IEMEK Journal of Embedded Systems and Applications (대한임베디드공학회논문지)

Institute of Embedded Engineering of Korea (대한임베디드공학회)
권호 표지
DOI QR코드

DOI QR Code

Failure Detection Method of Industrial Cartesian Coordinate Robots Based on a CNN Inference Window Using Ambient Sound

음향 데이터를 이용한 CNN 추론 윈도우 기반 산업용 직교 좌표 로봇의 고장 진단 기법

  • Received : 2023.09.30
  • Accepted : 2023.12.15
  • Published : 2024.02.28
Download PDF
⟨ Previous Next ⟩

Abstract

In the industrial field, robots are used to increase productivity by replacing labors with dangerous, difficult, and hard tasks. However, failures of individual industrial robots in the entire production process may cause product defects or malfunctions, and may cause dangerous disasters in the case of manufacturing parts used in automobiles and aircrafts. Although requirements for early diagnosis of industrial robot failures are steadily increasing, there are many limitations in early detection. This paper introduces methods for diagnosing robot failures using sound-based data and deep learning. This paper also analyzes, compares, and evaluates the performance of failure diagnosis using various deep learning technologies. Furthermore, in order to improve the performance of the fault diagnosis system using deep learning technology, we propose a method to increase the accuracy of fault diagnosis based on an inference window. When adopting the inference window of deep learning, the accuracy of the failure diagnosis was increased up to 94%.

Keywords

Acknowledgement

이 논문은 2022학년도 동명대학교 교내학술연구비 지원에 의하여 연구되었음 (과제번호 2022B010).

