Toward robust visual odometry using prior 2D map information and multiple hypothesis particle filtering

Edwards, S., Mihaylova, L. orcid.org/0000-0001-5856-2223, Aitken, J. orcid.org/0000-0003-4204-4020 et al. (1 more author) (2021) Toward robust visual odometry using prior 2D map information and multiple hypothesis particle filtering. In: Fox, C., Gao, J., Esfahani, G.H., Saaj, M., Hanheide, M. and Parsons, S., (eds.) Towards Autonomous Robotic Systems; 22nd Annual Conference, TAROS 2021, Lincoln, UK, September 8–10, 2021, Proceedings. Towards Autonomous Robotic Systems Conference (TAROS), 08-10 Sep 2021, Lincoln, UK (virtual conference). Lecture Notes in Computer Science (13054). Springer, Cham , pp. 188-192. ISBN 9783030891763

Abstract

Visual odometry can be used to estimate the pose of a robot from current and recent video frames. A problem with these methods is that they drift over time due to the accumulation of estimation errors at each time-step. In this short paper we propose and briefly demonstrate the potential benefit of using prior 2D, top-down map information combined with multiple hypothesis particle filtering to correct visual odometry estimates. The results demonstrate a substantial improvement in robustness and accuracy over the sole use of visual odometry.

Metadata

Authors/Creators:	Edwards, S. Mihaylova, L. https://orcid.org/0000-0001-5856-2223 Aitken, J. https://orcid.org/0000-0003-4204-4020 Anderson, S. https://orcid.org/0000-0002-7452-5681
Copyright, Publisher and Additional Information:	© 2021 Springer Nature Switzerland AG. This is an author-produced version of a paper subsequently published in TAROS 2021 Proceedings. Uploaded in accordance with the publisher's self-archiving policy.
Keywords:	Visual odometry; Deep learning; Multiple Hypothesis; Particle filter; Map prior
Dates:	Published (online): 31 October 2021 Published: 31 October 2021
Institution:	The University of Sheffield
Academic Units:	The University of Sheffield > Faculty of Engineering (Sheffield) > Department of Automatic Control and Systems Engineering (Sheffield)
Depositing User:	Symplectic Sheffield
Date Deposited:	06 Dec 2021 13:48
Last Modified:	06 Dec 2021 13:48
Status:	Published
Publisher:	Springer, Cham
Series Name:	Lecture Notes in Computer Science
Refereed:	Yes
Identification Number:	https://doi.org/10.1007/978-3-030-89177-0_19
Related URLs:	Conference

CORE (COnnecting REpositories)

Toward robust visual odometry using prior 2D map information and multiple hypothesis particle filtering

Abstract

Metadata

Download

Accepted Version

Export

Statistics