Extracting stairs and doors as natural landmarks for mobile robot localization from clouds of 3D edge-points

Natural landmarks are the main features in the next step of the research in localization of mobile robot platforms. The identification and recognition of these landmarks are crucial to better localize a robot. To help solving this problem, this work proposes an approach for the identification and r...

Fuld beskrivelse

Na minha lista:
Bibliografiske detaljer
Hovedforfatter: Souto, Leonardo Ângelo Virginio de
Andre forfattere: Gonçalves, Luiz Marcos Garcia
Format: doctoralThesis
Sprog:pt_BR
Udgivet: Universidade Federal do Rio Grande do Norte
Fag:
Marcos naturais
Sensores RGB-D
Nuvem de ponta 3D
Localização de robô
Online adgang:https://repositorio.ufrn.br/handle/123456789/31930
Tags: Tilføj Tag
Ingen Tags, Vær først til at tagge denne postø!
id ri-123456789-31930
record_format dspace
spelling ri-123456789-319302021-03-21T08:48:16Z Extracting stairs and doors as natural landmarks for mobile robot localization from clouds of 3D edge-points Souto, Leonardo Ângelo Virginio de Gonçalves, Luiz Marcos Garcia http://lattes.cnpq.br/7947681999420065 http://lattes.cnpq.br/1562357566810393 Nascimento, Tiago Pereira do http://lattes.cnpq.br/1641673656667170 Gomes, Rafael Beserra http://lattes.cnpq.br/5849107545126304 Silva, Bruno Marques Ferreira da http://lattes.cnpq.br/7878437620254155 Souza, Anderson Abner de Santana http://lattes.cnpq.br/2563070123322776 Marcos naturais Sensores RGB-D Nuvem de ponta 3D Localização de robô Natural landmarks are the main features in the next step of the research in localization of mobile robot platforms. The identification and recognition of these landmarks are crucial to better localize a robot. To help solving this problem, this work proposes an approach for the identification and recognition of natural marks included in the environment using images from RGB-D sensors. In the identification step, a structural analysis of the natural landmarks that are present in the environment is performed. The extraction of edge points of these landmarks is done using the 3D point cloud obtained from the RGBD sensor. These edge points are smoothed through the Sl0 algorithm, which minimizes the standard deviation of the normals at each point. Then, the second step of the proposed algorithm begins, which is the proper recognition of the natural landmarks. This recognition step is done as a real-time algorithm that extracts the points referring to the filtered edges and determines to which structure they belong to in the current scenario: stairs or doors. Finally, the geometrical characteristics that are intrinsic to the doors and stairs are identified. The approach proposed here has been validated with real robot experiments. The performed tests verify the efficacy of our proposed approach. Marcas naturais são as principais características na próxima etapa da pesquisa em localização de plataformas de robôs móveis. A identificação e o reconhecimento de marcas são cruciais para melhor localizar um robô. Para ajudar a resolver este problema, este trabalho propõe uma abordagem para a identificação e reconhecimento de marcas naturais incluídas no ambiente utilizando imagens de sensores RGB-D. Na etapa de identificação, é realizada uma análise estrutural dos marcos naturais que estão presentes no ambiente. A extração dos pontos de borda desses marcos é feita usando a nuvem de pontos 3D obtida do sensor RGB-D. Esses pontos de borda são suavizados por meio do algoritmo Sl0, que minimiza o desvio padrão das normais em cada ponto. Então, começa a segunda etapa do algoritmo proposto, que é o reconhecimento adequado dos marcos naturais. Esta etapa de reconhecimento é feita como um algoritmo em tempo real que extrai os pontos referentes às arestas filtradas e determina a qual estrutura pertencem no cenário atual: escadas ou portas. Por fim, são identificadas as características geométricas intrínsecas às portas e escadas. A abordagem proposta aqui foi validada com experimentos reais de robôs. Os testes realizados verificam a eficácia de nossa abordagem proposta. 2021-03-17T23:13:12Z 2021-03-17T23:13:12Z 2020-11-30 doctoralThesis SOUTO, Leonardo Ângelo Virginio de. Extracting stairs and doors as natural landmarks for mobile robot localization from clouds of 3D edge-points. 2020. 64f. Tese (Doutorado em Engenharia Elétrica e de Computação) - Centro de Tecnologia, Universidade Federal do Rio Grande do Norte, Natal, 2020. https://repositorio.ufrn.br/handle/123456789/31930 pt_BR Acesso Aberto application/pdf Universidade Federal do Rio Grande do Norte Brasil UFRN PROGRAMA DE PÓS-GRADUAÇÃO EM ENGENHARIA ELÉTRICA E DE COMPUTAÇÃO
institution Repositório Institucional
collection RI - UFRN
language pt_BR
topic Marcos naturais
Sensores RGB-D
Nuvem de ponta 3D
Localização de robô
spellingShingle Marcos naturais
Sensores RGB-D
Nuvem de ponta 3D
Localização de robô
Souto, Leonardo Ângelo Virginio de
Extracting stairs and doors as natural landmarks for mobile robot localization from clouds of 3D edge-points
description Natural landmarks are the main features in the next step of the research in localization of mobile robot platforms. The identification and recognition of these landmarks are crucial to better localize a robot. To help solving this problem, this work proposes an approach for the identification and recognition of natural marks included in the environment using images from RGB-D sensors. In the identification step, a structural analysis of the natural landmarks that are present in the environment is performed. The extraction of edge points of these landmarks is done using the 3D point cloud obtained from the RGBD sensor. These edge points are smoothed through the Sl0 algorithm, which minimizes the standard deviation of the normals at each point. Then, the second step of the proposed algorithm begins, which is the proper recognition of the natural landmarks. This recognition step is done as a real-time algorithm that extracts the points referring to the filtered edges and determines to which structure they belong to in the current scenario: stairs or doors. Finally, the geometrical characteristics that are intrinsic to the doors and stairs are identified. The approach proposed here has been validated with real robot experiments. The performed tests verify the efficacy of our proposed approach.
author2 Gonçalves, Luiz Marcos Garcia
author_facet Gonçalves, Luiz Marcos Garcia
Souto, Leonardo Ângelo Virginio de
format doctoralThesis
author Souto, Leonardo Ângelo Virginio de
author_sort Souto, Leonardo Ângelo Virginio de
title Extracting stairs and doors as natural landmarks for mobile robot localization from clouds of 3D edge-points
title_short Extracting stairs and doors as natural landmarks for mobile robot localization from clouds of 3D edge-points
title_full Extracting stairs and doors as natural landmarks for mobile robot localization from clouds of 3D edge-points
title_fullStr Extracting stairs and doors as natural landmarks for mobile robot localization from clouds of 3D edge-points
title_full_unstemmed Extracting stairs and doors as natural landmarks for mobile robot localization from clouds of 3D edge-points
title_sort extracting stairs and doors as natural landmarks for mobile robot localization from clouds of 3d edge-points
publisher Universidade Federal do Rio Grande do Norte
publishDate 2021
url https://repositorio.ufrn.br/handle/123456789/31930
work_keys_str_mv AT soutoleonardoangelovirginiode extractingstairsanddoorsasnaturallandmarksformobilerobotlocalizationfromcloudsof3dedgepoints
_version_ 1773966273531609088

Lignende værker