Shape completion with a 3D Convolutional Neural Network for multi-domain O&M activities in offshore wind farms.
Autor(a) principal: | |
---|---|
Data de Publicação: | 2020 |
Tipo de documento: | Dissertação |
Idioma: | eng |
Título da fonte: | Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
Texto Completo: | https://hdl.handle.net/10216/132652 |
Resumo: | An autonomous vehicle needs to understand its surrounding environment to plan routes and avoid collisions. For that purpose, they are equipped with appropriate sensors which allow them to capture the necessary information. The maritime environment presents additional which make it hard to have a clear picture of the nearby structures. In this work, the goal is to use the available sensor information to infer the complete shape of nearby structures. The approach is divided into three main components: clustering, classification, and registration. The clustering is used to detect sizeable structures and remove irrelevant ones. The resulting data is voxelized, and classified, by a 3D CNN, as one of the studied structures. Finally, a hybrid PSO-ICP registration method is used to fit a complete CAD model on the observed data. |
id |
RCAP_dc77caff1dec2f03bea4fbec6685a194 |
---|---|
oai_identifier_str |
oai:repositorio-aberto.up.pt:10216/132652 |
network_acronym_str |
RCAP |
network_name_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository_id_str |
7160 |
spelling |
Shape completion with a 3D Convolutional Neural Network for multi-domain O&M activities in offshore wind farms.Engenharia electrotécnica, electrónica e informáticaElectrical engineering, Electronic engineering, Information engineeringAn autonomous vehicle needs to understand its surrounding environment to plan routes and avoid collisions. For that purpose, they are equipped with appropriate sensors which allow them to capture the necessary information. The maritime environment presents additional which make it hard to have a clear picture of the nearby structures. In this work, the goal is to use the available sensor information to infer the complete shape of nearby structures. The approach is divided into three main components: clustering, classification, and registration. The clustering is used to detect sizeable structures and remove irrelevant ones. The resulting data is voxelized, and classified, by a 3D CNN, as one of the studied structures. Finally, a hybrid PSO-ICP registration method is used to fit a complete CAD model on the observed data.2020-07-212020-07-21T00:00:00Zinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisapplication/pdfhttps://hdl.handle.net/10216/132652TID:202594599engRicardo Fernando de Freitas Dinisinfo: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:RCAAP2023-11-29T12:43:18Zoai:repositorio-aberto.up.pt:10216/132652Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-19T23:25:27.439752Repositório 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 |
Shape completion with a 3D Convolutional Neural Network for multi-domain O&M activities in offshore wind farms. |
title |
Shape completion with a 3D Convolutional Neural Network for multi-domain O&M activities in offshore wind farms. |
spellingShingle |
Shape completion with a 3D Convolutional Neural Network for multi-domain O&M activities in offshore wind farms. Ricardo Fernando de Freitas Dinis Engenharia electrotécnica, electrónica e informática Electrical engineering, Electronic engineering, Information engineering |
title_short |
Shape completion with a 3D Convolutional Neural Network for multi-domain O&M activities in offshore wind farms. |
title_full |
Shape completion with a 3D Convolutional Neural Network for multi-domain O&M activities in offshore wind farms. |
title_fullStr |
Shape completion with a 3D Convolutional Neural Network for multi-domain O&M activities in offshore wind farms. |
title_full_unstemmed |
Shape completion with a 3D Convolutional Neural Network for multi-domain O&M activities in offshore wind farms. |
title_sort |
Shape completion with a 3D Convolutional Neural Network for multi-domain O&M activities in offshore wind farms. |
author |
Ricardo Fernando de Freitas Dinis |
author_facet |
Ricardo Fernando de Freitas Dinis |
author_role |
author |
dc.contributor.author.fl_str_mv |
Ricardo Fernando de Freitas Dinis |
dc.subject.por.fl_str_mv |
Engenharia electrotécnica, electrónica e informática Electrical engineering, Electronic engineering, Information engineering |
topic |
Engenharia electrotécnica, electrónica e informática Electrical engineering, Electronic engineering, Information engineering |
description |
An autonomous vehicle needs to understand its surrounding environment to plan routes and avoid collisions. For that purpose, they are equipped with appropriate sensors which allow them to capture the necessary information. The maritime environment presents additional which make it hard to have a clear picture of the nearby structures. In this work, the goal is to use the available sensor information to infer the complete shape of nearby structures. The approach is divided into three main components: clustering, classification, and registration. The clustering is used to detect sizeable structures and remove irrelevant ones. The resulting data is voxelized, and classified, by a 3D CNN, as one of the studied structures. Finally, a hybrid PSO-ICP registration method is used to fit a complete CAD model on the observed data. |
publishDate |
2020 |
dc.date.none.fl_str_mv |
2020-07-21 2020-07-21T00:00:00Z |
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 |
https://hdl.handle.net/10216/132652 TID:202594599 |
url |
https://hdl.handle.net/10216/132652 |
identifier_str_mv |
TID:202594599 |
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_ |
1799135559327678464 |