove. - Point cloud maps are
.pcdfiles (LIO-SAM output:GlobalMap.pcd,CornerMap.pcd,SurfMap.pcd,trajectory.pcd), in the map frame of the drive they came from. - Vehicle: the SeDriCa autonomous buggy at IIT Bombay. Sensors and frames are described in the SeDriCa repository docs.
Uploaded with upload_data.sh from the SeDriCa repository.
- Downloads last month
- 356