Options
Development of an optical object detection solution for defect prevention in a Learning Factory
Journal
Procedia Manufacturing
ISSN-Digital
10.1016/j.promfg.2017.04.037
Type
conference paper
Date Issued
2017-04
Author(s)
Abstract
This article investigates a potential application of low cost computer hardware and open source software in machining areas of
Learning Factories. Based on the implementation of an optical object detection to discover human errors in the setup process of a milling machine, the paper presents an example of how value stream improvements in Learning Factories can be achieved by
student projects.
Therefore, the process of identifying possible solutions which are suitable for solving a problem in the value stream of a Learning
Factory is described. In the presented case, this is an IT-solution to establish a Poka-Yoke system for depositing a work piece in a milling machine correctly. A solution, which uses a Raspberry Pi, is developed and integrated in the process. Furthermore, the ability of low cost hardware components and simple algorithms resting upon freely available software libraries to fulfil the requirements of modern manufacturing is demonstrated. Finally, this study illustrates that the implementation of in-house low cost digitalization rather rests on a profound understanding of the affected manufacturing process than on previous knowledge of programming or electronics
Learning Factories. Based on the implementation of an optical object detection to discover human errors in the setup process of a milling machine, the paper presents an example of how value stream improvements in Learning Factories can be achieved by
student projects.
Therefore, the process of identifying possible solutions which are suitable for solving a problem in the value stream of a Learning
Factory is described. In the presented case, this is an IT-solution to establish a Poka-Yoke system for depositing a work piece in a milling machine correctly. A solution, which uses a Raspberry Pi, is developed and integrated in the process. Furthermore, the ability of low cost hardware components and simple algorithms resting upon freely available software libraries to fulfil the requirements of modern manufacturing is demonstrated. Finally, this study illustrates that the implementation of in-house low cost digitalization rather rests on a profound understanding of the affected manufacturing process than on previous knowledge of programming or electronics
Language
English
HSG Classification
not classified
Volume
9/2017
Start page
190
End page
197
Event Title
7th Conference on Learning Factories, CLF 2017
Event Location
Darmstadt
Event Date
4-5 April 2017
Subject(s)
Division(s)
Eprints ID
253527