Control robótico inteligente para extracción de elementos flexibles

  1. Tapia Sal Paz, Benjamin 1
  2. Sorrosal, Gorka 1
  3. Mancisidor, Aitziber 2
  4. Cabanes, Itziar 2
  1. 1 IKERLAN
  2. 2 Universidad del País Vasco/Euskal Herriko Unibertsitatea
    info

    Universidad del País Vasco/Euskal Herriko Unibertsitatea

    Lejona, España

    ROR https://ror.org/000xsnr85

Journal:
Jornadas de Automática
  1. Cruz Martín, Ana María (coord.)
  2. Arévalo Espejo, V. (coord.)
  3. Fernández Lozano, Juan Jesús (coord.)

ISSN: 3045-4093

Year of publication: 2024

Issue: 45

Type: Article

DOI: 10.17979/JA-CEA.2024.45.10927 DIALNET GOOGLE SCHOLAR lock_openOpen access editor

Abstract

The automation of disassembly tasks presents significant challenges, mainly related to the dynamic and unstructured charac-teristics of the task, where adaptive actions are needed to ensure proper interaction between the robot and the task environment.This work proposes a reinforcement learning-based control to automate flexible element extraction tasks using robots, aimingto tackle the difficulties of working in these unstructured and dynamic environments. To achieve this, the proposed control willlearn to take appropriate actions in the robot’s movement that will extract flexible elements through low-force trajectories. As aresult, this work demonstrates how integrating a reinforcement learning-based controller can address the challenges of flexibleelement extraction, thereby contributing to the advancement of intelligent disassembly processes using robots.

Bibliographic References

  • Beltran-Hernandez, C. C., Petit, D., Ramirez-Alpizar, I. G., Harada, K., 10 2020. Variable compliance control for robotic peg-in-hole assembly: A deep-reinforcement-learning approach. Applied Sciences (Switzerland) 10, 1–17. DOI: 10.3390/app10196923 DOI: https://doi.org/10.3390/app10196923
  • Kristensen, C. B., Sørensen, F. A., Nielsen, H. B., Andersen, M. S., Bendtsen, S. P., Bøgh, S., 2019. Towards a robot simulation framework for e-waste disassembly using reinforcement learning. Vol. 38. Elsevier B.V., pp. 225–232. DOI: 10.1016/j.promfg.2020.01.030 DOI: https://doi.org/10.1016/j.promfg.2020.01.030
  • Kroemer, O., Niekum, S., Konidaris, G., 2021. A review of robot learning for manipulation: Challenges, representations, and algorithms. URL: http://jmlr.org/papers/v22/19-804.html.
  • Kurilova-Palisaitiene, J., Sundin, E., Poksinska, B., 1 2018. Remanufacturing challenges and possible lean improvements. Journal of Cleaner Production 172, 3225–3236. DOI: 10.1016/j.jclepro.2017.11.023 DOI: https://doi.org/10.1016/j.jclepro.2017.11.023
  • Lillicrap, T. P., Hunt, J. J., Pritzel, A., Heess, N., Erez, T., Tassa, Y., Silver, D., Wierstra, D., 9 2015. Continuous control with deep reinforcement learning. URL: http://arxiv.org/abs/1509.02971
  • Paz, B. T. S., Sorrosal, G., Mancisidor, A., Cabanes, I., 8 2023. Control híbrido para la manipulación robótica de elementos flexibles. Servizo de Publicacións. Universidade da Coruña, pp. 768–772. DOI: 10.17979/spudc.9788497498609.768 DOI: https://doi.org/10.17979/spudc.9788497498609.768
  • Poschmann, H., Br ̈uggemann, H., Goldmann, D., 4 2020. Disassembly 4.0: A review on using robotics in disassembly tasks as a way of automation. DOI: 10.1002/cite.201900107 DOI: https://doi.org/10.1002/cite.201900107
  • Sutton, R. S., Barto, A. G., 2018. Reinforcement learning : an introduction, 2nd Edition. The MIT Press.
  • Zachares, P. A., Lee, M. A., Lian, W., Bohg, J., 10 2021. Interpreting contact interactions to overcome failure in robot assembly tasks. Institute of Electrical and Electronics Engineers (IEEE), pp. 3410–3417. DOI: 10.1109/icra48506.2021.9560825 DOI: https://doi.org/10.1109/ICRA48506.2021.9560825
  • Zhou, Z., Ni, P., Zhu, X., Cao, Q., 11 2021. Compliant robotic assembly based on deep reinforcement learning. Institute of Electrical and Electronics Engineers (IEEE), pp. 6–9, la precision requerida en operaciones de peg-in-hole siguen siendo un problema abierto. DOI: 10.1109/mlise54096.2021.00009 DOI: https://doi.org/10.1109/MLISE54096.2021.00009
Data source: Dialnet