Memory-efficient real-time map building using octree of planes and points

Yonghyun Jo, Hanyoung Jang, Yeon Ho Kim, Joon Kee Cho, Hadi Moradi, Junghyun Han

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

This paper presents an octree-based map building algorithm for mobile home-service robots. The robot is equipped with a time-of-flight camera, which produces point clouds of the environment surfaces. Given the successive input of point clouds, a 3D map is incrementally computed in real time. The map is accurate and memory-efficient because the octree nodes containing points on a plane are merged and represented simply by an index to the plane. The real-time performance is achieved largely due to the parallel processing capability of the many-core Graphics Processing Unit used for plane extraction.

Original languageEnglish
Pages (from-to)301-308
Number of pages8
JournalAdvanced Robotics
Volume27
Issue number4
DOIs
Publication statusPublished - 2013 Mar 1

Fingerprint

Data storage equipment
Mobile homes
Robots
Cameras
Processing
Graphics processing unit

Keywords

  • GPU
  • home-service robot
  • map building
  • octree

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Human-Computer Interaction
  • Computer Science Applications
  • Hardware and Architecture
  • Software

Cite this

Memory-efficient real-time map building using octree of planes and points. / Jo, Yonghyun; Jang, Hanyoung; Kim, Yeon Ho; Cho, Joon Kee; Moradi, Hadi; Han, Junghyun.

In: Advanced Robotics, Vol. 27, No. 4, 01.03.2013, p. 301-308.

Research output: Contribution to journalArticle

Jo, Yonghyun ; Jang, Hanyoung ; Kim, Yeon Ho ; Cho, Joon Kee ; Moradi, Hadi ; Han, Junghyun. / Memory-efficient real-time map building using octree of planes and points. In: Advanced Robotics. 2013 ; Vol. 27, No. 4. pp. 301-308.
@article{16bb250c10bc46819bd2547a614274b0,
title = "Memory-efficient real-time map building using octree of planes and points",
abstract = "This paper presents an octree-based map building algorithm for mobile home-service robots. The robot is equipped with a time-of-flight camera, which produces point clouds of the environment surfaces. Given the successive input of point clouds, a 3D map is incrementally computed in real time. The map is accurate and memory-efficient because the octree nodes containing points on a plane are merged and represented simply by an index to the plane. The real-time performance is achieved largely due to the parallel processing capability of the many-core Graphics Processing Unit used for plane extraction.",
keywords = "GPU, home-service robot, map building, octree",
author = "Yonghyun Jo and Hanyoung Jang and Kim, {Yeon Ho} and Cho, {Joon Kee} and Hadi Moradi and Junghyun Han",
year = "2013",
month = "3",
day = "1",
doi = "10.1080/01691864.2013.755248",
language = "English",
volume = "27",
pages = "301--308",
journal = "Advanced Robotics",
issn = "0169-1864",
publisher = "Taylor and Francis Ltd.",
number = "4",

}

TY - JOUR

T1 - Memory-efficient real-time map building using octree of planes and points

AU - Jo, Yonghyun

AU - Jang, Hanyoung

AU - Kim, Yeon Ho

AU - Cho, Joon Kee

AU - Moradi, Hadi

AU - Han, Junghyun

PY - 2013/3/1

Y1 - 2013/3/1

N2 - This paper presents an octree-based map building algorithm for mobile home-service robots. The robot is equipped with a time-of-flight camera, which produces point clouds of the environment surfaces. Given the successive input of point clouds, a 3D map is incrementally computed in real time. The map is accurate and memory-efficient because the octree nodes containing points on a plane are merged and represented simply by an index to the plane. The real-time performance is achieved largely due to the parallel processing capability of the many-core Graphics Processing Unit used for plane extraction.

AB - This paper presents an octree-based map building algorithm for mobile home-service robots. The robot is equipped with a time-of-flight camera, which produces point clouds of the environment surfaces. Given the successive input of point clouds, a 3D map is incrementally computed in real time. The map is accurate and memory-efficient because the octree nodes containing points on a plane are merged and represented simply by an index to the plane. The real-time performance is achieved largely due to the parallel processing capability of the many-core Graphics Processing Unit used for plane extraction.

KW - GPU

KW - home-service robot

KW - map building

KW - octree

UR - http://www.scopus.com/inward/record.url?scp=84876192479&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84876192479&partnerID=8YFLogxK

U2 - 10.1080/01691864.2013.755248

DO - 10.1080/01691864.2013.755248

M3 - Article

AN - SCOPUS:84876192479

VL - 27

SP - 301

EP - 308

JO - Advanced Robotics

JF - Advanced Robotics

SN - 0169-1864

IS - 4

ER -