A Ground Texture-based Mapping and Localization Method for AGVs

Javier Macias-Sola, Sarah Uttendorf, Jan Olaf Blech

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

2 Citations (Scopus)

Abstract

We present a ground-texture based mapping and localization method that we implement and evaluate on an industrial AGV. Our method compares recorded textures with known pictures from a database prior estimating the pose. This comparison is based on feature histograms. Keeping this structure allows us to perform a live evaluation of the quality of the pose estimates. This is done by leveraging the conditions of the presented use case in which all images are known to have a similar orientation and scale. Furthermore, we provide an analysis of how the floor features are affected after cleaning, as well as their spatial distribution.

Original languageEnglish
Title of host publication2021 International Conference on Indoor Positioning and Indoor Navigation, IPIN 2021
PublisherIEEE
Number of pages6
ISBN (Electronic)978-1-6654-0402-0
DOIs
Publication statusPublished - Jan 2022
MoE publication typeA4 Conference publication
EventInternational Conference on Indoor Positioning and Indoor Navigation - Lloret de Mar, Spain
Duration: 29 Nov 20212 Dec 2021

Publication series

NameInternational Conference on Indoor Positioning and Indoor Navigation
ISSN (Electronic)2471-917X

Conference

ConferenceInternational Conference on Indoor Positioning and Indoor Navigation
Abbreviated titleIPIN
Country/TerritorySpain
CityLloret de Mar
Period29/11/202102/12/2021

Keywords

  • AGV
  • industrial floor
  • localization
  • pose estimation

Fingerprint

Dive into the research topics of 'A Ground Texture-based Mapping and Localization Method for AGVs'. Together they form a unique fingerprint.

Cite this