Research on SLAM System Based on Binocular Vision and IMU Information

Xiao Luo, Baoling Han*, Qingsheng Luo, Xinliang Zhong

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Simultaneous localization and mapping (SLAM) is one of the key technologies in the field of robotics, is the key to autonomous navigation of mobile robots, and is also the core and foundation of autonomous and intelligent mobile robots. The SLAM method which relied information solely on monocular cameras is too dependent on the characteristic information of the surrounding environment. For the lack of the scene texture, the dramatic changes of the illumination and the poor performance of the dynamic scene, and the low frame rate of the visual sensor, it cannot deal with the situation of fast motion. The inertial measurement unit (IMU) can output the acceleration and angular velocity of the sensor itself at a high frame rate, and is not affected by the environment, but with serious drift. To solve this problem, this paper aims to design a SLAM system which combines binocular vision information and IMU information. It can realize robust and precise positioning in unknown environment, and provide corresponding navigation map for navigation.

Original languageEnglish
Title of host publicationICGG 2020 - Proceedings of the 19th International Conference on Geometry and Graphics
EditorsLiang-Yee Cheng
PublisherSpringer Science and Business Media Deutschland GmbH
Pages557-567
Number of pages11
ISBN (Print)9783030634025
DOIs
Publication statusPublished - 2021
Event19th International Conference on Geometry and Graphics, ICGG 2020 - São Paulo, Brazil
Duration: 18 Jan 202122 Jan 2021

Publication series

NameAdvances in Intelligent Systems and Computing
Volume1296
ISSN (Print)2194-5357
ISSN (Electronic)2194-5365

Conference

Conference19th International Conference on Geometry and Graphics, ICGG 2020
Country/TerritoryBrazil
CitySão Paulo
Period18/01/2122/01/21

Keywords

  • Location
  • Loop detection
  • Nonlinear optimization
  • Visual SLAM

Fingerprint

Dive into the research topics of 'Research on SLAM System Based on Binocular Vision and IMU Information'. Together they form a unique fingerprint.

Cite this