USTC FLICAR: A Sensors Fusion Dataset of Lidar-Inertial-Camera for Heavy-duty Autonomous Aerial Work Robots (IJRR 2023)

Website: https://ustc-flicar.github.io/

📢 Updates

🚀 23/01/2025: Support the SOTA LIVO System: Fast-LIVO2!

🚀 01/09/2024: The author came to the HKU MaRS Lab for postgraduate study under supervised by Prof. Fu Zhang and continued his journey in robotics and SLAM

Introduction

This site presents the USTC FLICAR Dataset, collected from our research on heavy-duty autonomous aerial work robots, featuring a comprehensive set of sensors:

• Four 3D LiDARs (Velodyne HDL32/VLP32; LiVOX Avia; Ouster OS0-128)
• Two stereo cameras (Bumblebee XB3/XB2)
• Two monocular cameras (Hikvison; FILR IR)
• Multiple Inertial Measurement Units (IMUs)
• GNSS/INS system (Xsens MTI-G-710)
• Laser tracker for millimeter-level ground truth positions (API T3 Laser Tracker)

The dataset extends the typical autonomous driving sensing suite to aerial scenes, utilizing the “Giraffe” mapping robot based on a bucket truck. This platform is designed to explore the potential of combining autonomous driving perception systems with aerial work robots. Additionally, we introduce the Semantic FLICAR dataset, which provides fine-grained semantic segmentation annotations for multimodal continuous data in both temporal and spatial dimensions.

Citation

If you use some resource from this data suite, please cite it as

@article{wang2023ustc,
  title={USTC FLICAR: A sensors fusion dataset of LiDAR-inertial-camera for heavy-duty autonomous aerial work robots},
  author={Wang, Ziming and Liu, Yujiang and Duan, Yifan and Li, Xingchen and Zhang, Xinran and Ji, Jianmin and Dong, Erbao and Zhang, Yanyong},
  journal={The International Journal of Robotics Research},
  volume={42},
  number={11},
  pages={1015--1047},
  year={2023},
  publisher={SAGE Publications Sage UK: London, England}
}

[Journal][Preprint]

Downloads

Name	Link	Size	Duration	Remark
hf001	.bag	66.5 GB	192.5 s 26.46 m	Collected in complex aerial work scenes with power lines, trees, and houses with bucket truck motion; Sun
hf002	.bag	75.7 GB	217.8 s 33.5 m	The same scenes as above
hf003	.bag	83.2 GB	217.1 s 34.26 m	The same scenes as above
hf004	.bag	82.0 GB	155.9 s 24.1 m	The same scenes as above
hf005	.bag	90.3 GB	260.4 s 24.1 m	The same scenes as above
hf006	.bag	86.3 GB	230.6 s 33.9 m	The same scenes as above; cloud
hf007	.bag	67.5 GB	207.6 s 34.32 m	The same scenes as above; dusk
hf008	.bag	91.3 GB	210.6 s 30.78 m	The same scenes as above; night
hf009	.bag	101.3 GB	238.7 s 35.42 m	The same scenes as above; night
hf010	.bag	91.7 GB	210 s 16.06 m	The same scenes as above
hf011	.bag	25.5 GB	207 s 17.81 m	The same scenes as above; dusk
hf012	.bag	121.1 GB	231 s 26.15 m	The same scenes as above; dusk
hf013	.bag	100.9 GB	187 s 26.25 m	The same scenes as above; night
hf014	.bag	119.2 GB	201 s 25.57 m	The same scenes as above; night
hn001	.bag	79.2 GB	390 s 38.44 m	Collected in the 2th aerial work scene, including trucks, buildings, trees, etc.; IR
hn002	.bag	56.1 GB	395 s 44.97 m	Collected in the 3th aerial work scene, including trucks, buildings, etc.; IR
hn003	.bag	62.2 GB	442 s 38.64 m	The same scenes as above
hn004	.bag	59.1 GB	417 s 42.50 m	The same scenes as above
calib_data	.bag	-	-	IMU; Momo Stereo Camera; IMU-Camera; IMU-LiDAR, Camera-LiDAR; Multi-LiDARs Calib

Quick use

We have done some experiments of state-of-the-art methods on our the datasets. If you are seeking to do the same, please check out the following to get the work done quickly.

</style>

Method	Repository	Credit
Fast-LIO	https://github.com/ustc-flicar/ustcflicar-FAST-LIO	Forked from https://github.com/hku-mars/FAST_LIO
VINS-Fusion	https://github.com/ustc-flicar/ustcflicar-VINS-Fusion	Forked from https://github.com/HKUST-Aerial-Robotics/VINS-Fusion
VINS-Mono	https://github.com/brytsknguyen/VINS-Mono	Forked from https://github.com/HKUST-Aerial-Robotics/VINS-Mono
LIO-SAM	https://github.com/ustc-flicar/ustcflicar-VINS-Fusion	Forked from https://github.com/TixiaoShan/LIO-SAM
A-LOAM	https://github.com/ustc-flicar/ustcflicar-A-LOAM	Forked from https://github.com/HKUST-Aerial-Robotics/A-LOAM

Licence

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License and is intended for non-commercial academic use. If you are interested in using the dataset for commercial purposes please contact us.