Towards seamless pedestrian navigation

The best methods for processing a positioning solution vary between the positioning technologies. The main methods are examined and the best are chosen and used in the final implementation. The path-loss modelling is the faster method when using BLE. It is slightly more advantageous, although slight...

Full description

Bibliographic Details
Main Author: Peltola, Pekka
Published: University of Nottingham 2018
Subjects:
Online Access:https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.748485
id ndltd-bl.uk-oai-ethos.bl.uk-748485
record_format oai_dc
spelling ndltd-bl.uk-oai-ethos.bl.uk-7484852019-01-08T03:34:39ZTowards seamless pedestrian navigationPeltola, Pekka2018The best methods for processing a positioning solution vary between the positioning technologies. The main methods are examined and the best are chosen and used in the final implementation. The path-loss modelling is the faster method when using BLE. It is slightly more advantageous, although slightly less accurate, when compared with the very laborious Fingerprinting alternative. For an UWB system, the accurate timing method is the better option over the signal strength measures. These sensor subsystems produce the positioning solutions that are of different quality. The fusion filter is then the motor, where the information fusion of these subsystems happens. For the fusion, multiple methods exist. Accuracy and computational efficiency are the driving factors in this process of continuous development of an ultimate seamless navigation system. A truly seamless navigation system uses the best available sensors and methods in the derivation of the most accurate positioning solution. This acquisition relies on the context inference process. The system adapts to all contexts throughout the navigation process. The developed novel seamless positioning system in this thesis was implemented successfully and was proven to operate correctly. The accuracy of the implemented system reaches a level below 4 metres (95%), outdoors. Indoors, the accuracy reaches a level below 2 metres. The availability, the deployment and the existing infrastructure are an issue to be tackled in the future. The thesis work was conducted as a part of the Marie Curie Initial Training Network project, named Multi-Pos. This project consisted of 15 researchers at different locations in Europe. Part of the thesis is a result of the cooperation between the project partners. Especially the parts of the work done on context detection and the literature review were completed with common effort.G Geography (General)University of Nottinghamhttps://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.748485http://eprints.nottingham.ac.uk/51886/Electronic Thesis or Dissertation
collection NDLTD
sources NDLTD
topic G Geography (General)
spellingShingle G Geography (General)
Peltola, Pekka
Towards seamless pedestrian navigation
description The best methods for processing a positioning solution vary between the positioning technologies. The main methods are examined and the best are chosen and used in the final implementation. The path-loss modelling is the faster method when using BLE. It is slightly more advantageous, although slightly less accurate, when compared with the very laborious Fingerprinting alternative. For an UWB system, the accurate timing method is the better option over the signal strength measures. These sensor subsystems produce the positioning solutions that are of different quality. The fusion filter is then the motor, where the information fusion of these subsystems happens. For the fusion, multiple methods exist. Accuracy and computational efficiency are the driving factors in this process of continuous development of an ultimate seamless navigation system. A truly seamless navigation system uses the best available sensors and methods in the derivation of the most accurate positioning solution. This acquisition relies on the context inference process. The system adapts to all contexts throughout the navigation process. The developed novel seamless positioning system in this thesis was implemented successfully and was proven to operate correctly. The accuracy of the implemented system reaches a level below 4 metres (95%), outdoors. Indoors, the accuracy reaches a level below 2 metres. The availability, the deployment and the existing infrastructure are an issue to be tackled in the future. The thesis work was conducted as a part of the Marie Curie Initial Training Network project, named Multi-Pos. This project consisted of 15 researchers at different locations in Europe. Part of the thesis is a result of the cooperation between the project partners. Especially the parts of the work done on context detection and the literature review were completed with common effort.
author Peltola, Pekka
author_facet Peltola, Pekka
author_sort Peltola, Pekka
title Towards seamless pedestrian navigation
title_short Towards seamless pedestrian navigation
title_full Towards seamless pedestrian navigation
title_fullStr Towards seamless pedestrian navigation
title_full_unstemmed Towards seamless pedestrian navigation
title_sort towards seamless pedestrian navigation
publisher University of Nottingham
publishDate 2018
url https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.748485
work_keys_str_mv AT peltolapekka towardsseamlesspedestriannavigation
_version_ 1718808762709442560