The Journal of Korea Robotics Society (로봇학회논문지)

Korea Robotics Society (한국로봇학회)
권호 표지
DOI QR코드

DOI QR Code

Transformer based Collision Detection Approach by Torque Estimation using Joint Information

관절 정보를 이용한 토크 추정 방식의 트랜스포머 기반 로봇 충돌 검출 방법

  • Jiwon Park (Dept. of Electronic Engineering, Kyung Hee University) ;
  • Daegyu Lim (Robros Inc.) ;
  • Sumin Park (Robros Inc.) ;
  • Hyeonjun Park (Robros Inc.)
  • 박지원 ;
  • 임대규 ;
  • 박수민 ;
  • 박현준
  • Received : 2024.03.29
  • Accepted : 2024.05.22
  • Published : 2024.08.30
Download PDF
⟨ Previous Next ⟩

Abstract

With the rising interaction between robots and humans, detecting collisions has become increasingly vital for ensuring safety. In this paper, we propose a novel approach for detecting collisions without using force torque sensors or tactile sensors, utilizing a Transformer-based neural network architecture. The proposed collision detection approach comprises a torque estimator network that predicts the joint torque in a free-motion state using Synchronous time-step encoding, and a collision discriminator network that predicts collisions by leveraging the difference between estimated and actual torques. The collision discriminator finally creates a binary tensor that predicts collisions frame by frame. In simulations, the proposed network exhibited enhanced collision detection performance relative to the other kinds of networks both in terms of prediction speed and accuracy. This underscores the benefits of using Transformer networks for collision detection tasks, where rapid decision-making is essential.

Keywords

Acknowledgement

This research was supported by Deep-Tech Tips funded by Ministry of SMEs and Startups.

References

  1. S. Robla-Gomez, V. M. Becerra, J. R. Llata, E. Gonzalez-Sarabia, C. Torre-Ferrero, and J. Perez-Oria, "Working together: A review on safe human-robot collaboration in industrial environments," IEEE Access, vol. 5, pp. 26754-26773, Nov., 2017, DOI: 10.1109/ACCESS.2017.2773127. 
  2. J. Pan, I. A. Sucan, S. Chitta, and D. Manocha, "Real-time collision detection and distance computation on point cloud sensor data," IEEE Int. Conf. Robotics and Automation, Karlsruhe, Germany, pp. 3593-3599, 2013, DOI: 10.1109/ICRA.2013.6631081. 
  3. W. Li, Y. Han, J. Wu, and Z. Xiong, "Collision detection of robots based on a force/torque sensor at the bedplate," IEEE/ASME Transactions on Mechatronics, vol. 25, no. 5, pp. 2565-2573, Oct., 2020, DOI: 10.1109/TMECH.2020.2995904. 
  4. X. Wenzhong, X. Xi, and J. Xinjian, "Sensorless robot collision detection based on optimized velocity deviation," 2017 Chinese Automation Congress (CAC), Jinan, China, pp. 6200-6204, 2017, DOI: 10.1109/CAC.2017.8243894. 
  5. F. Flacco, T. Kroger, A. De Luca, and O. Khatib, "A depth space approach to human-robot collision avoidance," 2012 IEEE International Conference on Robotics and Automation, Saint Paul, MN, USA, pp. 338-345, 2012, DOI: 10.1109/ICRA.2012.6225245. 
  6. C.-N. Cho, S.-D. Lee, and J.-B. Song, "Collision Detection Algorithm based on Velocity Error," Journal of Korea Robotics Society, vol. 9, no. 2, pp. 111-116, May, 2014, DOI: 10.7746/jkros.2014.9.2.111. 
  7. B.-J. Jung, T.-K. Kim, G. Won, D. S. Kim, and J. Hwang, "Development of Joint Controller and Collision Detection Methods for Series Elastic Manipulator of Relief Robot," Journal of Korea Robotics Society, vol. 13, no. 3, pp. 157-163, Aug., 2018, DOI: 10.7746/jkros.2018.13.3.157. 
  8. A. D. Libera, E. Tosello, G. Pillonetto, S. Ghidoni, and R. Carli, "Proprioceptive Robot Collision Detection through Gaussian Process Regression," 2019 American Control Conference (ACC), Philadelphia, PA, USA, pp. 19-24, 2019, DOI: 10.23919/ACC.2019.8814361. 
  9. D. Lim, M.-J. Kim, J. Cha, D. Kim, and J. Park, "Proprioceptive External Torque Learning for Floating Base Robot and its Applications to Humanoid Locomotion," 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 8510-8517, 2023, DOI: 10.1109/IROS55552.2023.10342530. 
  10. D. Lim, D. Kim, and J. Park, "Momentum observer-based collision detection using lstm for model uncertainty learning," 2021 IEEE International Conference on Robotics and Automation (ICRA), Xi'an, China, pp. 4516-4522, 2021, DOI: 10.1109/ICRA48506.2021.9561667. 
  11. D. Kim, D. Lim, and J. Park, "Transferable Collision Detection Learning for Collaborative Manipulator Using Versatile Modularized Neural Network," IEEE Transactions on Robotics, vol. 38, no. 4, pp. 2426-2445, Aug., 2021, DOI: 10.1109/TRO.2021.3129630. 
  12. K. M. Park, Y. Park, S. Yoon, and F. C. Park, "Collision detection for robot manipulators using unsupervised anomaly detection algorithms," IEEE/ASME Transactions on Mechatronics, vol. 27, no. 5, pp. 2841-2851, Oct., 2022, DOI: 10.1109/TMECH.2021.3119057. 
  13. A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. Kaiser, and I. Polosukhin, "Attention is all you need," arXiv:1706.03762, 2017, DOI: 10.48550/arXiv.1706.03762. 
  14. E. Todorov, T. Erez, and Y. Tassa, "Mujoco: A physics engine for model-based control," 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems, Vilamoura-Algarve, Portugal, pp. 5026-5033, 2012, DOI: 10.1109/IROS.2012.6386109.