Object Shape Error Response using Bayesian 3D Convolutional Neural Networks for Assembly Systems with Compliant Parts

Sinha, Sumit; Franciosa, Pasquale; Ceglarek, Darek

doi:10.1109/indin45582.2020.9442245

Cited by 2 publications

(2 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This aids designers and engineers to make informed decision early on to improve product quality while reduce rework [44,45]. Further it can aid root cause diagnosis [46,47,48].…”

Section: Discussionmentioning

confidence: 99%

Modeling Variation in Multi-Station Compliant Assembly Using Parametric Space Envelope

Luo

Nie

Franciosa

et al. 2023

Journal of Manufacturing Science and Engineering

View full text Add to dashboard Cite

Non-rigid compliant parts are widely used in industries today. One of the biggest challenges facing the industries is geometric variation management of these compliant parts, which directly impacts product quality and functionality. Existing rigid body based variation modeling is not suitable for compliant assembly while finite element analysis based methods have the disadvantage of requiring heavy computation efforts. In view of that, this paper develops a novel methodology to evaluate geometric variation propagation in multi-station compliant assembly based upon parametric space envelope (i.e. variation tool constructed from parametric curves). Three sources of variation: location-led positional variation, assembly deformation-induced variation and station transition caused variation are analyzed. Under proposal, geometric variations are modeled indirectly through a compact set of boundary control points. Compared with existing methods where geometric variation is modeled through targeting key feature points on the manufacturing part, the proposed approach brings modeling accuracy and computation efficiency. The effectiveness of the method is illustrated and verified through an industrial case study on a multi-station compliant panel assembly. The developed method provides industries a new way to manage geometric variation from compliant assembly.

show abstract

“…This aids designers and engineers to make informed decision early on to improve product quality while reduce rework [44,45]. Further it can aid root cause diagnosis [46,47,48].…”

Section: Discussionmentioning

confidence: 99%

Modeling Variation in Multi-Station Compliant Assembly Using Parametric Space Envelope

Luo

Nie

Franciosa

et al. 2023

Journal of Manufacturing Science and Engineering

View full text Add to dashboard Cite

show abstract

“…Wang et al [5] proposed deep learning as a method for supporting visual observation of human workers' movements. Ceglarek et al [6] propose a novel Object Shape Error Response (OSER) approach to estimate the dimensional and geometric variation of the assembled products and [7] use deep learning for 3D object shape error modelling and estimate dimensional and geometric quality defects in multi-station assembly systems. Image based robot manipulator control system was developed by Copot et al [8], and particularly visual control robot manipulator using image moments.…”

Section: Introductionmentioning

confidence: 99%

Development of a neural network structure for identifying begin-end points in the assembly process

Kutschenreiter-Praszkiewicz¹

2023

Journal of Machine Engineering

View full text Add to dashboard Cite

The paper presents an approach to video-based assembly analysis using machine learning. A neural network is one of the machine learning methods that is widely studied in many engineering fields. The purpose of this paper is to develop a deep neural network structure for identifying begin-end points for a selected component assembly process. A neural network structure that effectively identifies begin-end points is proposed and an example from industry is presented. The proposed approach can prove useful in the assembly process analysis.

show abstract

Object Shape Error Response using Bayesian 3D Convolutional Neural Networks for Assembly Systems with Compliant Parts

Abstract: Please refer to published version for the most recent bibliographic citation information. If a published version is known of, the repository item page linked to above, will contain details on accessing it.

Cited by 2 publications

References 35 publications

Modeling Variation in Multi-Station Compliant Assembly Using Parametric Space Envelope

Modeling Variation in Multi-Station Compliant Assembly Using Parametric Space Envelope

Development of a neural network structure for identifying begin-end points in the assembly process

Contact Info

Product

Resources

About