Research on Visual ORB-SLAM Based on Dynamic Scene

Conference: ICMLCA 2021 - 2nd International Conference on Machine Learning and Computer Application
12/17/2021 - 12/19/2021 at Shenyang, China

Proceedings: ICMLCA 2021

Pages: 4Language: englishTyp: PDF

Personal VDE Members are entitled to a 10% discount on this title

Authors:
Zhang, Dongyang; Yu, Lifeng (Equipment Engineering College, Shenyang Ligong University, Shenyang, Liaoning, China)
Zhang, Hualiang (Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, Liaoning, China)

Abstract:
In recent years, simultaneous localization and mapping SLAM has achieved rapid development. However, when applied to dynamic or mobile targets, the localization and mapping accuracy of SLAM decreases greatly. To solve this problem, a dynamic adaptive SLAM method is proposed. The visual view geometry method was introduced into the classic ORB-SLAM2 method to reduce the influence of dynamic feature points in the tracking thread, save the static feature points and improve the accuracy of camera pose. Finally, simulation is carried out to verify. Experimental results on TUM data set show that compared with the ORB-SLAM2 system, this method can improve the tracking trajectory accuracy and map applicability.