Deep Learning Approach for UAV Visual Electrical Assets Inspection

Detalhes bibliográficos
Autor(a) principal: Barbosa, Joel Adão Pacheco
Data de Publicação: 2020
Tipo de documento: Dissertação
Idioma: eng
Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
Texto Completo: http://hdl.handle.net/10400.22/16898
Resumo: The growth in the electrical demand by most countries around the world requires bigger and more complex energy systems, which leads to the requirement of having even more monitoring, inspection and maintenance of those systems. To respond to this need, inspection methods based on Unmanned Aerial Vehicles (UAV) have emerged which, when equipped with the appropriate sensors, allow a greater reduction of costs and risks and an increase in efficiency and effectiveness compared to traditional methods, such as inspection with foot patrols or helicopter-assisted. To make the inspection process more autonomous and reliable, most of the methods apply visual detection methods that use highly complex Deep Learning based algorithms and that require a very large computational power. This dissertation intends to present a system for inspection of electrical assets, able to be integrated onboard the UAV, based on Deep Learning, which allows to collect visual samples grouped and aggregated for each electrical asset detected. To this end, a perception system capable of detecting electrical insulators or structures, such as poles or transmission towers, was developed, using the Movidius Neural Compute Stick portable platform that is capable of processing lightweight object detection Convolutional Neural Networks, allowing a modular, low-cost system that meets real-time processing requirements. In addition to this perception system, an electrical asset monitoring system has been implemented that allows tracking and mapping each asset throughout the inspection process, based on the previous system’s detections and a UAV navigation system. Finally, an autonomous inspection system is proposed, which consists of a set of trajectories that allow an efficient application of the monitoring system along a power line, through the mapping of structures and the gathering of insulator samples around that structure.
id RCAP_7713ac03402889a84187bc667ea77ad0
oai_identifier_str oai:recipp.ipp.pt:10400.22/16898
network_acronym_str RCAP
network_name_str Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
repository_id_str 7160
spelling Deep Learning Approach for UAV Visual Electrical Assets InspectionElectrical Assets InspectionDeep LearningObject DetectionMulti-Object TrackingUAVInspeção de Ativos ElétricosDeteção de ObjetosSeguimento de Múltiplos ObjetosVANTThe growth in the electrical demand by most countries around the world requires bigger and more complex energy systems, which leads to the requirement of having even more monitoring, inspection and maintenance of those systems. To respond to this need, inspection methods based on Unmanned Aerial Vehicles (UAV) have emerged which, when equipped with the appropriate sensors, allow a greater reduction of costs and risks and an increase in efficiency and effectiveness compared to traditional methods, such as inspection with foot patrols or helicopter-assisted. To make the inspection process more autonomous and reliable, most of the methods apply visual detection methods that use highly complex Deep Learning based algorithms and that require a very large computational power. This dissertation intends to present a system for inspection of electrical assets, able to be integrated onboard the UAV, based on Deep Learning, which allows to collect visual samples grouped and aggregated for each electrical asset detected. To this end, a perception system capable of detecting electrical insulators or structures, such as poles or transmission towers, was developed, using the Movidius Neural Compute Stick portable platform that is capable of processing lightweight object detection Convolutional Neural Networks, allowing a modular, low-cost system that meets real-time processing requirements. In addition to this perception system, an electrical asset monitoring system has been implemented that allows tracking and mapping each asset throughout the inspection process, based on the previous system’s detections and a UAV navigation system. Finally, an autonomous inspection system is proposed, which consists of a set of trajectories that allow an efficient application of the monitoring system along a power line, through the mapping of structures and the gathering of insulator samples around that structure.O grande crescimento da exigência elétrica pela maioria dos países por todo o mundo, requer que os sistemas de energia sejam maiores e mais complexos, o que conduz a uma maior necessidade de monitorização, inspeção e manutenção desses sistemas. Para responder a esta necessidade, surgiram métodos de inspeção baseados em Veículos Aéreos Não Tripulados (VANT) que, quando equipados com os sensores apropriados, permitem uma maior redução de custos e riscos e um grande aumento de eficiência e eficácia em comparação com os métodos tradicionais, como a inspeção com patrulhas pedonais ou assistida por helicóptero. Para tornar processo de inspeção mais autónomo e confiável, a maioria dos métodos realiza método de deteção visuais que utilizam algoritmos baseados em Deep Learning de elevada complexidade e que requerem um poder computacional muito grande. Nesta dissertação pretende-se apresentar um sistema de inspeção de ativos elétricos, para integração em VANTs, baseado em Apredizagem Profunda, que permite recolher amostras visuais agrupadas e agregadas por cada ativo elétrico detetado. Para tal foi desenvolvido um sistema de perceção capaz de detetar isoladores elétricos ou estruturas, como