| 133 | 1 | 64 |
| 下载次数 | 被引频次 | 阅读次数 |
针对视觉跟踪过程中出现的低纹理场景，以及由于运动模糊等因素导致同时定位与建图（SLAM）系统对环境特征的识别和跟踪偏差较大，甚至失效的问题，设计一种点线面多特征融合跟踪方法：改进快速直线检测算法，显化隐藏参数并使其可调整；然后针对短线特征进行筛选与合并，提高系统对环境信息的利用效率；最后在曼哈顿（Manhattan）世界假设中，提出一种结合几何与外观的线特征跟踪策略，利用线特征之间的几何关系建立约束优化匹配与跟踪，提高位姿估计的精度。实验结果表明，改进的特征跟踪方法相较于原算法在跟踪效率和整体定位精度方面都有明显提升，验证了该算法的有效性。
Abstract：Aiming at the low texture scene in the visual tracking process and the problem that the simultaneous localizationand mapping(SLAM) system has a large deviation or even failure in the recognition and tracking of environmental features dueto motion blur and other factors, the paper proposed a point-line-surface multi-feature fusion tracking method: the fast straight-line segment detection algorithm was improved, and the hidden parameters were displayed and made adjustable; then the short-line features were screened and combined to improve the utilization efficiency of environmental information; finally, in theManhattan world hypothesis, a line feature tracking strategy combining geometry and appearance was put forward, and thegeometric relationship between line features was used to establish the constraint optimization matching and tracking, so as toimprove the precision of pose estimation. Experimental result showed that the improved feature tracking method would haveobvious improvement in tracking efficiency and overall positioning accuracy compared with the original algorithm, which couldverify the effectiveness of the proposed algorithm.
[1] SHAN Z, LI R, SCHWERTFEGER S. Rgbd-inertial trajectory estimation and mapping for ground robots[J]. Sensors,2019, 19(10):2251.
[2] KLEIN G, MURRAY D. Parallel tracking and mapping for small AR workspaces[C]//IEEE&ACM International Symposium on Mixed&Augmented Reality. New York, NJ, USA:ACM, 2007:1-10.
[3] MUR A R, TARDOS J D. Orb-slam2:An open-source slam system for monocular, stereo, and RGB-D cameras[J].IEEE Transactions on Robotics, 2017, 33(5):1255-1262.
[4] ENGEL J, SCH?PS T, CREMERS D. LSD-SLAM:Large-scale direct monocular SLAM[C]//European Conference on Computer Vision. Cham:Springer International Publishing, 2014:834-849.
[5]金琼瑶.室内结构化环境下基于点线特征的RGB-DSLAM系统研究[D].济南:山东大学, 2020.
[6]黄平,曹镇,黄俊杰.一种基于线特征的RGB-D视觉里程计算法[J].中国惯性技术学报, 2021, 29(3):340-349.
[7] LIM H, KIM Y, JUNG K, et al. Avoiding degeneracy for monocular visual SLAM with point and line features[C]//Proc. IEEE International Conference on Robotics and Automation(ICRA).[S.l.]:IEEE, 2021:11675-11681.
[8]王丹,黄鲁,李垚.基于点线特征的单目视觉同时定位与地图构建算法[J].机器人, 2019, 41(3):392-403.
[9] PUMAROLA A, VAKHITOV A, AGUDO A, et al. PL-SLAM:Real-time monocular visual SLAM with points and lines[C]//2017 IEEE International Conference on Robotics and Automation(ICRA).[S.l.]:IEEE, 2017:4503-4508.
[10] VON GIOI R G, JAKUBOWICZ J, MOREL J M, et al. LSD:A fast line segment detector with a false detection control[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008, 32(4):722-732.
[11] SCH?PS T, SATTLER T, POLLEFEYS M. Surfelmeshing:Online surfel-based mesh reconstruction[C]//IEEE Transactions on Pattern Analysis and Machine Intelligence.[S.l.]:IEEE, 2019.
[12] YUNUS R, LI Y, TOMBARI F. Manhattanslam:Robust planar tracking and mapping leveraging mixture of manhattan frames[C]//2021 IEEE International Conference on Robotics and Automation(ICRA).[S.l.]:IEEE,2021:6687-6693.
[13]孙新柱,龚光强,陈孟元,等.室内场景下基于曼哈顿约束的多重特征视觉SLAM方法[J].中国惯性技术学报, 2023, 31(9):890-899. DOI:10.13695/j.cnki.12-1222/o3.2023.09.006.
[14] RUBLEE E, RABAUD V, KONOLIGE K, et al. ORB:An efficient alternative to SIFT or SURF[C]//2011International Conference on Computer Vision.[S.l.]:IEEE, 2011:2564-2571.
[15] ZHANG X, WANG W, QI X, et al. Stereo plane slam based on intersecting lines[C]//2021 IEEE/RSJ International Conference on Intelligent Robots and Systems(IROS).[S.l.]:IEEE, 2021:6566-6572.
[16] GOMEZ-OJEDA R, MORENO F A, ZUNIGA-NO?L D, et al. PL-SLAM:A stereo SLAM system through the combination of points and line segments[J]. IEEE Transactions on Robotics, 2019, 35(3):734-746.
[17]曹一波,赵鹏飞,朱海文,等.结构化环境下基于点线面特征融合的SLAM算法[J].计算机技术与发展, 2023,33(7):85-90.
[18] LIM H, JEON J, MYUNG H. UV-SLAM:Unconstrained line-based SLAM using vanishing points for structural mapping[C]//IEEE Robotics and Automation Letters.[S.l.]:IEEE, 2022, 7(2):1518–1525.
[19] FU Q, WANG J, YU H, et al. PL-VINS:Real-time monocular visual-inertial SLAM with point and line features[J].arXiv preprint arXiv:2009.07462, 2020.
[20] ZHANG Xiaoyu, WANG Wei, QI Xianyu, et al. Point-plane slam using supposed planes for indoor environments[J].Sensors, 2019, 19(17):3795.
基本信息:
DOI：10.16547/j.cnki.10-1096.20250221
中图分类号:TP391.41;TP242
引用信息:
[1]程鹏,王珂,邓甘霖,等.改进几何约束的多特征视觉Manhattan-SLAM[J].导航定位学报,2025,13(02):172-178.DOI:10.16547/j.cnki.10-1096.20250221.
2024-05-07
2024
2024-07-16
2024-08-19
2024
1
2025-04-20
2025-04-20