GitPedia

Lidar degeneracy datasets

LiDAR degeneracy dataset for LiDAR-radar-inertial fusion methods.

From ntnu-arl·Updated June 10, 2026·View on GitHub·

Dataset associated with submission for handling LiDAR-degenerative environments through LiDAR-radar-inertial fusion with factor graphs. The project is distributed under the BSD 3-Clause "New" or "Revised" License license, first published in 2023. Key topics include: imu, inertial-navigation-systems, lidar, lidar-degeneracy, odometry.

LiDAR Degeneracy Datasets

Dataset associated with submission for handling LiDAR-degenerative environments through LiDAR-radar-inertial fusion with factor graphs.

The datasets consist of sensor measurements (IMU, LiDAR, and FMCW radar) of an aerial robot manually flown through environments challenging for LiDAR sensors. These are a biking tunnel (geometric self-similarity) and a university corridor filled with fog (obscurants). In addition CAD extrinsics between the sensors are provided.

Environments

Fyllingsdalen Tunnel

Manual flight along a 8m-wide, 500m-long, straight section of a biking tunnel in Bergen, Norway.

fyllingsdalen bycicale tunnel image

Bag files can be found here

Fog-Filled University Corridor

Manual flight in a university environment in the Elektro building of NTNU in Trondheim, Norway. The first corridor flown through after takeoff is filled with dense fog.

fog-filled university corridor image

Bag files can be found here

Robot Equipiment

Drone

Sensors

Synchronized using microcontroller-based internally developed synchronization/triggering module

The header time stamp for IMU and radar originate from the triggering module, to replay the trigger-stamped LiDAR data from packet topics use this driver.

Topics
SourceTopicRate [Hz]
Triggering- /sensor_sync_node/trigger_0<br>- /sensor_sync_node/trigger_1- 800<br>- 10
IMU/vectornav_node/uncomp_imu200
LiDAR- /os_cloud_node/imu_packets<br>- /os_cloud_node/lidar_packets<br>- /os_cloud_node/metadata10
Radar/radar/cloud10
Extrinsics

All extrinsics are given with respect to the IMU

LiDAR

  • translation [x, y, z]
    • [-0.00171, 0.02149, 0.0358]
  • orientation [x, y, z, w]
    • [0.000462, 0.0008483, 0.0028835, 0.9999954]

Radar

  • translation [x, y, z]
    • [0.07771, 0.02141, -0.03631]
  • orientation [x, y, z, w]
    • [0.953717, 0, -0.3007058, 0]

Acknowledgements

We thank the Vestland Fylkeskommune for providing access to the Fyllingsdal sykkeltunnel.

Reference

If you use this work in your research, please cite the following publication:

Degradation Resilient LiDAR-Radar-Inertial Odometry (link)

bibtex
@inproceedings{nissov2024degradation, author={Nissov, Morten and Khedekar, Nikhil and Alexis, Kostas}, booktitle={2024 IEEE International Conference on Robotics and Automation (ICRA)}, title={Degradation Resilient LiDAR-Radar-Inertial Odometry}, year={2024}, pages={8587-8594}, doi={10.1109/ICRA57147.2024.10611444} }

Contributors

Showing top 2 contributors by commit count.

View all contributors on GitHub →

This article is auto-generated from ntnu-arl/lidar_degeneracy_datasets via the GitHub API.Last fetched: 6/19/2026