From a Point Cloud to a Simulation Model—Bayesian Segmentation and Entropy Based Uncertainty Estimation for 3D Modelling
The 3D modelling of indoor environments and the generation of process simulations play an important role in factory and assembly planning. In brownfield planning cases, existing data are often outdated and incomplete especially for older plants, which were mostly planned in 2D. Thus, current environ...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-03-01
|
Series: | Entropy |
Subjects: | |
Online Access: | https://www.mdpi.com/1099-4300/23/3/301 |
id |
doaj-ad9c4e2ea1e549b3a95c75012d90b49c |
---|---|
record_format |
Article |
spelling |
doaj-ad9c4e2ea1e549b3a95c75012d90b49c2021-03-04T00:02:25ZengMDPI AGEntropy1099-43002021-03-012330130110.3390/e23030301From a Point Cloud to a Simulation Model—Bayesian Segmentation and Entropy Based Uncertainty Estimation for 3D ModellingChristina Petschnigg0Markus Spitzner1Lucas Weitzendorf2Jürgen Pilz3BMW Group, Department of Factory Planning, Knorrstraße 147, 80788 Munich, GermanyBMW Group, Department of Factory Planning, Knorrstraße 147, 80788 Munich, GermanyBMW Group, Department of Factory Planning, Knorrstraße 147, 80788 Munich, GermanyDepartment of Statistics, Alpen-Adria-University Klagenfurt, Universitätsstraße 65-67, 9020 Klagenfurt, AustriaThe 3D modelling of indoor environments and the generation of process simulations play an important role in factory and assembly planning. In brownfield planning cases, existing data are often outdated and incomplete especially for older plants, which were mostly planned in 2D. Thus, current environment models cannot be generated directly on the basis of existing data and a holistic approach on how to build such a factory model in a highly automated fashion is mostly non-existent. Major steps in generating an environment model of a production plant include data collection, data pre-processing and object identification as well as pose estimation. In this work, we elaborate on a methodical modelling approach, which starts with the digitalization of large-scale indoor environments and ends with the generation of a static environment or simulation model. The object identification step is realized using a Bayesian neural network capable of point cloud segmentation. We elaborate on the impact of the uncertainty information estimated by a Bayesian segmentation framework on the accuracy of the generated environment model. The steps of data collection and point cloud segmentation as well as the resulting model accuracy are evaluated on a real-world data set collected at the assembly line of a large-scale automotive production plant. The Bayesian segmentation network clearly surpasses the performance of the frequentist baseline and allows us to considerably increase the accuracy of the model placement in a simulation scene.https://www.mdpi.com/1099-4300/23/3/301factory planningfactory simulationdigital factoryBayesian deep learninguncertainty estimationpoint clouds |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Christina Petschnigg Markus Spitzner Lucas Weitzendorf Jürgen Pilz |
spellingShingle |
Christina Petschnigg Markus Spitzner Lucas Weitzendorf Jürgen Pilz From a Point Cloud to a Simulation Model—Bayesian Segmentation and Entropy Based Uncertainty Estimation for 3D Modelling Entropy factory planning factory simulation digital factory Bayesian deep learning uncertainty estimation point clouds |
author_facet |
Christina Petschnigg Markus Spitzner Lucas Weitzendorf Jürgen Pilz |
author_sort |
Christina Petschnigg |
title |
From a Point Cloud to a Simulation Model—Bayesian Segmentation and Entropy Based Uncertainty Estimation for 3D Modelling |
title_short |
From a Point Cloud to a Simulation Model—Bayesian Segmentation and Entropy Based Uncertainty Estimation for 3D Modelling |
title_full |
From a Point Cloud to a Simulation Model—Bayesian Segmentation and Entropy Based Uncertainty Estimation for 3D Modelling |
title_fullStr |
From a Point Cloud to a Simulation Model—Bayesian Segmentation and Entropy Based Uncertainty Estimation for 3D Modelling |
title_full_unstemmed |
From a Point Cloud to a Simulation Model—Bayesian Segmentation and Entropy Based Uncertainty Estimation for 3D Modelling |
title_sort |
from a point cloud to a simulation model—bayesian segmentation and entropy based uncertainty estimation for 3d modelling |
publisher |
MDPI AG |
series |
Entropy |
issn |
1099-4300 |
publishDate |
2021-03-01 |
description |
The 3D modelling of indoor environments and the generation of process simulations play an important role in factory and assembly planning. In brownfield planning cases, existing data are often outdated and incomplete especially for older plants, which were mostly planned in 2D. Thus, current environment models cannot be generated directly on the basis of existing data and a holistic approach on how to build such a factory model in a highly automated fashion is mostly non-existent. Major steps in generating an environment model of a production plant include data collection, data pre-processing and object identification as well as pose estimation. In this work, we elaborate on a methodical modelling approach, which starts with the digitalization of large-scale indoor environments and ends with the generation of a static environment or simulation model. The object identification step is realized using a Bayesian neural network capable of point cloud segmentation. We elaborate on the impact of the uncertainty information estimated by a Bayesian segmentation framework on the accuracy of the generated environment model. The steps of data collection and point cloud segmentation as well as the resulting model accuracy are evaluated on a real-world data set collected at the assembly line of a large-scale automotive production plant. The Bayesian segmentation network clearly surpasses the performance of the frequentist baseline and allows us to considerably increase the accuracy of the model placement in a simulation scene. |
topic |
factory planning factory simulation digital factory Bayesian deep learning uncertainty estimation point clouds |
url |
https://www.mdpi.com/1099-4300/23/3/301 |
work_keys_str_mv |
AT christinapetschnigg fromapointcloudtoasimulationmodelbayesiansegmentationandentropybaseduncertaintyestimationfor3dmodelling AT markusspitzner fromapointcloudtoasimulationmodelbayesiansegmentationandentropybaseduncertaintyestimationfor3dmodelling AT lucasweitzendorf fromapointcloudtoasimulationmodelbayesiansegmentationandentropybaseduncertaintyestimationfor3dmodelling AT jurgenpilz fromapointcloudtoasimulationmodelbayesiansegmentationandentropybaseduncertaintyestimationfor3dmodelling |
_version_ |
1724232489422028800 |