liu.seSearch for publications in DiVA
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Failure detection in robotic arms using  statistical modeling, machine learning and hybrid gradient boosting
Universidade Federal de Minas Gerais, Brazil.
Robotics and Motion Division, ABB AB.
Robotics and Motion Division, ABB AB.
Linköping University, Department of Electrical Engineering, Automatic Control. Linköping University, Faculty of Science & Engineering.
2018 (English)Report (Other academic)
Abstract [en]

Modeling and failure prediction is an important task in manyengineering systems. For this task, the machine learning literaturepresents a large variety of models such as classification trees,random forest, artificial neural networks, fuzzy systems, amongothers. In addition, standard statistical models can be applied suchas the logistic regression, linear discriminant analysis, $k$-nearestneighbors, among others. This work evaluates advantages andlimitations of statistical and machine learning methods to predictfailures in industrial robots. The work is based on data from morethan five thousand robots in industrial use. Furthermore, a newapproach combining standard statistical and machine learning models,named \emph{hybrid gradient boosting}, is proposed. Results show thatthe a priori treatment of the database, i.e., outlier analysis,consistent database analysis and anomaly analysis have shown to becrucial to improve classification performance for statistical, machinelearning and hybrid models. Furthermore, local joint information hasbeen identified as the main driver for failure detection whereasfailure classification can be improved using additional informationfrom different joints and hybrid models.

Place, publisher, year, edition, pages
Linköping: Linköping University Electronic Press, 2018. , p. 33
Series
LiTH-ISY-R, ISSN 1400-3902 ; 3107
Keywords [en]
Failure detection in robotic arms using statistical modeling, machine learning and hybrid gradient boosting
National Category
Control Engineering
Identifiers
URN: urn:nbn:se:liu:diva-152437ISRN: LiTH-ISY-R-3107OAI: oai:DiVA.org:liu-152437DiVA, id: diva2:1259788
Projects
LINK-SIC
Funder
VINNOVA, 2016-05152Available from: 2018-10-31 Created: 2018-10-31 Last updated: 2018-10-31Bibliographically approved

Open Access in DiVA

fulltext(1591 kB)102 downloads
File information
File name FULLTEXT01.pdfFile size 1591 kBChecksum SHA-512
0068c9a94a0fac4ba42c4c9b817721012369e9f2c2a083966a8c3ce3fb2db48da29dfd49e7d00f824a9edeb012739604d939494768eb9ab34def60b10756f8d1
Type fulltextMimetype application/pdf

Authority records BETA

Gunnarsson, Svante

Search in DiVA

By author/editor
Gunnarsson, Svante
By organisation
Automatic ControlFaculty of Science & Engineering
Control Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 102 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 368 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf