Journal of Computer Applications ›› 2025, Vol. 45 ›› Issue (2): 610-615.DOI: 10.11772/j.issn.1001-9081.2024020227

• Multimedia computing and computer simulation • Previous Articles    

Dynamic visual SLAM algorithm incorporating object detection and feature point association

Shijia WEN, Shijun JING()   

  1. School of Instrument Science and Engineering,Southeast University,Nanjing Jiangsu 210096,China
  • Received:2024-03-05 Revised:2024-05-16 Accepted:2024-05-20 Online:2024-06-04 Published:2025-02-10
  • Contact: Shijun JING
  • About author:WEN Shijia, born in 2000, M. S. candidate. Her research interests include mobile robot navigation, visual simultaneous location and mapping, path planning.

结合目标检测和特征点关联的动态视觉SLAM算法

文诗佳, 金世俊()   

  1. 东南大学 仪器科学与工程学院,南京 210096
  • 通讯作者: 金世俊
  • 作者简介:文诗佳(2000—),女,湖南衡阳人,硕士研究生,主要研究方向:移动机器人导航、视觉同时定位与建图、路径规划;

Abstract:

Aiming at the problem that dynamic objects interfere with the normal operation of Simultaneous Localization And Mapping (SLAM) system seriously, a dynamic visual SLAM algorithm based on object detection and feature point association was proposed. Firstly, the YOLOv5 (You Only Look Once version 5) object detection network was used to obtain information about potential dynamic objects in environment, and the missed detection of the image was compensated on the basis of simple target tracking. Secondly, in order to solve the problem that the geometric constraint method of single feature point is prone to misjudgment, the feature point association was established according to the positional information and optical flow information of the image, and then combined with the epipolar constraint, dynamics of the relation network was judged. Thirdly, the two methods were combined to eliminate dynamic feature points in the image, and the remaining static feature points were weighted to estimate the camera pose. Finally, a dense point cloud map was established for the static environment. Experimental results of comparison and ablation on TUM (Technical University of Munich) public dataset demonstrate that the Root Mean Square Error (RMSE) in Absolute Trajectory Error (ATE) of the proposed algorithm is reduced by at least 95.22% and 5.61% respectively compared to ORB-SLAM2 and DS-SLAM (Dynamic Semantic SLAM) in highly dynamic scenarios. It can be seen that the proposed algorithm can improve accuracy and robustness while ensuring real-time performance.

Key words: dynamic environment, object detection, Simultaneous Localization And Mapping (SLAM), dense point cloud map, optical flow method

摘要:

针对动态物体严重干扰同时定位与建图(SLAM)系统正常运行的问题,提出一种基于目标检测和特征点关联的动态视觉SLAM算法。首先,利用YOLOv5目标检测网络得到环境中潜在动态物体的信息,并基于简易目标跟踪对图像漏检进行补偿;其次,为解决单一特征点的几何约束方法易出现误判的问题,依据图像的位置信息和光流信息建立特征点关联,再结合极线约束判断关系网的动态性;再次,结合两种方法剔除图像中的动态特征点,并用剩余的静态特征点加权估计位姿;最后,对静态环境建立稠密点云地图。在TUM(Technical University of Munich)公开数据集上的对比和消融实验的结果表明,与ORB-SLAM2和DS-SLAM(Dynamic Semantic SLAM)相比,所提算法在高动态场景下的绝对轨迹误差(ATE)中的均方根误差(RMSE)分别至少降低了95.22%和5.61%。可见,所提算法在保证实时性的同时提高了准确性和鲁棒性。

关键词: 动态环境, 目标检测, 同时定位与建图, 稠密点云地图, 光流法

CLC Number: