Panoramic visual odometry for ground vehicle

Mingyang Gao, Meiling Wang, Hao Zhu

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

1 Citation (Scopus)

Abstract

Motion estimation from image sensors is becoming an important component of navigation systems. The idea of making a low-cost, compact and self-contained device on vision attracts many researchers. We present our recent results on visual odometry using a panoramic camera in our unmanned ground vehicle. We exploit scanline intensity profiles to estimate the rotation by comparing their absolute difference. We transform the panoramic images into orthographic top view images, and perform feature detection to extract lanes and other manmade markers, then utilize the markers to estimate the horizontal speed. With the rotation and speed data, we determine the trajectory of the vehicle, and synthesise an orthographic top view image of the road plane along the trajectory. Finally, we verify our visual odometry system by experiment.

Original languageEnglish
Title of host publicationProceedings of the 35th Chinese Control Conference, CCC 2016
EditorsJie Chen, Qianchuan Zhao, Jie Chen
PublisherIEEE Computer Society
Pages5479-5484
Number of pages6
ISBN (Electronic)9789881563910
DOIs
Publication statusPublished - 26 Aug 2016
Event35th Chinese Control Conference, CCC 2016 - Chengdu, China
Duration: 27 Jul 201629 Jul 2016

Publication series

NameChinese Control Conference, CCC
Volume2016-August
ISSN (Print)1934-1768
ISSN (Electronic)2161-2927

Conference

Conference35th Chinese Control Conference, CCC 2016
Country/TerritoryChina
CityChengdu
Period27/07/1629/07/16

Keywords

  • IPM transformation
  • Panoramic camera
  • Scanline intensity profile
  • Visual odometry

Fingerprint

Dive into the research topics of 'Panoramic visual odometry for ground vehicle'. Together they form a unique fingerprint.

Cite this