postes ou torres de transmissão, com recurso `a plataforma portátil Movidius Neural Compute Stick que ´e capaz de processar Redes Neuronais Convolucionais leves de deteção de objetos, permitindo assim um sistema modular, de baixo custo e que cumpre requisitos de processamento em tempo real. Para além deste sistema de perceção, foi implementado um sistema de monitorização de ativos elétricos que permite seguir e mapear cada ativo ao longo do processo de inspeção, com base nas deteções do sistema anterior e no sistema de navegação do VANT. Por fim, ´e proposto um sistema de inspeção autónomo que consiste num conjunto de trajetórias que permitem aplicar o sistema de monitorização de ativos elétricos ao longo de uma linha elétrica, através do mapeamento de estruturas e na recolha de amostras de isoladores em torno dessa estrutura.Dias, André Miguel PinheiroRepositório Científico do Instituto Politécnico do PortoBarbosa, Joel Adão Pacheco2021-11-27T01:30:38Z20202020-01-01T00:00:00Zinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisapplication/pdfhttp://hdl.handle.net/10400.22/16898TID:202573672enginfo:eu-repo/semantics/openAccessreponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãoinstacron:RCAAP2023-03-13T13:05:11Zoai:recipp.ipp.pt:10400.22/16898Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-19T17:36:32.983196Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãofalse
dc.title.none.fl_str_mv Deep Learning Approach for UAV Visual Electrical Assets Inspection
title Deep Learning Approach for UAV Visual Electrical Assets Inspection
spellingShingle Deep Learning Approach for UAV Visual Electrical Assets Inspection
Barbosa, Joel Adão Pacheco
Electrical Assets Inspection
Deep Learning
Object Detection
Multi-Object Tracking
UAV
Inspeção de Ativos Elétricos
Deteção de Objetos
Seguimento de Múltiplos Objetos
VANT
title_short Deep Learning Approach for UAV Visual Electrical Assets Inspection
title_full Deep Learning Approach for UAV Visual Electrical Assets Inspection
title_fullStr Deep Learning Approach for UAV Visual Electrical Assets Inspection
title_full_unstemmed Deep Learning Approach for UAV Visual Electrical Assets Inspection
title_sort Deep Learning Approach for UAV Visual Electrical Assets Inspection
author Barbosa, Joel Adão Pacheco
author_facet Barbosa, Joel Adão Pacheco
author_role author
dc.contributor.none.fl_str_mv Dias, André Miguel Pinheiro
Repositório Científico do Instituto Politécnico do Porto
dc.contributor.author.fl_str_mv Barbosa, Joel Adão Pacheco
dc.subject.por.fl_str_mv Electrical Assets Inspection
Deep Learning
Object Detection
Multi-Object Tracking
UAV
Inspeção de Ativos Elétricos
Deteção de Objetos
Seguimento de Múltiplos Objetos
VANT
topic Electrical Assets Inspection
Deep Learning
Object Detection
Multi-Object Tracking
UAV
Inspeção de Ativos Elétricos
Deteção de Objetos
Seguimento de Múltiplos Objetos
VANT
description The growth in the electrical demand by most countries around the world requires bigger and more complex energy systems, which leads to the requirement of having even more monitoring, inspection and maintenance of those systems. To respond to this need, inspection methods based on Unmanned Aerial Vehicles (UAV) have emerged which, when equipped with the appropriate sensors, allow a greater reduction of costs and risks and an increase in efficiency and effectiveness compared to traditional methods, such as inspection with foot patrols or helicopter-assisted. To make the inspection process more autonomous and reliable, most of the methods apply visual detection methods that use highly complex Deep Learning based algorithms and that require a very large computational power. This dissertation intends to present a system for inspection of electrical assets, able to be integrated onboard the UAV, based on Deep Learning, which allows to collect visual samples grouped and aggregated for each electrical asset detected. To this end, a perception system capable of detecting electrical insulators or structures, such as poles or transmission towers, was developed, using the Movidius Neural Compute Stick portable platform that is capable of processing lightweight object detection Convolutional Neural Networks, allowing a modular, low-cost system that meets real-time processing requirements. In addition to this perception system, an electrical asset monitoring system has been implemented that allows tracking and mapping each asset throughout the inspection process, based on the previous system’s detections and a UAV navigation system. Finally, an autonomous inspection system is proposed, which consists of a set of trajectories that allow an efficient application of the monitoring system along a power line, through the mapping of structures and the gathering of insulator samples around that structure.
publishDate 2020
dc.date.none.fl_str_mv 2020
2020-01-01T00:00:00Z
2021-11-27T01:30:38Z
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
dc.type.driver.fl_str_mv info:eu-repo/semantics/masterThesis
format masterThesis
status_str publishedVersion
dc.identifier.uri.fl_str_mv http://hdl.handle.net/10400.22/16898
TID:202573672
url http://hdl.handle.net/10400.22/16898
identifier_str_mv TID:202573672
dc.language.iso.fl_str_mv eng
language eng
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.source.none.fl_str_mv reponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
instacron:RCAAP
instname_str Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
instacron_str RCAAP
institution RCAAP
reponame_str Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
collection Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
repository.name.fl_str_mv Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação
repository.mail.fl_str_mv
_version_ 1799131456580091904