跳到主要导航 跳到搜索 跳到主要内容

Robust SLAM Algorithm in Dynamic Environment Using Optical Flow

  • Beihang University

科研成果: 书/报告/会议事项章节会议稿件同行评审

摘要

Visual SLAM(Simultaneous Localization and Mapping) is one of the hottest research areas nowadays. Most of the SLAM methods assume that the scene is stationary. It is difficult to face the factual complex environment. To solve this problem, an improved RGB-D SLAM algorithm combined with Optical Flow and RANSAC (Random Sample Consensus) was proposed in this paper. Optical flow is used to detect moving objects in the scene. RANSAC is used to calculate the homography matrix and optimize matching results. We used a common data set for experimental verification. The results show that this algorithm can effectively detect dynamic objects, improve the accuracy of visual odometry and the robustness of the system.

源语言英语
主期刊名Proceedings of 2019 Chinese Intelligent Systems Conference - Volume I
编辑Yingmin Jia, Junping Du, Weicun Zhang
出版商Springer Verlag
681-689
页数9
ISBN(印刷版)9789813296817
DOI
出版状态已出版 - 2020
活动Chinese Intelligent Systems Conference, CISC 2019 - Haikou, 中国
期限: 26 10月 201927 10月 2019

出版系列

姓名Lecture Notes in Electrical Engineering
592
ISSN(印刷版)1876-1100
ISSN(电子版)1876-1119

会议

会议Chinese Intelligent Systems Conference, CISC 2019
国家/地区中国
Haikou
时期26/10/1927/10/19

指纹

探究 'Robust SLAM Algorithm in Dynamic Environment Using Optical Flow' 的科研主题。它们共同构成独一无二的指纹。

引用此