The Use of Joints Force Sensors to Determine the Collision Location and Type for an Industrial Robot
https://doi.org/10.17587/mau.20.171-179
Abstract
Recent advances in the development of sensors allowed to obtain robots with torque-sensitive sensors in each joint. At the moment, these sensors are used only to detect collision. This work shows the possibility of obtaining information on the collision point and it type. This information can subsequently be used to select the robot’s behavior strategy. The contact point localization is realized using two approaches: the analytical approach and machine learning. Analytical approach is based on finding point on the robot length and direction of applied external force where an equivalent torques will be the same as torques in a real robot. In the machine learning approach various learning technics were tested. For the collision type identification a classification tree was proposed that distinguish soft and hard collision, purposeful and accidental, single and continuous. The algorithm at the first stage detects presence of a collision, and if there is a collision localizes it and identify its type. The described algorithms were tested on an industrial manipulator Kuka iiwa LBR 14 R820, ground truth information about the experiments was obtained using a 3D lidar.
Keywords
human-robot interaction,
contact detection,
collision classification,
collision localization,
industrial robot,
neural network
Supporting agencies: Russian Science Foundation
About the Authors
D. I. Popov
Innopolis University
Russian Federation
Russian Federation
Popov D. I. - M. Sc., Junior researcher, Robotics Development Center.
Innopolis, 420500.
A. S. Klimchik
Innopolis University
Russian Federation
Innopolis, 420500.
Russian Federation
Innopolis, 420500.
References
1. De Santis A. et al. An atlas of physical human—robot interaction, Mechanism and Machine Theory, 2008, vol. 43, no. 3, pp. 253—270.
2. Wang X. et al. Robot manipulator self-identification for surrounding obstacle detection, Multimedia Tools and Applications, 2016, pp. 1—26.
3. Morikawa S. et al. Realtime collision avoidance using a robot manipulator with light-weight small high-speed vision systems, Robotics and Automation, 2007, pp. 794—799.
4. Part S. I. Impedance control: A n approach to manipulation, Journal of dynamic systems, measurement, and control, 1985, vol. 107, pp. 17.
5. Ebert D. M., Henrich D. D. Safe human-robot-cooperation: Image-based collision detection for industrial robots, Intelligent Robots and Systems, 2002, vol. 2, pp. 1826—1831.
6. Magrini E., Flacco F., De Luca A. Control of generalized contact motion and force in physical human-robot interaction, Robotics and Automation (ICR A), 2015, pp. 2298—2304.
7. Cannata G. et al. An embedded artificial skin for humanoid robots, Multisensor Fusion and Integration for Intelligent Systems, 2008, pp. 434-438.
8. Fritzsche M., Elkmann N., Schulenburg E. Tactile sensing: A key technology for safe physical human robot interaction, Proceedings of the 6th international conference on Human-robot interaction, ACM, 2011, pp. 139—140.
9. Fritzsche M., Saenz J., Penzlin F. A large scale tactile sensor for safe mobile robot manipulation, The Eleventh ACM/ IEEE International Conference on Human Robot Interaction, IEEE Press, 2016, pp. 427—428.
10. McMahan W., Romano J. M., Kuchenbecker K. J. Using accelerometers to localize tactile contact events on a robot arm, Proc. Workshop on Advances in Tactile Sensing and Touch-Based Human-Robot Interaction, 2012.
11. Petrovskaya A., Park J., Khatib O. Probabilistic estimation of whole body contacts for multi-contact robot control, Robotics and Automation, 2007, pp. 568—573.
12. Magrini E., Flacco F., De Luca A. Estimation of contact forces using a virtual force sensor, Intelligent Robots and Systems, 2014, pp. 2126—2133.
13. Dimeas F., Avendano-Valencia L. D., Aspragathos N. Human-robot collision detection and identification based on fuzzy and time series modelling, Robotica, 2015, vol. 33, no. 09, pp. 1886—1898.
14. Hur S., Oh S. R., Oh Y. Joint space torque controller based on time-delay control with collision detection, Intelligent Robots and Systems (IROS 2014), pp. 4710—4715.
15. Zelenak A., Pryor M., Schroeder K. An Extended Kalman Filter for Collision Detection During Manipulator Contact Tasks, ASME 2014 Dynamic Systems and Control Conference, American Society of Mechanical Engineers, 2014, pp. V001T11A005-V001T11A005.
16. Lee S. D., Kim M. C., Song J. B. Sensorless collision detection for safe human-robot collaboration, Intelligent Robots and Systems, 2015, pp. 2392—2397.
17. Geravand M., Flacco F., De Luca A. Human-robot physical interaction and collaboration using an industrial robot with a closed control architecture, Robotics and Automation (ICR A), IEEE, 2013, pp. 4000—4007.
18. De Luca A., Mattone R. Sensorless robot collision detection and hybrid force/motion control, Robotics and Automation, 2005, pp. 999—1004.
19. De Luca A. et al. Collision detection and safe reaction with the DLR-III lightweight manipulator arm, Intelligent Robots and Systems, 2006, pp. 1623—1630.
20. Haddadin S. et al. Collision detection and reaction: A contribution to safe physical human-robot interaction, Intelligent Robots and Systems, 2008, pp. 3356—3363.
21. Hennersperger C. et al. Towards MRI-based autonomous robotic US acquisitions: a first feasibility study, IEEE transactions on medical imaging, 2017, vol. 36, no. 2, pp. 538—548.
22. Likar N., Žlajpah L. External joint torque-based estimation of contact information, International Journal of Advanced Robotic Systems, 2014, vol. 11, no. 7, pp. 107.
23. Finkel D. E. DIRECT optimization algorithm user guide, Center for Research in Scientific Computation, North Carolina State University, 2003, vol. 2.
Review
For citations:
Popov D.I., Klimchik A.S. The Use of Joints Force Sensors to Determine the Collision Location and Type for an Industrial Robot. Mekhatronika, Avtomatizatsiya, Upravlenie. 2019;20(3):171-179. (In Russ.) https://doi.org/10.17587/mau.20.171-179
Views:
744
.png)
Email this article 