面向动态环境的紧耦合视觉惯性SLAM改进算法

doi:10.3778/j.issn.1002-8331.2310-0092

摘要/Abstract

摘要： SLAM（simultaneous localization and mapping）是无人载体实现自主导航定位的关键技术。针对传统视觉SLAM系统在动态场景下导航定位精度低的问题，在视觉SLAM系统的基础上引入惯性传感器（inertial measurement unit）。在ORB-SLAM3系统的基础上设计了一种面向动态环境的视觉惯性SLAM系统。提出一种基于向量场一致性（vector field consensus，VFC）的稀疏光流法来追踪图像的特征点并计算基础矩阵，分别利用光流对极几何约束和惯性传感器信息计算特征点的动态概率，提出一种联合的动态特征检测方法计算特征点的总动态概率，并将动态概率大于阈值的特征点进行剔除，在SLAM系统的前端实现了视觉信息与惯性运动信息的紧耦合。在数据集上的实验结果表明，该视觉惯性SLAM改进算法有良好的性能表现。

关键词: 同时定位与地图创建（SLAM）, 视觉惯性紧耦合, 动态环境, 向量场一致性, ORB-SLAM3

Abstract: SLAM (simultaneous localization and mapping) is a key technology for unmanned vehicles to achieve autonomous navigation and positioning. In response to the problem of low accuracy of traditional visual SLAM system in dynamic environment, inertial measurement unit (IMU) is imported. A visual-inertial SLAM system for dynamic environment is designed based on the ORB-SLAM3 system. A sparse optical flow method based on vector field consensus (VFC) is proposed to track the feature points and calculate the fundamental matrix between images. The dynamic probability of feature points is calculated by epipolar geometry constraints of optical flow and IMU. A united dynamic feature detection algorithm is proposed to calculate the dynamic probability of feature points, and feature points that dynamic probability is greater than threshold are removed. This algorithm achieves tight coupling of visual information and inertial information in the front-end of the SLAM system. The experimental test results on the datasets indicate that the improved visual-inertial SLAM algorithm has good performance.

Key words: simultaneous localization and mapping (SLAM), visual-inertial tightly-coupled, dynamic environment, vector field consensus, ORB-SLAM3

郭瑞奇, 修睿, 孙勇, 毛喆. 面向动态环境的紧耦合视觉惯性SLAM改进算法[J]. 计算机工程与应用, 2025, 61(4): 339-348.

GUO Ruiqi, XIU Rui, SUN Yong, MAO Zhe. Improved Tightly-Coupled Visual-Inertial SLAM Algorithm for Dynamic Environment[J]. Computer Engineering and Applications, 2025, 61(4): 339-348.

参考文献

[1] SMITH R, SELF M, CHEESEMAN P. Estimating uncertain spatial relationships in robotics[J]. Machine Intelligence & Pattern Recognition, 1988, 5(5): 435-461.
[2] CAMPOS C, ELVIRA R, RODRíGUEZ J J G, et al. 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] QIN T, LI P L, SHEN S J. VINS-mono: a robust and versatile monocular visual-inertial state estimator[J]. IEEE Transactions on Robotics, 2018, 34(4): 1004-1020.
[4] SUN K, MOHTA K, PFROMMER B, et al. Robust stereo visual inertial odometry for fast autonomous flight[J]. IEEE Robotics and Automation Letters, 2018, 3(2): 965-972.
[5] MUR-ARTAL R, TARDóS J D. Visual-inertial monocular SLAM with map reuse[J]. IEEE Robotics and Automation Letters, 2017, 2(2): 796-803.
[6] KUNDU A, KRISHNA K M, SIVASWAMY J. Moving object detection by multi-view geometric techniques from a single camera mounted robot[C]//Proceedings of the 2009 IEEE/RSJ International Conference on Intelligent Robots and Systems. Piscataway: IEEE, 2009: 4306-4312.
[7] DING X Y. Moving object detection in dynamic background[J]. Journal of Physics: Conference Series, 2021, 1955(1): 012113.
[8] 林凯, 梁新武, 蔡纪源. 基于重投影深度差累积图与静态概率的动态RGB-D SLAM算法[J]. 浙江大学学报 (工学版), 2022, 56(6): 1062-1070.
LIN K, LIANG X W, CAI J Y. Dynamic RGB-D SLAM algorithm based on reprojection depth difference cumulative map and static probability[J]. Journal of Zhejiang University (Engineering Science), 2022, 56(6): 1062-1070.
[9] 魏彤, 李绪. 动态环境下基于动态区域剔除的双目视觉SLAM算法[J]. 机器人, 2020, 42(3): 336-345.
WEI T, LI X. Binocular vision SLAM algorithm based on dynamic region elimination in dynamic environment[J]. Robot, 2020, 42(3): 336-345.
[10] BRASCH N, BOZIC A, LALLEMAND J, et al. Semantic monocular SLAM for highly dynamic environments[C]//Proceedings of the 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems. Piscataway: IEEE, 2018: 393-400.
[11] BESCOS B, FáCIL J M, CIVERA J, et al. DynaSLAM: tracking, mapping, and inpainting in dynamic scenes[J]. IEEE Robotics and Automation Letters, 2018, 3(4): 4076-4083.
[12] BESCOS B, CAMPOS C, TARDóS J D, et al. DynaSLAM II: tightly-coupled multi-object tracking and SLAM[J]. IEEE Robotics and Automation Letters, 2021, 6(3): 5191-5198.
[13] CUI L, MA C. SOF-SLAM: a semantic visual SLAM for dynamic environments[J]. IEEE Access, 2019, 7: 166528-166539.
[14] 王富强, 王强, 李敏, 等. 基于动态分级的自适应运动目标处理SLAM算法[J]. 计算机应用研究, 2023, 40(8): 2361-2366.
WANG F Q, WANG Q, LI M, et al. Adaptive moving object processing SLAM algorithm based on scene dynamic classification[J]. Application Research of Computers, 2023, 40(8): 2361-2366.
[15] YU C, LIU Z, LIU X J, et al. DS-SLAM: a semantic visual SLAM towards dynamic environments[C]//Proceedings of the 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems. Piscataway: IEEE, 2018: 1168-1174.
[16] MA J Y, ZHAO J, TIAN J W, et al. Robust point matching via vector field consensus[J]. IEEE Transactions on Image Processing, 2014, 23(4): 1706-1721.
[17] DEMPSTER A P, LAIRD N M, RUBIN D B. Maximum likelihood from incomplete data via the EM algorithm[J]. Journal of the Royal Statistical Society Series B: Statistical Methodology, 1977, 39(1): 1-22.
[18] BAKER S, MATTHEWS I. Lucas-Kanade 20 years on: a unifying framework[J]. International Journal of Computer Vision, 2004, 56(3): 221-255.
[19] STURM J, ENGELHARD N, ENDRES F, et al. A benchmark for the evaluation of RGB-D SLAM systems[C]//Proceedings of the 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems. Piscataway: IEEE, 2012: 573-580.
[20] GEIGER A, LENZ P, URTASUN R. Are we ready for autonomous driving? The KITTI vision benchmark suite[C]//Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2012: 3354-3361.
[21] BURRI M, NIKOLIC J, GOHL P, et al. The EuRoC micro aerial vehicle datasets[J]. International Journal of Robotics Research, 2016, 35(10): 1157-1163.