Indoor Global Localization Using Depth-Guided Photometric Edge Descriptor for Mobile Robot Navigation

Howon Cheong, Euntai Kim, Sung Kee Park

Research output: Contribution to journalArticle

Abstract

This paper suggests a new landmark descriptor for indoor mobile robot navigation with sensor fusion and a global localization method using it. In previous research on robot pose estimation, various landmarks such as geometric features, visual local-invariant features, or objects are utilized. However, in real-world situations, there is a possibility that distinctive landmarks are insufficient or there are many similar landmarks repeated in indoor environment, which makes accurate pose estimation difficult. In this work, we suggest a new landmark descriptor, called depth-guided photometric edge descriptor (DPED), which is composed of superpixels and approximated 3D depth information of photometric vertical edge. With this descriptor, we propose a global localization method based on coarse-to-fine strategy. In the coarse step, candidate nodes are found by place recognition using our pairwise constraint-based spectral matching technique, and the robot pose is estimated with a probabilistic scan matching in the fine step. The experimental results show that our method successfully estimates the robot pose in the real-world tests even when there is a lack of distinctive features and objects.

Original languageEnglish
Article number8782605
Pages (from-to)10837-10847
Number of pages11
JournalIEEE Sensors Journal
Volume19
Issue number22
DOIs
Publication statusPublished - 2019 Nov 15

Fingerprint

landmarks
navigation
robots
Mobile robots
Navigation
Robots
Fusion reactions
multisensor fusion
Sensors
estimates

All Science Journal Classification (ASJC) codes

  • Instrumentation
  • Electrical and Electronic Engineering

Cite this

@article{d81d5924a5a24a608f1e90678d4ab0b4,
title = "Indoor Global Localization Using Depth-Guided Photometric Edge Descriptor for Mobile Robot Navigation",
abstract = "This paper suggests a new landmark descriptor for indoor mobile robot navigation with sensor fusion and a global localization method using it. In previous research on robot pose estimation, various landmarks such as geometric features, visual local-invariant features, or objects are utilized. However, in real-world situations, there is a possibility that distinctive landmarks are insufficient or there are many similar landmarks repeated in indoor environment, which makes accurate pose estimation difficult. In this work, we suggest a new landmark descriptor, called depth-guided photometric edge descriptor (DPED), which is composed of superpixels and approximated 3D depth information of photometric vertical edge. With this descriptor, we propose a global localization method based on coarse-to-fine strategy. In the coarse step, candidate nodes are found by place recognition using our pairwise constraint-based spectral matching technique, and the robot pose is estimated with a probabilistic scan matching in the fine step. The experimental results show that our method successfully estimates the robot pose in the real-world tests even when there is a lack of distinctive features and objects.",
author = "Howon Cheong and Euntai Kim and Park, {Sung Kee}",
year = "2019",
month = "11",
day = "15",
doi = "10.1109/JSEN.2019.2932131",
language = "English",
volume = "19",
pages = "10837--10847",
journal = "IEEE Sensors Journal",
issn = "1530-437X",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "22",

}

Indoor Global Localization Using Depth-Guided Photometric Edge Descriptor for Mobile Robot Navigation. / Cheong, Howon; Kim, Euntai; Park, Sung Kee.

In: IEEE Sensors Journal, Vol. 19, No. 22, 8782605, 15.11.2019, p. 10837-10847.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Indoor Global Localization Using Depth-Guided Photometric Edge Descriptor for Mobile Robot Navigation

AU - Cheong, Howon

AU - Kim, Euntai

AU - Park, Sung Kee

PY - 2019/11/15

Y1 - 2019/11/15

N2 - This paper suggests a new landmark descriptor for indoor mobile robot navigation with sensor fusion and a global localization method using it. In previous research on robot pose estimation, various landmarks such as geometric features, visual local-invariant features, or objects are utilized. However, in real-world situations, there is a possibility that distinctive landmarks are insufficient or there are many similar landmarks repeated in indoor environment, which makes accurate pose estimation difficult. In this work, we suggest a new landmark descriptor, called depth-guided photometric edge descriptor (DPED), which is composed of superpixels and approximated 3D depth information of photometric vertical edge. With this descriptor, we propose a global localization method based on coarse-to-fine strategy. In the coarse step, candidate nodes are found by place recognition using our pairwise constraint-based spectral matching technique, and the robot pose is estimated with a probabilistic scan matching in the fine step. The experimental results show that our method successfully estimates the robot pose in the real-world tests even when there is a lack of distinctive features and objects.

AB - This paper suggests a new landmark descriptor for indoor mobile robot navigation with sensor fusion and a global localization method using it. In previous research on robot pose estimation, various landmarks such as geometric features, visual local-invariant features, or objects are utilized. However, in real-world situations, there is a possibility that distinctive landmarks are insufficient or there are many similar landmarks repeated in indoor environment, which makes accurate pose estimation difficult. In this work, we suggest a new landmark descriptor, called depth-guided photometric edge descriptor (DPED), which is composed of superpixels and approximated 3D depth information of photometric vertical edge. With this descriptor, we propose a global localization method based on coarse-to-fine strategy. In the coarse step, candidate nodes are found by place recognition using our pairwise constraint-based spectral matching technique, and the robot pose is estimated with a probabilistic scan matching in the fine step. The experimental results show that our method successfully estimates the robot pose in the real-world tests even when there is a lack of distinctive features and objects.

UR - http://www.scopus.com/inward/record.url?scp=85073880260&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85073880260&partnerID=8YFLogxK

U2 - 10.1109/JSEN.2019.2932131

DO - 10.1109/JSEN.2019.2932131

M3 - Article

AN - SCOPUS:85073880260

VL - 19

SP - 10837

EP - 10847

JO - IEEE Sensors Journal

JF - IEEE Sensors Journal

SN - 1530-437X

IS - 22

M1 - 8782605

ER -