@inproceedings{811f5fcce2e04b7090d98caedd22184a,
title = "Data-based assembly failure state estimation of mobile IT parts using a 6 DOF manipulator",
abstract = "This study discusses data-based failure state estimation of the mobile IT parts assembly using a 6 DOF manipulator. A position control-based robotic assembly is fast and simple for automation of production lines. However, when the assembly fails, it is very difficult to find the error that causes the assembly to fail. And the worker should stop and intervene in the assembly process to compensate the error. This is time-consuming and inefficient for the productivity of factory automation. To compensate the error without the aid of worker, this study presents a method for assembly failure state estimation. First, the failure state modeling of the mobile IT parts assembly is proposed. And the supervised learning was used for training whose input is the F/T sensor data and whose output is the failure state of the assembly. Furthermore, it is shown that artificial neural network (ANN) can lead to a higher classification accuracy for estimating the failure state and faster prediction.",
keywords = "Assembly failure state, Fault detection, Robotic assembly, State estimation, Supervised learning",
author = "Na, {Min Woo} and Kim, {Tae Jung} and Song, {Jae Bok}",
note = "Funding Information: This research was supported by the MOTIE under the Industrial Foundation Technology Development Program supervised by the KEIT (No. 10060110) Publisher Copyright: {\textcopyright} ICROS.; 18th International Conference on Control, Automation and Systems, ICCAS 2018 ; Conference date: 17-10-2018 Through 20-10-2018",
year = "2018",
month = dec,
day = "10",
language = "English",
series = "International Conference on Control, Automation and Systems",
publisher = "IEEE Computer Society",
pages = "1703--1707",
booktitle = "International Conference on Control, Automation and Systems",
}