Consistency analysis and improvement for single-camera localization

Joel A. Hesch, Stergios Roumeliotis

Research output: Chapter in Book/Report/Conference proceedingConference contribution

9 Scopus citations

Abstract

In this paper, we study the problem of estimator inconsistency in single-camera simultaneous localization and mapping (MonoSLAM) from a standpoint of system observability. Specifically, we postulate that a leading cause of inconsistency is the gain of spurious information along unobservable directions, resulting in smaller uncertainties, larger estimation errors, and divergence. Moreover, we introduce an Observability-Constrained MonoSLAM (OC-MonoSLAM) approach, which explicitly enforces the unobservable directions of the system, hence preventing spurious information gain and reducing inconsistency. Our analysis, along with the proposed method for reducing inconsistency, are validated with simulation trials and real-world experimentation.

Original languageEnglish (US)
Title of host publication2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2012
Pages15-22
Number of pages8
DOIs
StatePublished - Aug 20 2012
Event2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2012 - Providence, RI, United States
Duration: Jun 16 2012Jun 21 2012

Publication series

NameIEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
ISSN (Print)2160-7508
ISSN (Electronic)2160-7516

Other

Other2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2012
CountryUnited States
CityProvidence, RI
Period6/16/126/21/12

Fingerprint Dive into the research topics of 'Consistency analysis and improvement for single-camera localization'. Together they form a unique fingerprint.

Cite this