EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Meta reinforcement learning for robust quadruped locomotion under joint failures

Likai Wu iD, Chee-Onn Chow iD, Wei Ru Wong, Joon Huang Chuah, Jeevan Kanesan, Meina Zhang

PLOS One · 2026

DOI: 10.1371/journal.pone.0361099 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, 2026. Verfasst von Likai Wu und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Electrical Engineering, Faculty of Engineering, Universiti Malaya, Kuala Lumpur, Malaysia. Hauptthemen laut MEDLINE: Joints, Robotics, Locomotion. Die Arbeit nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Dezhou Engineering Research Center of Industrial Robot System Integration.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Joint failures can disrupt gait coordination, velocity tracking, and body stability in quadruped robots by reducing the available control authority. This study investigates faults in which joints remain locked at fixed angles and presents Adjudicate Reconfiguration and Evolving Regularization Meta Reinforcement Learning (ARER-MetaRL). The framework combines task inference from context, modular routing tailored to each task through an Adjudicate Reconfiguration (AR) network, and Evolving Gradient Regularization (ER) to support rapid adaptation. Tasks differ in the identities and locking angles of the faulty joints, the number of simultaneous faults, the commanded motion direction, and the terrain. Experiments in NVIDIA Isaac Sim cover all 12 actuated joints, with 0–3 stuck joints during meta-training and severity tests involving up to 6 simultaneously stuck joints. ARER-MetaRL achieves higher Mean Episode Linear Velocity Tracking Reward (MELV) and Mean Episode Roll–Pitch Tracking Reward (MERP) than single-task reinforcement learning, ablated variants, and the evaluated learning and adaptation baselines. Cumulative rewards further demonstrate the benefit of combining the proposed components. In the study of fault severity, ARER-MetaRL achieves MELV/MERP values of 0.86/0.84 with one stuck joint and 0.55/0.54 with six stuck joints, compared with 0.21/0.18 and 0.05/0.03, respectively, for the single-task baseline. These results show improved adaptation across different configurations of locked joints and terrain conditions.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

JointsRoboticsLocomotionReinforcement Machine LearningHumansAlgorithmsGait

Autor:innen und Institutionen

  • Department of Electrical Engineering, Faculty of Engineering, Universiti Malaya, Kuala Lumpur, Malaysia.
  • School of Electrical Engineering, Shandong Huayu University of Technology, Dezhou, China.
  • Department of Naval Architecture and Ocean Engineering, Pusan National University, Busan, South Korea.

ORCID-Profile: Likai Wu, Chee-Onn Chow

Förderung

  • Dezhou Engineering Research Center of Industrial Robot System Integration

Zitieren

APA

Wu, L., Chow, C. O., Wong, W. R., Chuah, J. H., Kanesan, J., & Zhang, M. (2026). Meta reinforcement learning for robust quadruped locomotion under joint failures. PLOS One. https://doi.org/10.1371/journal.pone.0361099

Vancouver

Wu L, Chow CO, Wong WR, Chuah JH, Kanesan J, Zhang M. Meta reinforcement learning for robust quadruped locomotion under joint failures. PLOS One. 2026. doi:10.1371/journal.pone.0361099

BibTeX
@article{Wu2026,
  author = {Wu, Likai and Chow, Chee-Onn and Wong, Wei Ru and Chuah, Joon Huang and Kanesan, Jeevan and Zhang, Meina},
  title = {Meta reinforcement learning for robust quadruped locomotion under joint failures},
  journal = {PLOS One},
  year = {2026},
  doi = {10.1371/journal.pone.0361099}
}
Kostenlos starten →