Localization and mapping for autonomous vehicles using bearing-only measurements

Kwok, Ngai Ming

Localization and mapping for autonomous vehicles using bearing-only measurements

Kwok, Ngai Ming

Permalink

Publication Type:: Thesis
Issue Date:: 2006

Closed Access

	Filename	Description	Size
	01Front.pdf	contents and abstract	523.61 kB	Adobe PDF	View/Open
	02Whole.pdf	thesis	16.36 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Kwok, Ngai Ming
dc.date.accessioned	2016-11-09T01:01:23Z
dc.date.available	2016-11-09T01:01:23Z
dc.date.issued	2006
dc.identifier.uri	http://hdl.handle.net/10453/60588
dc.description	University of Technology, Sydney. Faculty of Engineering.	en_AU
dc.description	NO FULL TEXT AVAILABLE. Access is restricted indefinitely. The hardcopy may be available for consultation at the UTS Library.
dc.description.abstract	NO FULL TEXT AVAILABLE. Access is restricted indefinitely. ----- This thesis is devoted to the fast-emerging area of robotics research on the simultaneous localization and mapping (SLAM) for an autonomous vehicle using bearing-only measurements. Research outcomes from this work are the developments of novel techniques for bearing-only SLAM in indoor environments. Based on the Bayesian estimation philosophy, estimation techniques reported in this thesis include the particle filter, the Gaussian sum filter and the multi-hypothesis filter that address two essential criteria in SLAM, namely, performance and efficiency. In general, sensing devices used by the autonomous vehicle for localization and mapping purpose can be classified as range-and-bearing, range-only and bearing-only sensors. The major difficulty in using the bearing-only sensors is the landmark initialization problem in the stochastic localization and mapping framework. The lack of range measurements to landmarks leads to the need for computationally expensive filters; such filters require efficiency improvements before they can be practically exploited. Out of substantial contributions to robotics and automation research resulting directly or in relevance to this project, the most significant findings in this thesis are listed in the following. • A new technique, based on evolutionary computations, is proposed to mitigate sample impoverishment when applying the particle filter to solve the bearing-only SLAM problem. • A successful implementation of bearing-only SLAM using a computationally efficient Gaussian sum filter realized by minimizing the number of Gaussians to be maintained in the filter. • An efficient approach to the matching of image features for vision-based SLAM by using the multi-observation decision making technique. • The development of a path control strategy that enhances the bearing-only SLAM performance by choosing an optimal trajectory for the autonomous vehicle.	en_AU
dc.format	Thesis (PhD)
dc.language.iso	en_AU	en_AU
dc.rights	The author owns the copyright in this thesis including all reproduction and reuse rights for the work. The work may not be altered without the permission of the copyright owner. Attribution is essential when quoting or paraphrasing from this thesis.
dc.rights	info:eu-repo/semantics/closedAccess
dc.title	Localization and mapping for autonomous vehicles using bearing-only measurements	en_AU
dc.type	Thesis	en_AU
utslib.copyright.status	closed_access

Abstract:

NO FULL TEXT AVAILABLE. Access is restricted indefinitely. ----- This thesis is devoted to the fast-emerging area of robotics research on the simultaneous localization and mapping (SLAM) for an autonomous vehicle using bearing-only measurements. Research outcomes from this work are the developments of novel techniques for bearing-only SLAM in indoor environments. Based on the Bayesian estimation philosophy, estimation techniques reported in this thesis include the particle filter, the Gaussian sum filter and the multi-hypothesis filter that address two essential criteria in SLAM, namely, performance and efficiency. In general, sensing devices used by the autonomous vehicle for localization and mapping purpose can be classified as range-and-bearing, range-only and bearing-only sensors. The major difficulty in using the bearing-only sensors is the landmark initialization problem in the stochastic localization and mapping framework. The lack of range measurements to landmarks leads to the need for computationally expensive filters; such filters require efficiency improvements before they can be practically exploited. Out of substantial contributions to robotics and automation research resulting directly or in relevance to this project, the most significant findings in this thesis are listed in the following. • A new technique, based on evolutionary computations, is proposed to mitigate sample impoverishment when applying the particle filter to solve the bearing-only SLAM problem. • A successful implementation of bearing-only SLAM using a computationally efficient Gaussian sum filter realized by minimizing the number of Gaussians to be maintained in the filter. • An efficient approach to the matching of image features for vision-based SLAM by using the multi-observation decision making technique. • The development of a path control strategy that enhances the bearing-only SLAM performance by choosing an optimal trajectory for the autonomous vehicle.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/60588