EUVIMEDEUVIMED entdecken
Crossref

Real-world humanoid locomotion with reinforcement learning

Ilija Radosavovic iD, Tete Xiao iD, Bike Zhang iD, Trevor Darrell, Jitendra Malik iD, Koushil Sreenath iD

Science Robotics · 17.04.2024 · 9 (89)

DOI: 10.1126/scirobotics.adi9579 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Science Robotics, erschienen am 17.04.2024. Verfasst von Ilija Radosavovic und 5 weiteren Autor:innen. Beteiligte Einrichtung: University of California, Berkeley CA, USA. Hauptthemen laut MEDLINE: Robotics. Die Arbeit wurde bisher 259-mal zitiert (Crossref) und nennt 61 Literaturangaben. Gefördert von: DARPA Machine Common Sense program, ONR MURI program.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Humanoid robots that can autonomously operate in diverse environments have the potential to help address labor shortages in factories, assist elderly at home, and colonize new planets. Although classical controllers for humanoid robots have shown impressive results in a number of settings, they are challenging to generalize and adapt to new environments. Here, we present a fully learning-based approach for real-world humanoid locomotion. Our controller is a causal transformer that takes the history of proprioceptive observations and actions as input and predicts the next action. We hypothesized that the observation-action history contains useful information about the world that a powerful transformer model can use to adapt its behavior in context, without updating its weights. We trained our model with large-scale model-free reinforcement learning on an ensemble of randomized environments in simulation and deployed it to the real-world zero-shot. Our controller could walk over various outdoor terrains, was robust to external disturbances, and could adapt in context.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

RoboticsHumansAgedLocomotionWalkingLearningReinforcement, Psychology

Autor:innen und Institutionen

  • University of California, Berkeley CA, USA.

ORCID-Profile: Ilija Radosavovic, Tete Xiao, Bike Zhang, Jitendra Malik, Koushil Sreenath

Förderung

  • DARPA Machine Common Sense program, ONR MURI program

Zitieren

APA

Radosavovic, I., Xiao, T., Zhang, B., Darrell, T., Malik, J., & Sreenath, K. (2024). Real-world humanoid locomotion with reinforcement learning. Science Robotics, 9(89). https://doi.org/10.1126/scirobotics.adi9579

Vancouver

Radosavovic I, Xiao T, Zhang B, Darrell T, Malik J, Sreenath K. Real-world humanoid locomotion with reinforcement learning. Science Robotics. 2024;9(89). doi:10.1126/scirobotics.adi9579

BibTeX
@article{Radosavovic2024,
  author = {Radosavovic, Ilija and Xiao, Tete and Zhang, Bike and Darrell, Trevor and Malik, Jitendra and Sreenath, Koushil},
  title = {Real-world humanoid locomotion with reinforcement learning},
  journal = {Science Robotics},
  year = {2024},
  volume = {9},
  number = {89},
  doi = {10.1126/scirobotics.adi9579}
}
Kostenlos starten →