Performance improvement of iterative closest point-based outdoor SLAM by rotation invariant descriptors of salient regions

Yong Ju Lee, Jae-Bok Song, Ji Hoon Choi

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

To navigate in an unknown environment, a robot should build a model for the environment. For outdoor environments, a three-dimensional (3-D) map is usually used as a main model. This study considers outdoor simultaneous localization and mapping (SLAM) to build a global 3-D map by matching local 3-D maps. An iterative closest point (ICP) algorithm is used to match local 3-D maps and estimate a robot pose, but an alignment error is generated by the ICP algorithm due to the false selection of corresponding points. We propose a new method to extract 3-D points that are valid for ICP matching. Rotation-invariant descriptors are introduced for robust correspondence. 3-D environmental data acquired by tilting a 2-D laser scanner are used to build local 3-D maps. Experimental results in real environments show the increased accuracy of the ICP-based matching and a reduction in matching time.

Original languageEnglish
Pages (from-to)349-360
Number of pages12
JournalJournal of Intelligent and Robotic Systems: Theory and Applications
Volume71
Issue number3-4
DOIs
Publication statusPublished - 2013 Sep 1

Fingerprint

Robots
Lasers

Keywords

  • 3-D maps
  • Iterative closest point (ICP)
  • Mapping
  • Outdoor navigation
  • SLAM

ASJC Scopus subject areas

  • Artificial Intelligence
  • Software
  • Control and Systems Engineering
  • Electrical and Electronic Engineering
  • Industrial and Manufacturing Engineering
  • Mechanical Engineering

Cite this

@article{6c38aae744d84655afa6a4366271a424,
title = "Performance improvement of iterative closest point-based outdoor SLAM by rotation invariant descriptors of salient regions",
abstract = "To navigate in an unknown environment, a robot should build a model for the environment. For outdoor environments, a three-dimensional (3-D) map is usually used as a main model. This study considers outdoor simultaneous localization and mapping (SLAM) to build a global 3-D map by matching local 3-D maps. An iterative closest point (ICP) algorithm is used to match local 3-D maps and estimate a robot pose, but an alignment error is generated by the ICP algorithm due to the false selection of corresponding points. We propose a new method to extract 3-D points that are valid for ICP matching. Rotation-invariant descriptors are introduced for robust correspondence. 3-D environmental data acquired by tilting a 2-D laser scanner are used to build local 3-D maps. Experimental results in real environments show the increased accuracy of the ICP-based matching and a reduction in matching time.",
keywords = "3-D maps, Iterative closest point (ICP), Mapping, Outdoor navigation, SLAM",
author = "Lee, {Yong Ju} and Jae-Bok Song and Choi, {Ji Hoon}",
year = "2013",
month = "9",
day = "1",
doi = "10.1007/s10846-012-9786-2",
language = "English",
volume = "71",
pages = "349--360",
journal = "Journal of Intelligent and Robotic Systems: Theory and Applications",
issn = "0921-0296",
publisher = "Springer Netherlands",
number = "3-4",

}

TY - JOUR

T1 - Performance improvement of iterative closest point-based outdoor SLAM by rotation invariant descriptors of salient regions

AU - Lee, Yong Ju

AU - Song, Jae-Bok

AU - Choi, Ji Hoon

PY - 2013/9/1

Y1 - 2013/9/1

N2 - To navigate in an unknown environment, a robot should build a model for the environment. For outdoor environments, a three-dimensional (3-D) map is usually used as a main model. This study considers outdoor simultaneous localization and mapping (SLAM) to build a global 3-D map by matching local 3-D maps. An iterative closest point (ICP) algorithm is used to match local 3-D maps and estimate a robot pose, but an alignment error is generated by the ICP algorithm due to the false selection of corresponding points. We propose a new method to extract 3-D points that are valid for ICP matching. Rotation-invariant descriptors are introduced for robust correspondence. 3-D environmental data acquired by tilting a 2-D laser scanner are used to build local 3-D maps. Experimental results in real environments show the increased accuracy of the ICP-based matching and a reduction in matching time.

AB - To navigate in an unknown environment, a robot should build a model for the environment. For outdoor environments, a three-dimensional (3-D) map is usually used as a main model. This study considers outdoor simultaneous localization and mapping (SLAM) to build a global 3-D map by matching local 3-D maps. An iterative closest point (ICP) algorithm is used to match local 3-D maps and estimate a robot pose, but an alignment error is generated by the ICP algorithm due to the false selection of corresponding points. We propose a new method to extract 3-D points that are valid for ICP matching. Rotation-invariant descriptors are introduced for robust correspondence. 3-D environmental data acquired by tilting a 2-D laser scanner are used to build local 3-D maps. Experimental results in real environments show the increased accuracy of the ICP-based matching and a reduction in matching time.

KW - 3-D maps

KW - Iterative closest point (ICP)

KW - Mapping

KW - Outdoor navigation

KW - SLAM

UR - http://www.scopus.com/inward/record.url?scp=84883134659&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84883134659&partnerID=8YFLogxK

U2 - 10.1007/s10846-012-9786-2

DO - 10.1007/s10846-012-9786-2

M3 - Article

VL - 71

SP - 349

EP - 360

JO - Journal of Intelligent and Robotic Systems: Theory and Applications

JF - Journal of Intelligent and Robotic Systems: Theory and Applications

SN - 0921-0296

IS - 3-4

ER -