EUVIMEDEUVIMED entdecken
Crossref

Learning agile soccer skills for a bipedal robot with deep reinforcement learning

Tuomas Haarnoja iD, Ben Moran iD, Guy Lever iD, Sandy H Huang iD, Dhruva Tirumala, Jan Humplik, Markus Wulfmeier iD, Saran Tunyasuvunakool iD, Noah Y Siegel iD, Roland Hafner iD, Michael Bloesch iD, Kristian Hartikainen und 16 weitere

Science Robotics · 10.04.2024 · 9 (89)

DOI: 10.1126/scirobotics.adi8022 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Science Robotics, erschienen am 10.04.2024. Verfasst von Tuomas Haarnoja und 27 weiteren Autor:innen. Beteiligte Einrichtung: Google DeepMind, London, UK. Hauptthemen laut MEDLINE: Robotics, Soccer. Die Arbeit wurde bisher 189-mal zitiert (Crossref) und nennt 112 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We investigated whether deep reinforcement learning (deep RL) is able to synthesize sophisticated and safe movement skills for a low-cost, miniature humanoid robot that can be composed into complex behavioral strategies. We used deep RL to train a humanoid robot to play a simplified one-versus-one soccer game. The resulting agent exhibits robust and dynamic movement skills, such as rapid fall recovery, walking, turning, and kicking, and it transitions between them in a smooth and efficient manner. It also learned to anticipate ball movements and block opponent shots. The agent’s tactical behavior adapts to specific game contexts in a way that would be impractical to manually design. Our agent was trained in simulation and transferred to real robots zero-shot. A combination of sufficiently high-frequency control, targeted dynamics randomization, and perturbations during training enabled good-quality transfer. In experiments, the agent walked 181% faster, turned 302% faster, took 63% less time to get up, and kicked a ball 34% faster than a scripted baseline.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

RoboticsSoccerLearningWalkingComputer Simulation

Autor:innen und Institutionen

  • Google DeepMind, London, UK.
  • University College London, London, UK.
  • Proactive Global, London, UK.

ORCID-Profile: Tuomas Haarnoja, Ben Moran, Guy Lever, Sandy H Huang, Markus Wulfmeier, Saran Tunyasuvunakool, Noah Y Siegel, Roland Hafner, Michael Bloesch, Leonard Hasenclever, Yuval Tassa, Charles Game, Andrea Huber, Francesco Nori, Raia Hadsell, Nicolas Heess

Zitieren

APA

Haarnoja, T., Moran, B., Lever, G., Huang, S. H., Tirumala, D., Humplik, J., Wulfmeier, M., Tunyasuvunakool, S., Siegel, N. Y., Hafner, R., Bloesch, M., Hartikainen, K., Byravan, A., Hasenclever, L., Tassa, Y., Sadeghi, F., Batchelor, N., Casarini, F., Saliceti, S., …, & Heess, N. (2024). Learning agile soccer skills for a bipedal robot with deep reinforcement learning. Science Robotics, 9(89). https://doi.org/10.1126/scirobotics.adi8022

Vancouver

Haarnoja T, Moran B, Lever G, Huang SH, Tirumala D, Humplik J, et al. Learning agile soccer skills for a bipedal robot with deep reinforcement learning. Science Robotics. 2024;9(89). doi:10.1126/scirobotics.adi8022

BibTeX
@article{Haarnoja2024,
  author = {Haarnoja, Tuomas and Moran, Ben and Lever, Guy and Huang, Sandy H and Tirumala, Dhruva and Humplik, Jan and Wulfmeier, Markus and Tunyasuvunakool, Saran and Siegel, Noah Y and Hafner, Roland and Bloesch, Michael and Hartikainen, Kristian and Byravan, Arunkumar and Hasenclever, Leonard and Tassa, Yuval and Sadeghi, Fereshteh and Batchelor, Nathan and Casarini, Federico and Saliceti, Stefano and Game, Charles and Sreendra, Neil and Patel, Kushal and Gwira, Marlon and Huber, Andrea and Hurley, Nicole and Nori, Francesco and Hadsell, Raia and Heess, Nicolas},
  title = {Learning agile soccer skills for a bipedal robot with deep reinforcement learning},
  journal = {Science Robotics},
  year = {2024},
  volume = {9},
  number = {89},
  doi = {10.1126/scirobotics.adi8022}
}
Kostenlos starten →