EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

High-Speed Humanoid Running with Trajectory-Optimized Motion Priors and Reinforcement Learning

Chengwei Zhuo, Shaowen Cheng, Chengkai Su, Chaojie Fu, Lei Jiang, Yongbin Jin, Hongtao Wang

Biomimetics · 05.11.2026 · 11 (11) · 811

DOI: 10.3390/biomimetics11110811 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biomimetics, online erschienen am 05.11.2026. Verfasst von Chengwei Zhuo und 6 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

High-speed human running data are difficult to acquire, limiting the availability of motion priors for high-speed humanoid running. To reduce dependence on motion capture and retargeting, this study proposes a framework that combines trajectory optimization with reinforcement learning. A planar multibody model generates reference motions across speeds subject to contact, periodicity, and actuator constraints. The optimized motions guide the training of a three-dimensional running policy through adversarial imitation and proximal policy optimization. Hardware experiments demonstrate treadmill running by the full-size BOLT humanoid at 10m/s. In 2s simulation windows under a 10m/s forward velocity command, comparison with a policy trained on retargeted human motions shows a more balanced distribution of joint loads relative to actuator capacity. The coefficient of variation of joint root-mean-square (RMS) utilization decreases from 0.397 to 0.317 across twelve joints, a reduction of 20.1%, with a corresponding reduction of 35.6% across the six leg joints. These results support trajectory optimization as an effective source of motion priors for high-speed humanoid running and demonstrate a route from reference generation to hardware deployment without high-speed human motion data.

Zitieren

APA

Zhuo, C., Cheng, S., Su, C., Fu, C., Jiang, L., Jin, Y., & Wang, H. (2026). High-Speed Humanoid Running with Trajectory-Optimized Motion Priors and Reinforcement Learning. Biomimetics, 11(11), 811. https://doi.org/10.3390/biomimetics11110811

Vancouver

Zhuo C, Cheng S, Su C, Fu C, Jiang L, Jin Y, et al. High-Speed Humanoid Running with Trajectory-Optimized Motion Priors and Reinforcement Learning. Biomimetics. 2026;11(11):811. doi:10.3390/biomimetics11110811

BibTeX
@article{Zhuo2026,
  author = {Zhuo, Chengwei and Cheng, Shaowen and Su, Chengkai and Fu, Chaojie and Jiang, Lei and Jin, Yongbin and Wang, Hongtao},
  title = {High-Speed Humanoid Running with Trajectory-Optimized Motion Priors and Reinforcement Learning},
  journal = {Biomimetics},
  year = {2026},
  volume = {11},
  number = {11},
  pages = {811},
  doi = {10.3390/biomimetics11110811}
}
Kostenlos starten →