Use this URL to cite or link to this record in EThOS:
Title: Investigation of context determination for advanced navigation using smartphone sensors
Author: Gao, Han
ISNI:       0000 0004 7660 5866
Awarding Body: UCL (University College London)
Current Institution: University College London (University of London)
Date of Award: 2019
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Navigation and positioning is inherently dependent on the context, which comprises both the operating environment and the behaviour of the host vehicle or user. The environment determines the type and quality of radio signals available for positioning, while the behaviour can contribute additional information to the navigation solution. Although many navigation and positioning techniques have been developed, no single one is capable of providing reliable and accurate positioning in all contexts. Therefore, it is necessary for a navigation system to be able to operate across different types of contexts. Context adaptive navigation offers a solution to this problem by detecting the operating contexts and adopting different positioning techniques accordingly. This study focuses on context determination with the available sensors on smartphone, through framework design, behavioural and environmental context detection, context association, comprehensive experimental tests, and system demonstration, building the foundation for a context-adaptive navigation system. In this thesis, the overall framework of context determination is first designed. Following the framework, the behavioural contexts, covering different human activities and vehicle motions, are recognised by different machine learning classifiers in hierarchy. Their classification results are further enhanced by feature selection and a connectivity dependent filter. Environmental contexts are detected from GNSS measurements. Indoor and outdoor environments are first distinguished based on the availability and strength of GNSS signals using a hidden Markov model based method. Within the model, the different levels of connections between environments are exploited as well. Then a fuzzy inference system is designed to enable the further classification of outdoor environments into urban and open-sky. As behaviours and environments are not completely independent, this study also considers context association, investigating how behaviours can be associated within environment detection. Tests in a series of multi-context scenarios have shown that the association mechanism can further improve the reliability of context detection. Finally, the proposed context determination system has been demonstrated in daily scenarios.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available