Multi-dimensional Error Identification during Robotic Snap Assembly

  • Yusuke HayamiEmail author
  • Peihao Shi
  • Weiwei Wan
  • Ixchel G. Ramirez-Alpizar
  • Kensuke Harada
Conference paper
Part of the Mechanisms and Machine Science book series (Mechan. Machine Science, volume 73)


In this research, we propose a novel error identification method during robotic snap assembly aiming at automated recovery from error states. In the proposed method, we first obtain the feature quantities of force/torque from the simulated snap assembly by using functional principal component analysis (fPCA). Then, we cluster these data into success and several different error states based on the k-means clustering by using the decision tree considering the multi-dimensional feature of the force/torque signal. Furthermore, we try to predict an error state of a snap assembly task before the error actually happens. Finally, we show simulation results to show the effectiveness of the proposed method.


Snap assembly K-means clustering Functional PCA 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    A.Rodriguez, D.Bourne, M.Mason, G.F.Rossano, J.Wang. “Failure Detection in Assembly: Force Signature Analysis ”. IEEE International Conference on Automation Science and Engineering, 2010.Google Scholar
  2. 2.
    A.Scolt, M.Linderoth, A.Robertsson, R.Johansson. “Detection of contact force transients in robotic assembly ”. IEEE International Conference on Robotics and Automation (ICRA), 2015.Google Scholar
  3. 3.
    J.Rojas, K.Harada, H.Onda, N.Yamanobe, E.Yoshida, K.Nagata. “Early failure characterization of cantilever snap assemblies using the PA-RCBHT ”. IEEE International Conference on Robotics and Automation (ICRA), 2014.Google Scholar
  4. 4.
    Ge Ming, Xu Yangsheng, Du Ruxu. “ An Intelligent Online Monitoring and Diagnostic System for Manufacturing Automation ”. IEEE Transactions on Automation Science and Engineering, 2008.Google Scholar
  5. 5.
    L.Rusli, A.Luscher, C.Sommerich. “ Force and tactile feedback in preloaded cantilever snap-fits under manual assembly ”. International Journal of Industrial Ergonomics, 2010.Google Scholar
  6. 6.
    S.Gen, R.W.M.Jr., G.A.Gabriele. “A hierarchical classification scheme to define and order the design space for integral snap-fit assembly ”. Research in Engineering Design,1998.Google Scholar
  7. 7.
    J.Ji, K.-M.Lee, S.Zhang. “Cantilever Snap-Fit Performance Analysis for Haptic Evaluation”. Journal of Mechanical Design, 2011.Google Scholar
  8. 8.
    J.Rojas, K.Harada, H.Onda, N.Yamanobe, E.Yoshida, K.Nagata, Y.Kawai. “ A Relative-Change-Based Hierarchical Taxonomy for Cantilever-Snap Assembly Verification. ”. IEEE/RSJ International Conference on Intelligent Robots and Systems, 2012.Google Scholar
  9. 9.
    E.D.Lello, M.Klotzbucher, T.D.LAet, H.Bruyninckx . “Bayesian Time-Series Models for Continuous Fault Detection andRecognition in Industrial Robotic Tasks ”. IEEE/RSJ International Conference on Intelligent Robots and Systems, 2013.Google Scholar
  10. 10.
    W.Luo, J.Rojas, T.Guan, K.Harada, K.Nagata. “ Cantilever snap assemblies failure detection using SVMs and the RCBHT”. IEEE International Conference on Mechatronics and Automation, 2014.Google Scholar
  11. 11.
  12. 12.
    Y.Yamanishi, Y.Tanaka, “ Principal Component Analysis for Functional Data ” Journal of Faculty of Environmental Science and Technology, Vol.6, 25-34, 2000Google Scholar
  13. 13.
    Besse, P. and Ramsay, J. O. ”Principal components analysis of sampled functions”, Psychometrika, 51, 285-311, 1986Google Scholar
  14. 14.
    Ramsay, Y. O. and Silverman, B.W. Functional Data Analysis, Springer, 1997Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Yusuke Hayami
    • 1
    Email author
  • Peihao Shi
    • 1
  • Weiwei Wan
    • 1
    • 2
  • Ixchel G. Ramirez-Alpizar
    • 1
  • Kensuke Harada
    • 1
    • 2
  1. 1.Osaka UniversityToyonakaJapan
  2. 2.National Inst. of Advanced Industrial Science and TechnologyTsukubaJapan

Personalised recommendations