MMAUD: A Comprehensive Multi-Modal Anti-UAV Dataset for Detection, Classification, Tracking and Trajectory Estimation of Compact Commercially Available Drones Threats
This site presents the datasets collected from our research platform, featuring an extensive set of sensors:
- Two 3D lidars ( Conic LIDAR and Peripheral LIDAR)
- Two time-synchronized cameras
- One mmWave Radar
- Four Audio Array Nodes
If you use some resource from this data suite, please cite it as
@INPROCEEDINGS{yuan2024MMAUD,
author={Yuan, Shenghai and Yang, Yizhuo and Nguyen, Thien Hoang and Nguyen, Thien-Minh and Yang, Jianfei and Liu, Fen and Li, Jianping and Wang, Han and Xie, Lihua},
booktitle={2024 IEEE International Conference on Robotics and Automation (ICRA)},
title={MMAUD: A Comprehensive Multi-Modal Anti-UAV Dataset for Modern Miniature Drone Threats},
year={2024},
pages={2745-2751},
doi={10.1109/ICRA57147.2024.10610957}
}
The files below are hosted on OneDrive. If you are having a problem downloading from one drive, do raise an issue.
Note: All rosbag data has been compressed using 'rosbag compress' to reduce its size by a factor of 3. If you directly run 'rosbag play,' the playback frequency will be reduced. To restore the bag to its full rate, please use the 'rosbag decompress' command.
<style type="text/css"> .tg {border-collapse:collapse;border-spacing:0;} .tg td{border-color:black;border-style:solid;border-width:1px;font-family:Arial, sans-serif;font-size:14px; overflow:hidden;padding:10px 5px;word-break:normal;} .tg th{border-color:black;border-style:solid;border-width:1px;font-family:Arial, sans-serif;font-size:14px; font-weight:normal;overflow:hidden;padding:10px 5px;word-break:normal;} .tg .tg-6ibf{border-color:inherit;font-size:18px;text-align:center;vertical-align:top} .tg .tg-v8dz{border-color:inherit;font-size:18px;text-align:left;vertical-align:top} .tg .tg-9m02{border-color:inherit;color:#00E;font-size:18px;text-align:center;text-decoration:underline;vertical-align:top} </style>| Name | ROSBag Data | Folder Data | Ground truth | Size | Duration | Remark |
|---|---|---|---|---|---|---|
| DJI Mavic2 | .bag | .zip | .bag | 14.1 GB | 198s | MMAUD V1 Rooftop Simple |
| DJI Mavic3 | .bag | .zip | .bag | 11.1 GB | 321.1 s | MMAUD V1 Rooftop Simple |
| DJI Phantom4 | .bag | .zip | .bag | 13.2 GB | 181.4 s | MMAUD V1 Rooftop Simple |
| DJI Avata | .bag | .zip | .bag | 19.7 GB | 396.3 s | MMAUD V1 Rooftop Simple |
| DJI M300 | .bag | .zip | .bag | 14.4 GB | 428.7 s | MMAUD V1 Rooftop Simple |
| DJI Mavic3 | .bag | .zip | .bag | ?? GB | ?? s | MMAUD V2 Carpark Hard |
| DJI Phantom4 | .bag | .zip | .bag | ?? GB | ?? s | MMAUD V2 Carpark Hard |
| DJI Avata | .bag | .zip | .bag | ?? GB | ?? s | MMAUD V2 Carpark Hard |
| DJI M300 | .bag | .zip | .bag | ?? GB | ?? s | MMAUD V2 Carpark Hard |
| DJI Mavic3 | .bag | .zip | .bag | ?? GB | ?? s | MMAUD V3 Carpark Moderate |
| DJI Phantom4 | .bag | .zip | .bag | ?? GB | ?? s | MMAUD V3 Carpark Moderate |
| DJI Avata | .bag | .zip | .bag | ?? GB | ?? s | MMAUD V3 Carpark Moderate |
| DJI M300 | .bag | .zip | .bag | ?? GB | ?? s | MMAUD V3 Carpark Moderate |
V2 and V3 were used for the CVPR UG2+ challenge, which is more challenging than V1. V1 mostly flies below 30m, while V2 and V3 are designed for actual warfare simulation, reaching up to 70m. UAV Precision hits are unlikely to go beyond 70m. The results of the UG2 CVPR 2024 challenge are now available.
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.
Remark: The 2D detection baseline is considered too trivial to be provided separately. Interested party can download the data here https://drive.google.com/drive/folders/1_LpPyIfETQS-k2vlSsbzI9pzyVzZScSx?usp=sharing. Then follow the official code of yolov5 to train the model: https://github.com/ultralytics/yolov5. For rest of other 2D detection pipelines, it should be similar.
| 3D Pose Estimation | Repository | Remark |
|---|---|---|
| ResNet | ResNet | Visual-Only/Supervised |
| Darknet | Darknet | Visual-Only/Supervised |
| VorasNet | DroneChase | Audio-Only/Supervised |
| AV-PED | AV-PED | Audio-Visual/Self-Supervised |
| Audionet | Audionet | Audio-Only/Supervised |
| AV-FDTI | AV-FDTI | Audio-Visual/Supervised |
| TAME | TAME | Audio-Only/Self-Supervised |
| AAUTE | AAUTE | Audio-Only/Self-Supervised |
| UnLiDAR | UnLiDAR | LiDAR-Only/Unsupervised |
| MMUDCT | MMUDCT | LiDAR-Visual/Unsupervised |
| AV-DTEC | AV-DTEC | LiDAR-Visual-Audio/Mamba/Self-Supervised |
The CAD drawing can be found here.
Since there are multiple ethernet devices. It is recommended to set 2 livox lidar and MMwave radar to be at 192.168.10.xx , 192.168.11.xx , and 192.168.12.xx.
The microphone and camera can be obtained from Taobao. Whereas other LIDAR and RADAR need to find your local distributor to get it.
If you have any issues in recreating this rig, feel free to drop an issue in this dataset repo
For more information on the sensors and how to use the dataset, please checkout the other sections.
For resources and other works of our group please checkout our github.
If you have some inquiry, please raise an issue on github.
Each part of the code is obtained separately from a different source. Please help cite their corresponding works accordingly.
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.