PDF
Full
Outline
Aiming at the problem of robot pose estimation bias and imperfect map construction caused by moving objects in dynamic scenes, an ORB-SLAM3 algorithm for dynamic scene optimization is proposed. Firstly, the dynamic object is detected by the improved YOLOv5s algorithm and the associated feature points are preliminarily removed. Then, the missing dynamic feature points are further filtered by combining LK optical flow tracking and epipolar geometric constraint analysis based on fundamental matrix, so as to improve the accuracy of environment perception and pose estimation. At the same time, the corresponding point cloud information is generated by filtering the key frames of dynamic information to realize the construction of 3D dense static map. The test results in indoor dynamic scenes show that compared with the traditional ORB-SLAM3, the absolute trajectory error and relative pose error of the proposed algorithm are reduced by 55.2% and 93.7% respectively in the office environment, and by 24.3% and 40.2% in the corridor scene, which verifies the robustness advantage of the proposed algorithm in dynamic scenes.
PDF
179
83
| 科 Family | 属数 Number of genus | 种数 Number of species | 占总种数比例 Percentage of total species (%) | 属 Genus | 种数 Number of species | 占总种数比例 Percentage of total species (%) |
|---|---|---|---|---|---|---|
| 鹅膏菌科Amanitaceae | 2 | 11 | 5.26 | 鹅膏菌属 Amanita | 10 | 4.78 |
| 小菇科 Mycenaceae | 2 | 12 | 5.74 | 丝盖伞属 Inocybe | 5 | 2.39 |
| 多孔菌科 Polyporaceae | 8 | 14 | 6.70 | 蜡蘑属 Laccaria | 5 | 2.39 |
| 红菇科 Russulaceae | 3 | 23 | 11.00 | 小皮伞属 Marasmius | 6 | 2.87 |
| 小菇属 Mycena | 11 | 5.26 | ||||
| 光柄菇属 Pluteus | 5 | 2.39 | ||||
| 红菇属 Russula | 17 | 8.13 | ||||
| 栓菌属 Trametes | 5 | 2.39 |
关闭全屏
No. 86 Xueyuan South Road, Haidian District, Beijing
010-62199257
qkjq@cast.org.cn
Copyright © 2025 China Association for Science and Technology. All rights reserved. For all open access content, the relevant licensing terms apply.
Sponsored by the Office of the Leading Group for Cybersecurity and Informatization of CAST, and supported by Science and Technology Review Publishing House
