SGDO-SLAM: A Semantic RGB-D SLAM System with Coarse-to-Fine Dynamic Rejection and Static Weighted Optimization
Vision sensor-based simultaneous localization and mapping (SLAM) systems are essential for mobile robots to locate and generate spatial models of their surroundings. However, the majority of visual SLAM systems assume static settings, leading to significant accuracy degradation in dynamic scenes. We...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-06-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/25/12/3734 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850165034527752192 |
|---|---|
| author | Qiming Hu Shuwen Wang Nanxing Chen Wei Li Jiayu Yuan Enhui Zheng Guirong Wang Weimin Chen |
| author_facet | Qiming Hu Shuwen Wang Nanxing Chen Wei Li Jiayu Yuan Enhui Zheng Guirong Wang Weimin Chen |
| author_sort | Qiming Hu |
| collection | DOAJ |
| description | Vision sensor-based simultaneous localization and mapping (SLAM) systems are essential for mobile robots to locate and generate spatial models of their surroundings. However, the majority of visual SLAM systems assume static settings, leading to significant accuracy degradation in dynamic scenes. We present SGDO-SLAM, a real-time RGB-D semantic-aware SLAM framework, building upon ORB-SLAM2 to address non-static environments. Firstly, a multi-constraint dynamic rejection method from coarse to fine is proposed. The method starts with coarse rejection by combining semantic and geometric information, followed by detailed rejection using depth information, where static quality weights are quantified based on depth consistency constraints. The method achieves accurate dynamic scene perceptions and improves the accuracy of the system’s positioning. Then, a position optimization method driven by static quality weights is proposed, which prioritizes high-quality static features to enhance pose estimation. Finally, a visualized dense point cloud map is established. We performed experimental evaluations on the TUM RGB-D dataset and the Bonn dataset. The experimental results demonstrate that SGDO-SLAM reduces the absolute trajectory error performance metrics by 95% compared to the ORB-SLAM2 algorithm, while maintaining real-time efficiency and achieving state-of-the-art accuracy in dynamic scenarios. |
| format | Article |
| id | doaj-art-456daf30459f4904a64fcfca39c4e336 |
| institution | OA Journals |
| issn | 1424-8220 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-456daf30459f4904a64fcfca39c4e3362025-08-20T02:21:50ZengMDPI AGSensors1424-82202025-06-012512373410.3390/s25123734SGDO-SLAM: A Semantic RGB-D SLAM System with Coarse-to-Fine Dynamic Rejection and Static Weighted OptimizationQiming Hu0Shuwen Wang1Nanxing Chen2Wei Li3Jiayu Yuan4Enhui Zheng5Guirong Wang6Weimin Chen7School of Mechanical and Electrical Engineering, China Jiliang University, Hangzhou 310018, ChinaSchool of Mechanical and Electrical Engineering, China Jiliang University, Hangzhou 310018, ChinaSchool of Mechanical and Electrical Engineering, China Jiliang University, Hangzhou 310018, ChinaSchool of Mechanical and Electrical Engineering, China Jiliang University, Hangzhou 310018, ChinaSchool of Mechanical and Electrical Engineering, China Jiliang University, Hangzhou 310018, ChinaSchool of Mechanical and Electrical Engineering, China Jiliang University, Hangzhou 310018, ChinaSchool of Mechanical and Electrical Engineering, China Jiliang University, Hangzhou 310018, ChinaSchool of Mechanical and Electrical Engineering, China Jiliang University, Hangzhou 310018, ChinaVision sensor-based simultaneous localization and mapping (SLAM) systems are essential for mobile robots to locate and generate spatial models of their surroundings. However, the majority of visual SLAM systems assume static settings, leading to significant accuracy degradation in dynamic scenes. We present SGDO-SLAM, a real-time RGB-D semantic-aware SLAM framework, building upon ORB-SLAM2 to address non-static environments. Firstly, a multi-constraint dynamic rejection method from coarse to fine is proposed. The method starts with coarse rejection by combining semantic and geometric information, followed by detailed rejection using depth information, where static quality weights are quantified based on depth consistency constraints. The method achieves accurate dynamic scene perceptions and improves the accuracy of the system’s positioning. Then, a position optimization method driven by static quality weights is proposed, which prioritizes high-quality static features to enhance pose estimation. Finally, a visualized dense point cloud map is established. We performed experimental evaluations on the TUM RGB-D dataset and the Bonn dataset. The experimental results demonstrate that SGDO-SLAM reduces the absolute trajectory error performance metrics by 95% compared to the ORB-SLAM2 algorithm, while maintaining real-time efficiency and achieving state-of-the-art accuracy in dynamic scenarios.https://www.mdpi.com/1424-8220/25/12/3734simultaneous localization and mapping (SLAM)dynamic environmentsdepth consistency constraintsstatic quality weightsvision sensor |
| spellingShingle | Qiming Hu Shuwen Wang Nanxing Chen Wei Li Jiayu Yuan Enhui Zheng Guirong Wang Weimin Chen SGDO-SLAM: A Semantic RGB-D SLAM System with Coarse-to-Fine Dynamic Rejection and Static Weighted Optimization Sensors simultaneous localization and mapping (SLAM) dynamic environments depth consistency constraints static quality weights vision sensor |
| title | SGDO-SLAM: A Semantic RGB-D SLAM System with Coarse-to-Fine Dynamic Rejection and Static Weighted Optimization |
| title_full | SGDO-SLAM: A Semantic RGB-D SLAM System with Coarse-to-Fine Dynamic Rejection and Static Weighted Optimization |
| title_fullStr | SGDO-SLAM: A Semantic RGB-D SLAM System with Coarse-to-Fine Dynamic Rejection and Static Weighted Optimization |
| title_full_unstemmed | SGDO-SLAM: A Semantic RGB-D SLAM System with Coarse-to-Fine Dynamic Rejection and Static Weighted Optimization |
| title_short | SGDO-SLAM: A Semantic RGB-D SLAM System with Coarse-to-Fine Dynamic Rejection and Static Weighted Optimization |
| title_sort | sgdo slam a semantic rgb d slam system with coarse to fine dynamic rejection and static weighted optimization |
| topic | simultaneous localization and mapping (SLAM) dynamic environments depth consistency constraints static quality weights vision sensor |
| url | https://www.mdpi.com/1424-8220/25/12/3734 |
| work_keys_str_mv | AT qiminghu sgdoslamasemanticrgbdslamsystemwithcoarsetofinedynamicrejectionandstaticweightedoptimization AT shuwenwang sgdoslamasemanticrgbdslamsystemwithcoarsetofinedynamicrejectionandstaticweightedoptimization AT nanxingchen sgdoslamasemanticrgbdslamsystemwithcoarsetofinedynamicrejectionandstaticweightedoptimization AT weili sgdoslamasemanticrgbdslamsystemwithcoarsetofinedynamicrejectionandstaticweightedoptimization AT jiayuyuan sgdoslamasemanticrgbdslamsystemwithcoarsetofinedynamicrejectionandstaticweightedoptimization AT enhuizheng sgdoslamasemanticrgbdslamsystemwithcoarsetofinedynamicrejectionandstaticweightedoptimization AT guirongwang sgdoslamasemanticrgbdslamsystemwithcoarsetofinedynamicrejectionandstaticweightedoptimization AT weiminchen sgdoslamasemanticrgbdslamsystemwithcoarsetofinedynamicrejectionandstaticweightedoptimization |