Journal of Applied Science and Engineering

Published by Tamkang University Press

ESCI jase impact factor scopus logo open access rate of Scopus journal

Robot Work Behavior Perception Using Position–Direction Baseline Compensation and Lightweight Multi-Scale Temporal Convolution

 2026

 Volume 35

Lili Li, Gengzhao Wang, Qi Yang, and Chonglei Shao

School of Mechanical Engineering, Shenyang Ligong University, Shenyang 110159, China

Received: August 18, 2026
Accepted: September 11, 2026
Publication Date: September 25, 2026

上傳圖片

Overall framework of position–direction baseline compensation and lightweight multi-scale temporal convolution for four-state perception

 Copyright The Author(s). This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are cited.

Download Citation: BibTeX | http://dx.doi.org/10.6180/jase.202612_35.062

Download PDF

Drive-current-only perception of robot joint states is complicated by payload, friction, and motion changes that can resemble soft collisions. This study proposes four-state perception combining position-direction baseline compensation with a lightweight multi-scale temporal convolutional network: a normal-current baseline is fitted per position and direction from the frozen training set, 13-point windows use raw current, compensated residuals, and residual first differences, and 12 statistical features are fused late. Each submodel has 4,600 parameters, with probabilities averaged over three fixed seeds. On 60 frozen independent-test windows, accuracy, macro-F1, and balanced accuracy are 0.9333, 0.9347, and 0.9479, with soft-collision recall of 1.0000 and all 20 runs correct. Results support four-state separability under the present acquisition system and frozen split, but not generalization across operating levels, devices, or industrial environments.

Keywords: robot joint; soft-collision recognition; position-direction baseline compensation; multi-scale temporal convolution; work behavior perception

