Chapter Topological Relationship Modelling for Industrial Facility Digitisation Using Graph Neural Networks

There is rising demand for automated digital twin construction based on point cloud scans, especially in the domain of industrial facilities. Yet, current automation approaches focus almost exclusively on geometric modelling. The output of these methods is a disjoint cluster of individual elements,...

Full beskrivning

Sparad:
Bibliografiska uppgifter
Huvudupphov: Jayasinghe, Haritha, Brilakis, Ioannis
Materialtyp: Online
Språk:engelska
Utgiven: Firenze University Press 2024
Ämnen:
Länkar:ONIX_20240402_9791221502893_13
Taggar: Lägg till en tagg
Inga taggar, Lägg till första taggen!
_version_ 1869515110531203072
author Jayasinghe, Haritha
Brilakis, Ioannis
author_browse Brilakis, Ioannis
Jayasinghe, Haritha
author_facet Jayasinghe, Haritha
Brilakis, Ioannis
author_sort Jayasinghe, Haritha
collection Directory of Open Access Books
description There is rising demand for automated digital twin construction based on point cloud scans, especially in the domain of industrial facilities. Yet, current automation approaches focus almost exclusively on geometric modelling. The output of these methods is a disjoint cluster of individual elements, while element relationships are ignored. This research demonstrates the feasibility of adopting Graph Neural Networks (GNN) for automated detection of connectivity relationships between elements in industrial facility scans. We propose a novel method which represents elements and relationships as graph nodes and edges respectively. Element geometry is encoded into graph node features. This allows relationship inference to be modelled as a graph link prediction task. We thereby demonstrate that connectivity relationships can be learned from existing design files, without requiring domain specific, hand-coded rules, or manual annotations. Preliminary results show that our method performs successfully on a synthetic point cloud testset generated from design files with a 0.64 F1 score. We further demonstrate that the method adapts to occluded real-world scans. The method can be further extended with the introduction of more descriptive node features. Additionally, we present tools for relationship annotation and visualisation to aid relationship detection
format Online
id doab-20.500.12854ir-136916
institution Directory of Open Access Books
language eng
publishDate 2024
publishDateRange 2024
publishDateSort 2024
publisher Firenze University Press
publisherStr Firenze University Press
record_format ojs
spelling doab-20.500.12854ir-1369162024-05-10T18:13:54Z Chapter Topological Relationship Modelling for Industrial Facility Digitisation Using Graph Neural Networks Jayasinghe, Haritha Brilakis, Ioannis BIM Digital twin GNN machine learning thema EDItEUR::U Computing and Information Technology::UT Computer networking and communications::UTV Virtualization There is rising demand for automated digital twin construction based on point cloud scans, especially in the domain of industrial facilities. Yet, current automation approaches focus almost exclusively on geometric modelling. The output of these methods is a disjoint cluster of individual elements, while element relationships are ignored. This research demonstrates the feasibility of adopting Graph Neural Networks (GNN) for automated detection of connectivity relationships between elements in industrial facility scans. We propose a novel method which represents elements and relationships as graph nodes and edges respectively. Element geometry is encoded into graph node features. This allows relationship inference to be modelled as a graph link prediction task. We thereby demonstrate that connectivity relationships can be learned from existing design files, without requiring domain specific, hand-coded rules, or manual annotations. Preliminary results show that our method performs successfully on a synthetic point cloud testset generated from design files with a 0.64 F1 score. We further demonstrate that the method adapts to occluded real-world scans. The method can be further extended with the introduction of more descriptive node features. Additionally, we present tools for relationship annotation and visualisation to aid relationship detection 2024-05-10T18:13:51Z 2024-05-10T18:13:51Z 2024-04-02T15:44:40Z 2023 chapter ONIX_20240402_9791221502893_13 2704-5846 https://library.oapen.org/handle/20.500.12657/89044 9791221502893 https://directory.doabooks.org/handle/20.500.12854/136916 eng Proceedings e report open access image/jpeg n/a https://library.oapen.org/bitstream/20.500.12657/89044/1/9791221502893_88.pdf Firenze University Press 10.36253/979-12-215-0289-3.88 10.36253/979-12-215-0289-3.88 2ec4474d-93b1-4cfa-b313-9c6019b51b1a 9791221502893 8 Florence open access
spellingShingle BIM
Digital twin
GNN
machine learning
thema EDItEUR::U Computing and Information Technology::UT Computer networking and communications::UTV Virtualization
Jayasinghe, Haritha
Brilakis, Ioannis
Chapter Topological Relationship Modelling for Industrial Facility Digitisation Using Graph Neural Networks
title Chapter Topological Relationship Modelling for Industrial Facility Digitisation Using Graph Neural Networks
title_full Chapter Topological Relationship Modelling for Industrial Facility Digitisation Using Graph Neural Networks
title_fullStr Chapter Topological Relationship Modelling for Industrial Facility Digitisation Using Graph Neural Networks
title_full_unstemmed Chapter Topological Relationship Modelling for Industrial Facility Digitisation Using Graph Neural Networks
title_short Chapter Topological Relationship Modelling for Industrial Facility Digitisation Using Graph Neural Networks
title_sort chapter topological relationship modelling for industrial facility digitisation using graph neural networks
topic BIM
Digital twin
GNN
machine learning
thema EDItEUR::U Computing and Information Technology::UT Computer networking and communications::UTV Virtualization
topic_facet BIM
Digital twin
GNN
machine learning
thema EDItEUR::U Computing and Information Technology::UT Computer networking and communications::UTV Virtualization
url ONIX_20240402_9791221502893_13
work_keys_str_mv AT jayasingheharitha chaptertopologicalrelationshipmodellingforindustrialfacilitydigitisationusinggraphneuralnetworks
AT brilakisioannis chaptertopologicalrelationshipmodellingforindustrialfacilitydigitisationusinggraphneuralnetworks