elib
DLR-Header
DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

Stereo vision-based simultaneous localization and mapping with ranging aid

Lee, Young-Hee and Zhu, Chen and Giorgi, Gabriele and Günther, Christoph (2018) Stereo vision-based simultaneous localization and mapping with ranging aid. In: Proceedings of the IEEE/ION Position, Location and Navigation Symposium, PLANS 2018. IEEE/ION Position, Location and Navigation Symposium (PLANS), 2018, Monterey, California, USA. ISBN 978-153861647-5

[img] PDF
866kB

Abstract

We propose fusion of stereo visual odometry and ranging for Simultaneous Localization And Mapping (SLAM). Two basic processes of feature- and keyframe-based stereo visual odometry (Tracking and Local mapping) are operated in the algorithm, while saving all local keyframes, map points, and visual constraints in a global map database, as well as the available ranging constraints of the keyframes. At the end of the sequence, state estimates by visual odometry are fused with ranging measurements to mitigate the inherent accumulating errors in the process and achieve global consistency. We formulate a simple graphical representation for the fusion, and perform least squares estimation with the sparse Levenberg-Marquardt algorithm to minimize the summation of the re-projection and distance squared errors over all the defined constraints in the global graph. The proposed algorithm is evaluated both qualitatively and quantitatively on a real stereo image dataset with synthetically generated distance measurements with superimposed Gaussian white noise. The experimental results show that the proposed SLAM algorithm effectively compensates the cumulative bias in visual odometry. Furthermore, the global accuracy of the trajectory estimation is comparable to the one of stereo vision-only SLAM with closing loops.

Item URL in elib:https://elib.dlr.de/123287/
Document Type:Conference or Workshop Item (Lecture)
Title:Stereo vision-based simultaneous localization and mapping with ranging aid
Authors:
AuthorsInstitution or Email of AuthorsAuthors ORCID iD
Lee, Young-Heeyounghee.lee (at) tum.deUNSPECIFIED
Zhu, ChenChen.Zhu (at) dlr.dehttps://orcid.org/0000-0002-4320-4826
Giorgi, GabrieleGabriele.Giorgi (at) dlr.deUNSPECIFIED
Günther, Christophchristoph.guenther (at) dlr.deUNSPECIFIED
Date:23 November 2018
Journal or Publication Title:Proceedings of the IEEE/ION Position, Location and Navigation Symposium, PLANS 2018
Refereed publication:Yes
Open Access:Yes
Gold Open Access:No
In SCOPUS:Yes
In ISI Web of Science:Yes
ISBN:978-153861647-5
Status:Published
Keywords:stereo vision, sensor fusion, ranging sensor, graph-based SLAM
Event Title:IEEE/ION Position, Location and Navigation Symposium (PLANS), 2018
Event Location:Monterey, California, USA
Event Type:international Conference
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Space
HGF - Program Themes:Communication and Navigation
DLR - Research area:Raumfahrt
DLR - Program:R KN - Kommunikation und Navigation
DLR - Research theme (Project):R - Project Navigation 4.0
Location: Oberpfaffenhofen
Institutes and Institutions:Institute of Communication and Navigation > Navigation
Deposited By: Zhu, Chen
Deposited On:23 Nov 2018 15:13
Last Modified:31 Jul 2019 20:21

Repository Staff Only: item control page

Browse
Search
Help & Contact
Information
electronic library is running on EPrints 3.3.12
Copyright © 2008-2017 German Aerospace Center (DLR). All rights reserved.