The invariance of state estimation for robot navigation

Zhang, Teng

The invariance of state estimation for robot navigation

Zhang, Teng

Permalink

Publication Type:: Thesis
Issue Date:: 2018

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

Adobe PDF

Download contents and abstractAdobe PDF (216.89 kB)

Adobe PDF

Download thesisAdobe PDF (3.33 MB)

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Zhang, Teng
dc.date.accessioned	2018-04-09T01:39:59Z
dc.date.available	2018-04-09T01:39:59Z
dc.date.issued	2018
dc.identifier.uri	http://hdl.handle.net/10453/123279
dc.description	University of Technology Sydney. Faculty of Engineering and Information Technology.	en_AU
dc.description.abstract	We are living in an era that are being changed by mobile robots such as unmanned aerial vehicles and self-driving cars. State estimation for navigation is one of the fundamental problems in mobile robot's applications. This work entirely focuses on two problems of state estimation for robot navigation, i.e., simultaneous localization and mapping (SLAM), and visual-inertial navigation systems (VINS). The SLAM problem asks whether it is possible for a robot to build a map of an unknown environment and simultaneously work out its own location within the map. The VINS problem aims at the estimates of a robot's pose and velocity by the on-board sensor fusion of a camera and an inertial measurement unit (IMU). After data association, both SLAM and VINS need a back-end solver to estimate the state of a robot and the environment, which mainly includes two methods: extended Kalman filter and optimization. The main contribution of this thesis is the invariance theory, which proposes the basic principles for state estimation, i.e., the actual estimates should be invariant under unobservable (deterministic or stochastic) transformations. The invariance theory does not only provide compact, elegant, profound explanations and insights of extended Kalman filter and optimization in SLAM and VINS, but also help to design new algorithms to improve the existing methods.	en_AU
dc.format	Thesis (PhD)
dc.language.iso	en_AU	en_AU
dc.relation	https://opus.lib.uts.edu.au/bitstream/10453/123279/7/02whole.pdf
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	au.edu.uts.lib/ppc
dc.rights	info:eu-repo/semantics/openAccess
dc.title	The invariance of state estimation for robot navigation	en_AU
dc.type	Thesis	en_AU
utslib.copyright.status	open_access

Abstract:

We are living in an era that are being changed by mobile robots such as unmanned aerial vehicles and self-driving cars. State estimation for navigation is one of the fundamental problems in mobile robot's applications. This work entirely focuses on two problems of state estimation for robot navigation, i.e., simultaneous localization and mapping (SLAM), and visual-inertial navigation systems (VINS). The SLAM problem asks whether it is possible for a robot to build a map of an unknown environment and simultaneously work out its own location within the map. The VINS problem aims at the estimates of a robot's pose and velocity by the on-board sensor fusion of a camera and an inertial measurement unit (IMU). After data association, both SLAM and VINS need a back-end solver to estimate the state of a robot and the environment, which mainly includes two methods: extended Kalman filter and optimization. The main contribution of this thesis is the invariance theory, which proposes the basic principles for state estimation, i.e., the actual estimates should be invariant under unobservable (deterministic or stochastic) transformations. The invariance theory does not only provide compact, elegant, profound explanations and insights of extended Kalman filter and optimization in SLAM and VINS, but also help to design new algorithms to improve the existing methods.

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

http://hdl.handle.net/10453/123279