  1. [1] S. Haddadin, A. De Luca, and A. Albu-Schäffer, (2017) “Robot Collisions: A Survey on Detection, Isolation, and Identification” IEEE Transactions on Robotics 33(6): 1292-1312. DOI: https://doi.org/10.1109/TRO.2017.2723903.
  2. [2] A. De Luca and R. Mattone. “Sensorless Robot Collision Detection and Hybrid Force/Motion Control”. In: Proceedings of the 2005 IEEE International Conference on Robotics and Automation. 2005, 999-1004. DOI: https://doi.org/10.1109/ROBOT.2005.1570247.
  3. [3] S. Haddadin, A. Albu-Schäffer, A. De Luca, and G. Hirzinger. “Collision Detection and Reaction: A Contribution to Safe Physical Human-Robot Interaction”. In: 2008 IEEE/RSJ International Conference on Intelligent Robots and Systems. 2008, 3356-3363. DOI: https://doi.org/10.1109/IROS.2008.4650764.
  4. [4] T. Ren, Y. Dong, D. Wu, and K. Chen, (2018) “Collision Detection and Identification for Robot Manipulators Based on Extended State Observer” Control Engineering Practice 79: 144-153. DOI: https://doi.org/10.1016/j.conengprac.2018.07.004.
  5. [5] T. Xu, H. Tuo, Q. Fang, D. Shan, H. Jin, J. Fan, Y. Zhu, and J. Zhao, (2024) “A Novel Collision Detection Method Based on Current Residuals for Robots Without Joint Torque Sensors: A Case Study on UR10 Robot” Robotics and Computer-Integrated Manufacturing 89: 102777. DOI: https://doi.org/10.1016/j.rcim.2024.102777.
  6. [6] P. Aivaliotis, S. Aivaliotis, C. Gkournelos, K. Kokkalis, G. Michalos, and S. Makris, (2019) “Power and Force Limiting on Industrial Robots for Human-Robot Collaboration” Robotics and Computer-Integrated Manufacturing 59: 346-360. DOI: https://doi.org/10.1016/j.rcim.2019.05.001.
  7. [7] K. Katsampiris-Salgado, K. Haninger, C. Gkrizis, N. Dimitropoulos, J. Krüger, G. Michalos, and S. Makris, (2024) “Collision Detection for Collaborative Assembly Operations on High-Payload Robots” Robotics and Computer-Integrated Manufacturing 87: 102708. DOI: https://doi.org/10.1016/j.rcim.2023.102708.
  8. [8] Y. J. Heo, D. Kim, W. Lee, H. Kim, J. Park, and W. K. Chung, (2019) “Collision Detection for Industrial Collaborative Robots: A Deep Learning Approach” IEEE Robotics and Automation Letters 4(2): 740-746. DOI: https://doi.org/10.1109/LRA.2019.2893400.
  9. [9] J. H. Lee, M. Hong, and K. M. Park, (2026) “Sensorless Collision Detection and Classification in Collaborative Robots Using Stacked GRU Networks” Actuators 15(4): 206. DOI: https://doi.org/10.3390/act15040206.
  10. [10] S. Bai, J. Z. Kolter, and V. Koltun. An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling. arXiv:1803.01271. 2018. DOI: https://doi.org/10.48550/arXiv.1803.01271.
  11. [11] H. Ismail Fawaz, B. Lucas, G. Forestier, C. Pelletier, D. F. Schmidt, J. Weber, G. I. Webb, L. Idoumghar, P.-A. Muller, and F. Petitjean, (2020) “InceptionTime: Finding AlexNet for Time Series Classification” Data Mining and Knowledge Discovery 34(6): 1936-1962. DOI: https://doi.org/10.1007/s10618-020-00710-y.
  12. [12] K. He, X. Zhang, S. Ren, and J. Sun. “Deep Residual Learning for Image Recognition”. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2016, 770-778. DOI: https://doi.org/10.1109/CVPR.2016.90.
  13. [13] J. Hu, L. Shen, and G. Sun. “Squeeze-and-Excitation Networks”. In: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2018, 7132-7141. DOI: https://doi.org/10.1109/CVPR.2018.00745.
  14. [14] G. C. Cawley and N. L. C. Talbot, (2010) “On Over-fitting in Model Selection and Subsequent Selection Bias in Performance Evaluation” Journal of Machine Learning Research 11(70): 2079-2107. URL: https://www.jmlr.org/papers/v11/cawley10a.html.
  15. [15] A. Savitzky and M. J. E. Golay, (1964) “Smoothing and Differentiation of Data by Simplified Least Squares Procedures” Analytical Chemistry 36(8): 1627-1639. DOI: https://doi.org/10.1021/ac60214a047.
  16. [16] K. H. Brodersen, C. S. Ong, K. E. Stephan, and J. M. Buhmann. “The Balanced Accuracy and Its Posterior Distribution”. In: 2010 20th International Conference on Pattern Recognition. 2010, 3121-3124. DOI: https://doi.org/10.1109/ICPR.2010.764.
  17. [17] I. Loshchilov and F. Hutter. “Decoupled Weight Decay Regularization”. In: International Conference on Learning Representations. 2019. URL: https://openreview.net/forum?id=Bkg6RiCqY7.
  18. [18] F. Pedregosa, G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V. Dubourg, J. Vanderplas, A. Passos, D. Cournapeau, M. Brucher, M. Perrot, and É. Duchesnay, (2011) “Scikit-learn: Machine Learning in Python” Journal of Machine Learning Research 12(85): 2825-2830. URL: https://www.jmlr.org/papers/v12/pedregosa11a.html.
  19. [19] L. Ma, L. Dong, Y. Han, J. Yan, and J. Zhou, (2020) “A Method of Sensorless Collision Detection Based on Motor Current for Robot Manipulator” International Journal of Applied Electromagnetics and Mechanics 64(1-4): 237-244. DOI: https://doi.org/10.3233/JAE-209327.
  20. [20] M. Czubenko and Z. Kowalczuk, (2021) “A Simple Neural Network for Collision Detection of Collaborative Robots” Sensors 21(12): 4235. DOI: https://doi.org/10.3390/s21124235.
  21. [21] S. Kaufman, S. Rosset, C. Perlich, and O. Stitelman, (2012) “Leakage in Data Mining: Formulation, Detection, and Avoidance” ACM Transactions on Knowledge Discovery from Data 6(4): 15:1-15:21. DOI: https://doi.org/10.1145/2382577.2382579.
  22. [22] S. Varma and R. Simon, (2006) “Bias in Error Estimation When Using Cross-Validation for Model Selection” BMC Bioinformatics 7: 91. DOI: https://doi.org/10.1186/1471-2105-7-91.
  23. [23] V. Villani, F. Pini, F. Leali, and C. Secchi, (2018) “Survey on Human-Robot Collaboration in Industrial Settings: Safety, Intuitive Interfaces and Applications” Mechatronics 55: 248-266. DOI: https://doi.org/10.1016/j.mechatronics.2018.02.009.