Visual SLAM Algorithm Optimized by Dynamic Feature Elimination and Keyframe Selection
Main Article Content
Keywords
visual slam, dynamic feature point removal, keyframe optimization, dynamic environment
Abstract
Traditional visual SLAM algorithms struggle to maintain stability in dynamic scenes, often resulting in tracking failures and low positioning accuracy. This paper proposes the YDK-SLAM algorithm, which is suitable for dynamic environments. Building upon ORB-SLAM3, a dynamic feature detection thread has been added. It employs the YOLOv8 object detection network to identify potential moving objects, precisely removing dynamic feature points through depth information and multi-view geometry methods, and utilizing the remaining static feature points for pose estimation. Simultaneously, by employing optical flow methods and adaptive keyframe insertion strategies in the tracking thread based on camera motion characteristics, thereby enhancing the operational speed of YDK-SLAM. Experimental results on the public TUM dataset demonstrate that compared to ORB-SLAM3, YDK-SLAM achieves an average reduction of 70.73% in absolute trajectory error. In high-dynamic scenes, the error is reduced by an average of 93.73%, effectively enhancing the system's positioning accuracy and robustness.
References
- [1] Placed J A, et al. A survey on active simultaneous localization and mapping: State of the art and new frontiers[J]. IEEE Transactions on Robotics, 2023, 39(3): 1686-1705.
- [2] Campos C, Elvira R, Rodríguez J J G, Montiel J M M, Tardós J D. ORB-SLAM3: An Accurate Open-Source Library for Visual, Visual–Inertial, and Multimap SLAM[J]. IEEE Transactions on Robotics, 2021, 37(6): 1874-1890.
- [3] Yu C, et al. DS-SLAM: A Semantic Visual SLAM towards Dynamic Environments[C]. Madrid, Spain: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2018: 1168-1174.
- [4] Bescos B, Fácil J M, Civera J, Neira J. DynaSLAM: Tracking, Mapping, and Inpainting in Dynamic Scenes[J]. IEEE Robotics and Automation Letters, 2018, 3(4): 4076-4083.
- [5] Pan Z, Hou J, Yu L. Optimization RGB-D 3-D Reconstruction Algorithm Based on Dynamic SLAM[J]. IEEE Transactions on Instrumentation and Measurement, 2023, 72: 1-13.
- [6] Cheng S, Sun C, Zhang S, Zhang D. SG-SLAM: A Real-Time RGB-D Visual SLAM Toward Dynamic Scenes With Semantic and Geometric Information[J]. IEEE Transactions on Instrumentation and Measurement, 2023, 72: 1-12.
- [7] Su P, Luo S, Huang X. Real-Time Dynamic SLAM Algorithm Based on Deep Learning[J]. IEEE Access, 2022, 10: 87754-87766.
- [8] Chang Z, Wu H, Li C. YOLOv4-tiny-based robust RGB-D SLAM approach with point and surface feature fusion in complex indoor environments[J]. Journal of Field Robotics, 2022, 40(3): 521-534.
- [9] Wen S, Tao S, Liu X, Babiarz A, Yu F R. CD-SLAM: A Real-Time Stereo Visual–Inertial SLAM for Complex Dynamic Environments With Semantic and Geometric Information[J]. IEEE Transactions on Instrumentation and Measurement, 2024, 73: 1-8.
- [10] Fu J, Ma P, Jia Z. An Improved Monocular ORB-SLAM Using Scaled-Invariant Features[C]. Wulumuqi, China: IEEE International Conference on Advanced Information, Mechanical Engineering, Robotics and Automation (AIMERA), 2024: 125-130.
- [11] Fu X, Zhang K, Shen C, Zhu J, Zang L. Recognition and Location Based on Fusion of FAST Algorithm and SIFT Algorithm[C]. Tianjin, China: IEEE 21st International Conference on Communication Technology (ICCT), 2021: 422-426.
- [12] Hsieh J-W, Chen L-C, Chen D-Y. Symmetrical SURF and Its Applications to Vehicle Detection and Vehicle Make and Model Recognition[J]. IEEE Transactions on Intelligent Transportation Systems, 2014, 15(1): 6-20.
- [13] Zu L, Wei C, Sun Q, Zhang M, Sheng N, Ge S S. Adaptive Keyframe Selection Strategy of Visual SLAM in Complex Poses[J]. IEEE Sensors Journal, 2025, 25(1): 1756-1767.
Article Sidebar
This work is licensed under a Creative Commons Attribution 4.0 International License.