Binocular stereo vision-based relative positioning algorithm for drone swarm
Abstract To address the challenges of high computational complexity and poor real-time performance in binocular vision-based Unmanned Aerial Vehicle (UAV) formation flight, this paper introduces a UAV localization algorithm based on a lightweight object detection model. Firstly, we optimized the YOL...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-86981-1 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571723892916224 |
|---|---|
| author | Qing Cheng Yazhe Wang |
| author_facet | Qing Cheng Yazhe Wang |
| author_sort | Qing Cheng |
| collection | DOAJ |
| description | Abstract To address the challenges of high computational complexity and poor real-time performance in binocular vision-based Unmanned Aerial Vehicle (UAV) formation flight, this paper introduces a UAV localization algorithm based on a lightweight object detection model. Firstly, we optimized the YOLOv5s model using lightweight design principles, resulting in Yolo-SGN. This model achieves a 65.5% reduction in parameter count, a 62.7% reduction in FLOPs, and a 1.8% increase in accuracy compared to the original detection model. Subsequently, Yolo-SGN is utilized to extract target regions from binocular images, and feature point matching is exclusively conducted within these regions to minimize unnecessary computations in non-target areas. Experimental results demonstrate that the combination of Yolo-SGN and the Oriented FAST and Rotated BRIEF (ORB) algorithm reduces feature point matching computations to only a quarter of those in the original ORB algorithm, significantly enhancing real-time performance. Finally, the extracted feature points from UAVs are input into a binocular vision localization model to compute their three-dimensional coordinates. The average of the three-dimensional coordinates of all feature points is used to determine the three-dimensional position of the target UAV. Experimental results confirm that the UAV binocular vision localization algorithm, based on a lightweight object detection model, exhibits exceptional performance in terms of precision and real-time capabilities. |
| format | Article |
| id | doaj-art-59ba89eadccb43fca305f372917bb8f3 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-59ba89eadccb43fca305f372917bb8f32025-02-02T12:20:37ZengNature PortfolioScientific Reports2045-23222025-01-0115112110.1038/s41598-025-86981-1Binocular stereo vision-based relative positioning algorithm for drone swarmQing Cheng0Yazhe Wang1School of Air Traffic Management, Civil Aviation Flight University of ChinaSchool of Air Traffic Management, Civil Aviation Flight University of ChinaAbstract To address the challenges of high computational complexity and poor real-time performance in binocular vision-based Unmanned Aerial Vehicle (UAV) formation flight, this paper introduces a UAV localization algorithm based on a lightweight object detection model. Firstly, we optimized the YOLOv5s model using lightweight design principles, resulting in Yolo-SGN. This model achieves a 65.5% reduction in parameter count, a 62.7% reduction in FLOPs, and a 1.8% increase in accuracy compared to the original detection model. Subsequently, Yolo-SGN is utilized to extract target regions from binocular images, and feature point matching is exclusively conducted within these regions to minimize unnecessary computations in non-target areas. Experimental results demonstrate that the combination of Yolo-SGN and the Oriented FAST and Rotated BRIEF (ORB) algorithm reduces feature point matching computations to only a quarter of those in the original ORB algorithm, significantly enhancing real-time performance. Finally, the extracted feature points from UAVs are input into a binocular vision localization model to compute their three-dimensional coordinates. The average of the three-dimensional coordinates of all feature points is used to determine the three-dimensional position of the target UAV. Experimental results confirm that the UAV binocular vision localization algorithm, based on a lightweight object detection model, exhibits exceptional performance in terms of precision and real-time capabilities.https://doi.org/10.1038/s41598-025-86981-1Unmanned aerial vehicle detectionLightweight networkBinocular stereo visionDeep learning |
| spellingShingle | Qing Cheng Yazhe Wang Binocular stereo vision-based relative positioning algorithm for drone swarm Scientific Reports Unmanned aerial vehicle detection Lightweight network Binocular stereo vision Deep learning |
| title | Binocular stereo vision-based relative positioning algorithm for drone swarm |
| title_full | Binocular stereo vision-based relative positioning algorithm for drone swarm |
| title_fullStr | Binocular stereo vision-based relative positioning algorithm for drone swarm |
| title_full_unstemmed | Binocular stereo vision-based relative positioning algorithm for drone swarm |
| title_short | Binocular stereo vision-based relative positioning algorithm for drone swarm |
| title_sort | binocular stereo vision based relative positioning algorithm for drone swarm |
| topic | Unmanned aerial vehicle detection Lightweight network Binocular stereo vision Deep learning |
| url | https://doi.org/10.1038/s41598-025-86981-1 |
| work_keys_str_mv | AT qingcheng binocularstereovisionbasedrelativepositioningalgorithmfordroneswarm AT yazhewang binocularstereovisionbasedrelativepositioningalgorithmfordroneswarm |