赵健,李虹

• 1:太原科技大学电子信息工程学院

摘要(Abstract)：

针对同步定位与地图建立(simultaneous localization and mapping, SLAM)算法在动态环境下存在位姿估计和地图构建误差较大的问题，提出一种光流语义分割方法用于增加动态场景下的可运行性。将ORB-SLAM2系统与YOLOv5模型结合，对传入图像提取特征点的同时将YOLOv5网络模型语义分割后的物体分为高、中、低动态物体。利用运动一致性检测算法，对三种检测物体动态阈值判断，辨别其是否需要剔除特征点，增加ORB-SLAM2算法在动态环境下的运行精度。为加快系统运行速度，用LK光流法加快普通帧与普通帧之间的匹配，其原理为使用LK光流匹配特征点代替ORB特征点匹配，大大的缩小运行时间，同时运行误差变化不大。实验在TUM数据集下测试，平均每一帧提取2 000个特征点，在增加LK光流后缩短0.01 s以上，若在900帧数据集下，可缩短9 s.其绝对轨迹误差对比于ORB-SLAM2和DS-SLAM平均提升在95%与30%以上，证明了算法在动态场景下良好的运行精度与鲁棒性。

关键词(KeyWords)： 动态环境;语义分割;一致性检测算法;LK光流;动态阈值

Abstract:

Keywords:

基金项目(Foundation): 山西省重点研发计划(201803D121123)

作者(Author): 赵健,李虹

参考文献(References)：

• [1] MUR-ARTAL R,MONTIEL J M M,TARDOS J D.ORB-SLAM:A Versatile and Accurate Monocular SLAM System[J].IEEE Transactions on Robotics,2015,31(5):1147-1163.
• [2] ENGEL J,SCHPS T,CREMERS D.LSD-SLAM:Large-scale direct monocular SLAM[C]//European Conference on Computer Vision.Springer,Cham,2014.
• [3] FANG Y,DAI B.An improved moving target detecting and tracking based on Optical Flow technique and Kalman filter[C]//(2009 4th International Conference on Computer Science Education).IEEE,2009.
• [4] KUNDU A,KRISHNA K M,SIVASWAMY J.Moving object detection by multi-view geometric techniques from a single camera mounted robot[C]//2009 IEEE/RSJ International Conference on Intelligent Robots and Systems.Piscataway,NJ,USA:IEEE,2009:4306-4312.
• [5] LIM H,CHOI Y,KHAC C N,et al.Dense Optical flow based Moving Object Detection at Dynamic Scenes[J].IEMEK Journal of Embedded Systems and Applications,2016,11(5):277-285.
• [6] ZHONG F,SHENG W,ZHANG Z,et al.Detect-SLAM:Making Object Detection and SLAM Mutually Beneficial[C]//2018 IEEE Winter Conference on Applications of Computer Vision(WACV).IEEE,2018.
• [7] YU C,LIU Z,LIU X,et al.DS-SLAM:A Semantic Visual SLAM towards Dynamic Environments[C]//2018 IEEE/RSJ International Conference on Intelligent Robots and Systems(IROS).IEEE,2018.
• [8] 郑思诚，孔令华，游通飞，等.动态环境下基于深度学习的语义SLAM算法[J].计算机应用，2021,41(10):2945-2951.
• [9] BESCóS B,FáCIL J M,CIVERA J,et al.DynSLAM:Tracking,Mapping and Inpainting in Dynamic Scenes[J].IEEE Robotics and Automation Letters,2018,3(4):1-10.
• [10] 付豪，徐和根，张志明，等.动态环境下基于语义和光流约束的视觉同步定位与地图构建[J].计算机应用，2021,41(11):3337-3344.
• [11] 姜昊辰，刘衍青，彭镜铨，等.基于语义先验和深度约束的室内动态场景RGB-D SLAM算法[J].信息与控制，2021,50(3):275-286.
• [12] 徐晓苏，安仲帅.基于深度学习的室内动态场景下的VSLAM方法[J].中国惯性技术学报，2020,28(4):480-486.
• [13] REDMON J,DIVVALA S,GIRSHICK R,et al.You Only Look Once:Unified,Real-Time Object Detection[C]//Computer Vision & Pattern Recognition,2016.
• [14] LIU W,ANGUELOV D,ERHAN D,et al.SSD:Single Shot MultiBox Detector[C]//European Conference on Computer Vision.Springer,2016.