Using MATLAB as auxiliary tool in visual localization for mobile robots

Shen Ho Liu, Yin Tien Wang, Chen Chien Hsu, Wei Yen Wang, Chiang Heng Chien

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

A feature-based visual robot localization algorithm is generally responsible for estimating the robot pose based on features from images using cameras. To address the issues of estimation accuracy and required computational time, an improved visual odometry (VO) system is proposed using direct approach of perspective-3-point (P3P) algorithm as well as map management method. Furthermore, a GUI platform is designed using MATLAB so that performances of the proposed VO system can be easily observed by users. In order to validate the reliability of the proposed system, various experiments are conducted using a binocular camera, where experimental results show that the proposed VO system outperforms conventional algorithms in terms of estimation accuracy and runtime efficiency.

Original languageEnglish
Title of host publication2018 International Symposium on Consumer Technologies, ISCT 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages24-25
Number of pages2
ISBN (Electronic)9781538646151
DOIs
Publication statusPublished - 2018 Jul 9
Event2018 International Symposium on Consumer Technologies, ISCT 2018 - St. Petersburg, Russian Federation
Duration: 2018 May 112018 May 12

Publication series

Name2018 International Symposium on Consumer Technologies, ISCT 2018

Conference

Conference2018 International Symposium on Consumer Technologies, ISCT 2018
Country/TerritoryRussian Federation
CitySt. Petersburg
Period2018/05/112018/05/12

Keywords

  • MATLAB
  • Perspective-3-point
  • Visual Odometry

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Computer Vision and Pattern Recognition
  • Engineering (miscellaneous)
  • Computer Science Applications

Fingerprint

Dive into the research topics of 'Using MATLAB as auxiliary tool in visual localization for mobile robots'. Together they form a unique fingerprint.

Cite this