References

  1. Korea Instute for Robot Industry Advancement, "Survey on Robot Industry 2021", 2021. (in Korean)
  2. KITECH, "AI-based Failure Diagnostic Device and Method," Korea patent No. 10-2289212-0000, 2021. 08. 06. (in Korean)
  3. D. T. Hoang, H. J. Kang, "A Survey on Deep Learning Based Bearing Fault Diagnosis," Neurocomputing, Vol. 335, No. 28, pp. 327-335, 2019. (in Korean) https://doi.org/10.1016/j.neucom.2018.06.078
  4. H. Cho, H. Choi, H. Kim, K. Nam, J. Ryu, K. Ha, "Feature Selection for Unmanned Surface Vehicle Fault Diagnosis Research and Experimental Verification," Journal of Institute of Control, Robotics and Systems, Vol. 28, No. 6, 542-550.
  5. R. Chalapathy, S. Chawla, "Deep Learning for Anomaly Detection: A Survey," arXiv preprint arXiv:1901.03407, 2019.
  6. V. Chandola, A. Banerjee, V. Kumar, "Anomaly Detection, A Survey," ACM Computing Surveys, Vol. 41, Iss. 3, No. 15, pp. 1-58, 2009.
  7. F. Mattia, P. Galeone, M. Simoni, E. Ghelfi, "A Survey on Gans for Anomaly Detection," arXiv preprint arXiv:1906.11632.
  8. V. Wilmet, S. Verma, T. Redl, H. Sandaker, Z. Li, "A Comparison of Supervised and Unsupervised Deep Learning Methods for Anomaly Detection in Images," arXiv preprint arXiv:2107.09204.
  9. N. Goernitz, M. M. Kloft, K. Rieck, U. Brefeld, "Toward Supervised Anomaly Detection," Journal Of Artificial Intelligence Research, Vol. 46, pp. 235-262, 2013.
  10. J. Yoon, K. Sohn, C. L. Li, S. Arik, C. Y. Lee, T. Pfister, "Self-supervise, Refine, Repeat: Improving Unsupervised Anomaly Detection," Transactions on Machine Learning Research, arXiv:2106.06115, 2022.
  11. B. Zong, Q. Song, M. R. Min, W. Cheng, C. Lumezanu, D. Cho, H. Chen, "Deep Autoencoding Gaussian Mixture Model for Unsupervised Anomaly Detection," International Conference on Learning Representations, 2018.
  12. L. Ruff, R. A. Vandermeulen, N. Gornitz, A. Binder, E. Muller, K.-R. Muller, M. Kloft, "Deep Semi-Supervised Anomaly Detection," arXiv:1906.02694, 2019.
  13. S. Kcay, A. Atapour-Abarghouei, T. P. Breckon, "GANomaly: Semi-supervised Anomaly Detection via Adversarial Training," Computer Vision - ACCV 2018, Lecture Notes in Computer Science, Vol 11363. Springer, Cham., 2018.
  14. T. Salimans, I. Goodfellow, W. Zaremba, V. Cheung, A. Radford, X. Chen, "Improved Techniques for Training GANs," International Conference on Neural Information Processing Systems, (NeuIPS2016), https://arxiv.org/abs/1606.03498, 2016.
  15. D. Li, D. Chen, L. Shi, B. Jin, J. Goh, S. K. Ng, "MAD-GAN: Multivariate Anomaly Detection for Time Series Data with Generative Adversarial Networks," arXiv:1901.04997, 2019.
  16. C. Zhou, R. C. Paffenroth, "Anomaly Detection with Robust Deep Autoencoders," In Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 665-674, 2017.
  17. M. Sakurada, T. Yairi, "Anomaly Detection Using Autoencoders with Nonlinear Dimensionality Reduction," International Workshop on Machine Learning for Sensory Data Analysis (MLSDA), pp. 4-11, 2014.
  18. Z. Chen, C. K. Yeo, B. S. Lee, C. T. Lau, "Autoencoder-based Network Anomaly Detection," In 2018 Wireless Telecommunications Symposium (WTS), pp. 1-5, 2018.
  19. A. A. Pol, V. Berger, C. Germain, G. Cerminara, M. Pierini, "Anomaly Detection with Conditional Variational Autoencoders," 18th IEEE International Conference on Machine Learning and Applications (ICMLA), pp. 1651-1657, 2019.
  20. O. I. Provotar, Y. M. Linder, M. M. Veres, "Unsupervised Anomaly Detection in Time Series Using LSTM-Based Autoencoders," In 2019 IEEE International Conference on Advanced Trends in Information Theory (ATIT), pp. 513-517, 2019.
  21. S. Yan, H. Shao, Y. Xiao, B. Liu, J. Wan, "Hybrid Robust Convolutional Autoencoder for Unsupervised Anomaly Detection of Machine Tools Under Noises," Robotics and Computer-Integrated Manufacturing, Vol. 79, pp. 102441, 2013.
  22. H. Purohit, R. Tanabe, K. Ichige, T. Endo, Y. Nikaido, K. Suefusa, Y. Kawaguchi, "MIMII Dataset: Sound Dataset for Malfunctioning Industrial Machine Investigation and Inspection," arXiv preprint arXiv:1909.09347, 2019.
  23. H. Purohit, R. Tanabe, K. Ichige, T. Endo, Y. Nikaido, K. Suefusa, Y. Kawaguchi, "MIMII Dataset: Sound Dataset for Malfunctioning Industrial Machine Investigation and Inspection," in Proc. 4th Workshop on Detection and Classification of Acoustic Scenes and Events (DCASE), 2019.
  24. Y. Kawaguchi, K. Imoto, Y. Koizumi, N. Harada, D. Niizumi, K. Dohi, R. Tanabe, H. Purohit, T. Endo, "Description and Discussion on DCASE 2021 Challenge Task 2: Unsupervised Anomalous Sound Detection for Machine Condition Monitoring under Domain Shifted Conditions," in arXiv e-prints: 2106.04492, 2021.
  25. R. Tanabe, H. Purohit, K. Dohi, T. Endo, Y. Nikaido, T. Nakamura, Y. Kawaguchi, "MIMII DUE: Sound Dataset for Malfunctioning Industrial Machine Investigation and Inspection with Domain Shifts due to Changes in Operational and Environmental Conditions," 2021 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA). IEEE, pp. 21-25, 2021.
  26. H. Cho, "CNN-based Autoencoder and Post-Training Quantization for On-Device Anomaly Detection of Cartesian Coordinate Robots," IEEE Computing and Communication Workshop and Conference (CCWC), 2024.
  27. I. Kuroyanagi, T. Hayashi, Y. Adachi, T. Yoshimura, K. Takeda, T. Toda, "Anomalous Sound Detection With Ensemble of Autoencoder and Binary Classification Approaches," Technical Report, Detection and Classification of Acoustic Scenes and Events, 2021.