Coarse Semantic-Based Motion Removal for Robust Mapping in Dynamic Environments
SLAM in dynamic environments is still a severe challenge for most feature-based SLAM systems. Moving objects will lead to terrible errors in the calculation of frame tracking and local mapping. We propose a novel method for keypoints selection to lower the negative effect brought by moving objects d...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9075196/ |
| _version_ | 1819169958867763200 |
|---|---|
| author | Shuo Wang Xudong Lv Junbao Li Dong Ye |
| author_facet | Shuo Wang Xudong Lv Junbao Li Dong Ye |
| author_sort | Shuo Wang |
| collection | DOAJ |
| description | SLAM in dynamic environments is still a severe challenge for most feature-based SLAM systems. Moving objects will lead to terrible errors in the calculation of frame tracking and local mapping. We propose a novel method for keypoints selection to lower the negative effect brought by moving objects during map construction. To address this challenge, we concentrate on the combination of coarse semantic information and a feature-based SLAM system. In this article, a modified CenterNet object detector is proposed as the moving object detection thread for providing coarse semantic information and 2D location. The modified CenterNet can provide a faster and more accurate prediction. For each frame in a sequence, objects in a scenario will be classified into two motion states, non-static and static, according to the category prediction from the moving object detection thread. Then important processing called semantic data association is presented for motion removal in frame tracking thread and local mapping thread. In this way, the keypoints on the non-static objects will not be chosen for calculation. Finally, on this basis, a modified real-time SLAM system is presented. Experimental results in TUM indoor dataset show that the proposed SLAM system outperforms the state-of-the-art ORB-SLAM2 in dynamic environments. Besides that, our method has a better speed-accuracy trade-off than other deep learning-based SLAM systems in dynamic environments. |
| first_indexed | 2024-12-22T19:27:46Z |
| format | Article |
| id | doaj.art-1da15e10a390471f826a02deb74ffedf |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-22T19:27:46Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-1da15e10a390471f826a02deb74ffedf2022-12-21T18:15:11ZengIEEEIEEE Access2169-35362020-01-018740487406410.1109/ACCESS.2020.29893179075196Coarse Semantic-Based Motion Removal for Robust Mapping in Dynamic EnvironmentsShuo Wang0https://orcid.org/0000-0003-4058-6897Xudong Lv1https://orcid.org/0000-0002-7850-6883Junbao Li2https://orcid.org/0000-0003-3988-3675Dong Ye3https://orcid.org/0000-0003-3715-5885School of Instrumentation Science and Engineering, Harbin Institute of Technology, Harbin~, ChinaSchool of Instrumentation Science and Engineering, Harbin Institute of Technology, Harbin~, ChinaSchool of Electronic Engineering and Automation, Harbin Institute of Technology, Harbin~, ChinaSchool of Instrumentation Science and Engineering, Harbin Institute of Technology, Harbin~, ChinaSLAM in dynamic environments is still a severe challenge for most feature-based SLAM systems. Moving objects will lead to terrible errors in the calculation of frame tracking and local mapping. We propose a novel method for keypoints selection to lower the negative effect brought by moving objects during map construction. To address this challenge, we concentrate on the combination of coarse semantic information and a feature-based SLAM system. In this article, a modified CenterNet object detector is proposed as the moving object detection thread for providing coarse semantic information and 2D location. The modified CenterNet can provide a faster and more accurate prediction. For each frame in a sequence, objects in a scenario will be classified into two motion states, non-static and static, according to the category prediction from the moving object detection thread. Then important processing called semantic data association is presented for motion removal in frame tracking thread and local mapping thread. In this way, the keypoints on the non-static objects will not be chosen for calculation. Finally, on this basis, a modified real-time SLAM system is presented. Experimental results in TUM indoor dataset show that the proposed SLAM system outperforms the state-of-the-art ORB-SLAM2 in dynamic environments. Besides that, our method has a better speed-accuracy trade-off than other deep learning-based SLAM systems in dynamic environments.https://ieeexplore.ieee.org/document/9075196/Simultaneous localization and mappingmotion removaldynamic environmentsobject detection |
| spellingShingle | Shuo Wang Xudong Lv Junbao Li Dong Ye Coarse Semantic-Based Motion Removal for Robust Mapping in Dynamic Environments IEEE Access Simultaneous localization and mapping motion removal dynamic environments object detection |
| title | Coarse Semantic-Based Motion Removal for Robust Mapping in Dynamic Environments |
| title_full | Coarse Semantic-Based Motion Removal for Robust Mapping in Dynamic Environments |
| title_fullStr | Coarse Semantic-Based Motion Removal for Robust Mapping in Dynamic Environments |
| title_full_unstemmed | Coarse Semantic-Based Motion Removal for Robust Mapping in Dynamic Environments |
| title_short | Coarse Semantic-Based Motion Removal for Robust Mapping in Dynamic Environments |
| title_sort | coarse semantic based motion removal for robust mapping in dynamic environments |
| topic | Simultaneous localization and mapping motion removal dynamic environments object detection |
| url | https://ieeexplore.ieee.org/document/9075196/ |
| work_keys_str_mv | AT shuowang coarsesemanticbasedmotionremovalforrobustmappingindynamicenvironments AT xudonglv coarsesemanticbasedmotionremovalforrobustmappingindynamicenvironments AT junbaoli coarsesemanticbasedmotionremovalforrobustmappingindynamicenvironments AT dongye coarsesemanticbasedmotionremovalforrobustmappingindynamicenvironments |