EUVIMEDEUVIMED entdecken
Crossref

Parkour in the wild: Learning a general and extensible agile locomotion policy using multi-expert distillation and RL fine-tuning

Nikita Rudin, Junzhe He, Joshua Aurand, Marco Hutter

The International Journal of Robotics Research · 03.06.2026

DOI: 10.1177/02783649261455067 ↗

Kurz gefasst

Fachartikel in der Zeitschrift The International Journal of Robotics Research, online erschienen am 03.06.2026. Verfasst von Nikita Rudin und 3 weiteren Autor:innen. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 38 Literaturangaben. Gefördert von: Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung, Nvidia.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Legged robots are well-suited for navigating terrains inaccessible to wheeled robots, making them ideal for applications in search and rescue or space exploration. However, current control methods often struggle to generalize across diverse, unstructured environments. This paper introduces a novel framework for agile locomotion of legged robots by combining multi-expert distillation with reinforcement learning (RL) fine-tuning to achieve robust generalization. Initially, terrain-specific expert policies are trained to develop specialized locomotion skills. These policies are then distilled into a unified generalist policy via the DAgger algorithm. The distilled policy is subsequently fine-tuned using RL on a broader terrain set, including real-world 3D scans. The framework allows further adaptation to new terrains through repeated fine-tuning. The proposed policy leverages depth images as exteroceptive inputs, enabling robust navigation across diverse, unstructured terrains. Experimental results demonstrate significant performance improvements over existing methods in synthesizing multi-terrain skills into a single controller. Deployment on the ANYmal D robot validates the policy’s ability to navigate complex environments with agility and robustness, setting a new benchmark for legged robot locomotion.

Förderung

  • Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung
  • Nvidia

Zitieren

APA

Rudin, N., He, J., Aurand, J., & Hutter, M. (2026). Parkour in the wild: Learning a general and extensible agile locomotion policy using multi-expert distillation and RL fine-tuning. The International Journal of Robotics Research. https://doi.org/10.1177/02783649261455067

Vancouver

Rudin N, He J, Aurand J, Hutter M. Parkour in the wild: Learning a general and extensible agile locomotion policy using multi-expert distillation and RL fine-tuning. The International Journal of Robotics Research. 2026. doi:10.1177/02783649261455067

BibTeX
@article{Rudin2026,
  author = {Rudin, Nikita and He, Junzhe and Aurand, Joshua and Hutter, Marco},
  title = {Parkour in the wild: Learning a general and extensible agile locomotion policy using multi-expert distillation and RL fine-tuning},
  journal = {The International Journal of Robotics Research},
  year = {2026},
  doi = {10.1177/02783649261455067}
}
Kostenlos starten →