Publicação

Ground truth system for validation of indoor positioning systems

Ver documento

Detalhes bibliográficos
Resumo:This dissertation presents a solution, based on Visible Light Positioning, for the establishment of a ground truth in the testing environment of Indoor Positioning Systems. Firstly, in order to understand the requirements of current Indoor Positioning Systems, a study of their technologies and accuracy capabilities is presented. Subsequently, this document details the theory behind the proposed system, as well as its development steps. The proposed system is comprised of two cameras which can determine their own positions automatically, using ArUco markers as reference. The cameras have the capability of detecting a LED target through image processing, with one of the cameras transferring data to the other by means of a TCP socket. The target’s position in the 3D space is determined through a series of algorithms and mathematical principles. A thorough, simulation-based, characterization of uncertainty is presented, used to determine the impact of the main error sources encountered during development, on the precision and accuracy of the system. Experimental accuracy results, obtained from the laboratory setup, are also presented and compared to the simulation results. To conclude, a discussion about the obtained results and the validity of the system as a ground truth establishing system is given.
Autores principais:Vala, Diogo Alexandre Agostinho
Assunto:Ground truth Indoor positioning systems Camera Visible light positioning
Ano:2022
País:Portugal
Tipo de documento:dissertação de mestrado
Tipo de acesso:acesso aberto
Instituição associada:Universidade de Aveiro
Idioma:inglês
Origem:RIA - Repositório Institucional da Universidade de Aveiro
Descrição
Resumo:This dissertation presents a solution, based on Visible Light Positioning, for the establishment of a ground truth in the testing environment of Indoor Positioning Systems. Firstly, in order to understand the requirements of current Indoor Positioning Systems, a study of their technologies and accuracy capabilities is presented. Subsequently, this document details the theory behind the proposed system, as well as its development steps. The proposed system is comprised of two cameras which can determine their own positions automatically, using ArUco markers as reference. The cameras have the capability of detecting a LED target through image processing, with one of the cameras transferring data to the other by means of a TCP socket. The target’s position in the 3D space is determined through a series of algorithms and mathematical principles. A thorough, simulation-based, characterization of uncertainty is presented, used to determine the impact of the main error sources encountered during development, on the precision and accuracy of the system. Experimental accuracy results, obtained from the laboratory setup, are also presented and compared to the simulation results. To conclude, a discussion about the obtained results and the validity of the system as a ground truth establishing system is given.