SOLO-SLAM: A Parallel Semantic SLAM Algorithm for Dynamic Scenes
Simultaneous localization and mapping (SLAM) is a core technology for mobile robots working in unknown environments. Most existing SLAM techniques can achieve good localization accuracy in static scenes, as they are designed based on the assumption that unknown scenes are rigid. However, real-world...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-09-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/22/18/6977 |
_version_ | 1797482447913353216 |
---|---|
author | Liuxin Sun Junyu Wei Shaojing Su Peng Wu |
author_facet | Liuxin Sun Junyu Wei Shaojing Su Peng Wu |
author_sort | Liuxin Sun |
collection | DOAJ |
description | Simultaneous localization and mapping (SLAM) is a core technology for mobile robots working in unknown environments. Most existing SLAM techniques can achieve good localization accuracy in static scenes, as they are designed based on the assumption that unknown scenes are rigid. However, real-world environments are dynamic, resulting in poor performance of SLAM algorithms. Thus, to optimize the performance of SLAM techniques, we propose a new parallel processing system, named SOLO-SLAM, based on the existing ORB-SLAM3 algorithm. By improving the semantic threads and designing a new dynamic point filtering strategy, SOLO-SLAM completes the tasks of semantic and SLAM threads in parallel, thereby effectively improving the real-time performance of SLAM systems. Additionally, we further enhance the filtering effect for dynamic points using a combination of regional dynamic degree and geometric constraints. The designed system adds a new semantic constraint based on semantic attributes of map points, which solves, to some extent, the problem of fewer optimization constraints caused by dynamic information filtering. Using the publicly available TUM dataset, SOLO-SLAM is compared with other state-of-the-art schemes. Our algorithm outperforms ORB-SLAM3 in accuracy (maximum improvement is 97.16%) and achieves better results than Dyna-SLAM with respect to time efficiency (maximum improvement is 90.07%). |
first_indexed | 2024-03-09T22:33:26Z |
format | Article |
id | doaj.art-06e9b655d2e14c1fa217618fe636cd33 |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-03-09T22:33:26Z |
publishDate | 2022-09-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-06e9b655d2e14c1fa217618fe636cd332023-11-23T18:52:25ZengMDPI AGSensors1424-82202022-09-012218697710.3390/s22186977SOLO-SLAM: A Parallel Semantic SLAM Algorithm for Dynamic ScenesLiuxin Sun0Junyu Wei1Shaojing Su2Peng Wu3College of Intelligence Science and Technology, National University of Defense Technology, Changsha 410073, ChinaCollege of Intelligence Science and Technology, National University of Defense Technology, Changsha 410073, ChinaCollege of Intelligence Science and Technology, National University of Defense Technology, Changsha 410073, ChinaCollege of Intelligence Science and Technology, National University of Defense Technology, Changsha 410073, ChinaSimultaneous localization and mapping (SLAM) is a core technology for mobile robots working in unknown environments. Most existing SLAM techniques can achieve good localization accuracy in static scenes, as they are designed based on the assumption that unknown scenes are rigid. However, real-world environments are dynamic, resulting in poor performance of SLAM algorithms. Thus, to optimize the performance of SLAM techniques, we propose a new parallel processing system, named SOLO-SLAM, based on the existing ORB-SLAM3 algorithm. By improving the semantic threads and designing a new dynamic point filtering strategy, SOLO-SLAM completes the tasks of semantic and SLAM threads in parallel, thereby effectively improving the real-time performance of SLAM systems. Additionally, we further enhance the filtering effect for dynamic points using a combination of regional dynamic degree and geometric constraints. The designed system adds a new semantic constraint based on semantic attributes of map points, which solves, to some extent, the problem of fewer optimization constraints caused by dynamic information filtering. Using the publicly available TUM dataset, SOLO-SLAM is compared with other state-of-the-art schemes. Our algorithm outperforms ORB-SLAM3 in accuracy (maximum improvement is 97.16%) and achieves better results than Dyna-SLAM with respect to time efficiency (maximum improvement is 90.07%).https://www.mdpi.com/1424-8220/22/18/6977SLAMSOLO-SLAMSOLO_V2deep learningnavigationrobotics |
spellingShingle | Liuxin Sun Junyu Wei Shaojing Su Peng Wu SOLO-SLAM: A Parallel Semantic SLAM Algorithm for Dynamic Scenes Sensors SLAM SOLO-SLAM SOLO_V2 deep learning navigation robotics |
title | SOLO-SLAM: A Parallel Semantic SLAM Algorithm for Dynamic Scenes |
title_full | SOLO-SLAM: A Parallel Semantic SLAM Algorithm for Dynamic Scenes |
title_fullStr | SOLO-SLAM: A Parallel Semantic SLAM Algorithm for Dynamic Scenes |
title_full_unstemmed | SOLO-SLAM: A Parallel Semantic SLAM Algorithm for Dynamic Scenes |
title_short | SOLO-SLAM: A Parallel Semantic SLAM Algorithm for Dynamic Scenes |
title_sort | solo slam a parallel semantic slam algorithm for dynamic scenes |
topic | SLAM SOLO-SLAM SOLO_V2 deep learning navigation robotics |
url | https://www.mdpi.com/1424-8220/22/18/6977 |
work_keys_str_mv | AT liuxinsun soloslamaparallelsemanticslamalgorithmfordynamicscenes AT junyuwei soloslamaparallelsemanticslamalgorithmfordynamicscenes AT shaojingsu soloslamaparallelsemanticslamalgorithmfordynamicscenes AT pengwu soloslamaparallelsemanticslamalgorithmfordynamicscenes |