YDD-SLAM: Indoor Dynamic Visual SLAM Fusing YOLOv5 with Depth Information
Simultaneous location and mapping (SLAM) technology is key in robot autonomous navigation. Most visual SLAM (VSLAM) algorithms for dynamic environments cannot achieve sufficient positioning accuracy and real-time performance simultaneously. When the dynamic object proportion is too high, the VSLAM a...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-12-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/23/23/9592 |
| _version_ | 1797399590365822976 |
|---|---|
| author | Peichao Cong Junjie Liu Jiaxing Li Yixuan Xiao Xilai Chen Xinjie Feng Xin Zhang |
| author_facet | Peichao Cong Junjie Liu Jiaxing Li Yixuan Xiao Xilai Chen Xinjie Feng Xin Zhang |
| author_sort | Peichao Cong |
| collection | DOAJ |
| description | Simultaneous location and mapping (SLAM) technology is key in robot autonomous navigation. Most visual SLAM (VSLAM) algorithms for dynamic environments cannot achieve sufficient positioning accuracy and real-time performance simultaneously. When the dynamic object proportion is too high, the VSLAM algorithm will collapse. To solve the above problems, this paper proposes an indoor dynamic VSLAM algorithm called YDD-SLAM based on ORB-SLAM3, which introduces the YOLOv5 object detection algorithm and integrates deep information. Firstly, the objects detected by YOLOv5 are divided into eight subcategories according to their motion characteristics and depth values. Secondly, the depth ranges of the dynamic object and potentially dynamic object in the moving state in the scene are calculated. Simultaneously, the depth value of the feature point in the detection box is compared with that of the feature point in the detection box to determine whether the point is a dynamic feature point; if it is, the dynamic feature point is eliminated. Further, multiple feature point optimization strategies were developed for VSLAM in dynamic environments. A public data set and an actual dynamic scenario were used for testing. The accuracy of the proposed algorithm was significantly improved compared to that of ORB-SLAM3. This work provides a theoretical foundation for the practical application of a dynamic VSLAM algorithm. |
| first_indexed | 2024-03-09T01:42:18Z |
| format | Article |
| id | doaj.art-a81f37992b634ee8a8a0a1daa1fdeb20 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T01:42:18Z |
| publishDate | 2023-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-a81f37992b634ee8a8a0a1daa1fdeb202023-12-08T15:26:32ZengMDPI AGSensors1424-82202023-12-012323959210.3390/s23239592YDD-SLAM: Indoor Dynamic Visual SLAM Fusing YOLOv5 with Depth InformationPeichao Cong0Junjie Liu1Jiaxing Li2Yixuan Xiao3Xilai Chen4Xinjie Feng5Xin Zhang6School of Mechanical and Automotive Engineering, Guangxi University of Science and Technology, Liuzhou 545006, ChinaSchool of Mechanical and Automotive Engineering, Guangxi University of Science and Technology, Liuzhou 545006, ChinaSchool of Mechanical and Automotive Engineering, Guangxi University of Science and Technology, Liuzhou 545006, ChinaSchool of Mechanical and Automotive Engineering, Guangxi University of Science and Technology, Liuzhou 545006, ChinaSchool of Mechanical and Automotive Engineering, Guangxi University of Science and Technology, Liuzhou 545006, ChinaSchool of Mechanical and Automotive Engineering, Guangxi University of Science and Technology, Liuzhou 545006, ChinaSchool of Mechanical and Automotive Engineering, Guangxi University of Science and Technology, Liuzhou 545006, ChinaSimultaneous location and mapping (SLAM) technology is key in robot autonomous navigation. Most visual SLAM (VSLAM) algorithms for dynamic environments cannot achieve sufficient positioning accuracy and real-time performance simultaneously. When the dynamic object proportion is too high, the VSLAM algorithm will collapse. To solve the above problems, this paper proposes an indoor dynamic VSLAM algorithm called YDD-SLAM based on ORB-SLAM3, which introduces the YOLOv5 object detection algorithm and integrates deep information. Firstly, the objects detected by YOLOv5 are divided into eight subcategories according to their motion characteristics and depth values. Secondly, the depth ranges of the dynamic object and potentially dynamic object in the moving state in the scene are calculated. Simultaneously, the depth value of the feature point in the detection box is compared with that of the feature point in the detection box to determine whether the point is a dynamic feature point; if it is, the dynamic feature point is eliminated. Further, multiple feature point optimization strategies were developed for VSLAM in dynamic environments. A public data set and an actual dynamic scenario were used for testing. The accuracy of the proposed algorithm was significantly improved compared to that of ORB-SLAM3. This work provides a theoretical foundation for the practical application of a dynamic VSLAM algorithm.https://www.mdpi.com/1424-8220/23/23/9592dynamic VSLAMORB-SLAM3YOLOv5object classificationdepth information fusion |
| spellingShingle | Peichao Cong Junjie Liu Jiaxing Li Yixuan Xiao Xilai Chen Xinjie Feng Xin Zhang YDD-SLAM: Indoor Dynamic Visual SLAM Fusing YOLOv5 with Depth Information Sensors dynamic VSLAM ORB-SLAM3 YOLOv5 object classification depth information fusion |
| title | YDD-SLAM: Indoor Dynamic Visual SLAM Fusing YOLOv5 with Depth Information |
| title_full | YDD-SLAM: Indoor Dynamic Visual SLAM Fusing YOLOv5 with Depth Information |
| title_fullStr | YDD-SLAM: Indoor Dynamic Visual SLAM Fusing YOLOv5 with Depth Information |
| title_full_unstemmed | YDD-SLAM: Indoor Dynamic Visual SLAM Fusing YOLOv5 with Depth Information |
| title_short | YDD-SLAM: Indoor Dynamic Visual SLAM Fusing YOLOv5 with Depth Information |
| title_sort | ydd slam indoor dynamic visual slam fusing yolov5 with depth information |
| topic | dynamic VSLAM ORB-SLAM3 YOLOv5 object classification depth information fusion |
| url | https://www.mdpi.com/1424-8220/23/23/9592 |
| work_keys_str_mv | AT peichaocong yddslamindoordynamicvisualslamfusingyolov5withdepthinformation AT junjieliu yddslamindoordynamicvisualslamfusingyolov5withdepthinformation AT jiaxingli yddslamindoordynamicvisualslamfusingyolov5withdepthinformation AT yixuanxiao yddslamindoordynamicvisualslamfusingyolov5withdepthinformation AT xilaichen yddslamindoordynamicvisualslamfusingyolov5withdepthinformation AT xinjiefeng yddslamindoordynamicvisualslamfusingyolov5withdepthinformation AT xinzhang yddslamindoordynamicvisualslamfusingyolov5withdepthinformation |