EUVIMEDEUVIMED entdecken
Crossref

Legs-as-Arms: A Multitask Reinforcement Learning Framework for Simultaneous Loco-Manipulation in Hexapod Robots

Zhelin Zhang, Liang Ding, Peng Xu, Tie Liu, Ao Zhang, Huaiguang Yang, Haibo Gao, Zongquan Deng

IEEE Transactions on Industrial Electronics · 2026-08 · 73 (8) · 11775-11786

DOI: 10.1109/tie.2026.3663691 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Industrial Electronics, 2026-08. Verfasst von Zhelin Zhang und 7 weiteren Autor:innen. Die Arbeit nennt 37 Literaturangaben. Gefördert von: National Key R&D Program of China, National Natural Science Foundation of China, National Natural Science Foundation of China, Basic Research Program for Young Students.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Legged robots are designed to perform operational tasks in complex, rugged terrains. These tasks generally encompass two fundamental modalities: locomotion and manipulation. Humanoid robots, benefiting from their distinctive morphological configuration (independent arm and leg structures), can decouple the control of these two task types. In contrast, conventional quadruped/hexapod robots, lacking dedicated end effectors, are typically optimized for locomotion. While manipulation capabilities can be added via auxiliary robotic arms, this approach leads to unnecessary energy consumption and sacrifices payload capacity. Adopting a limb reuse architecture offers a promising solution. This work proposes Legs-as-Arms, a multitask reinforcement learning framework for hexapod robots that enables simultaneous locomotion and manipulation. The framework implements effective limb reuse, enabling any leg(s) to track desired trajectories while maintaining stable body movement. By leveraging a multihead critic network and value normalization, we integrate three key capabilities into a single control policy: 1) base locomotion control; 2) stationary whole-body manipulation; and 3) dynamic mobile manipulation. Our method demonstrates fault-tolerant gaits, workspace expansion, centimeter-level trajectory tracking accuracy, and significant load adaptability. We validate the framework’s effectiveness and practical application value through comprehensive hardware experiments across diverse multimodal scenarios.

Förderung

  • National Key R&D Program of China
  • National Natural Science Foundation of China
  • National Natural Science Foundation of China, Basic Research Program for Young Students
  • National Key Laboratory of Aerospace Mechanism
  • Fundamental Research Funds for the Central University

Zitieren

APA

Zhang, Z., Ding, L., Xu, P., Liu, T., Zhang, A., Yang, H., Gao, H., & Deng, Z. (2026). Legs-as-Arms: A Multitask Reinforcement Learning Framework for Simultaneous Loco-Manipulation in Hexapod Robots. IEEE Transactions on Industrial Electronics, 73(8), 11775-11786. https://doi.org/10.1109/tie.2026.3663691

Vancouver

Zhang Z, Ding L, Xu P, Liu T, Zhang A, Yang H, et al. Legs-as-Arms: A Multitask Reinforcement Learning Framework for Simultaneous Loco-Manipulation in Hexapod Robots. IEEE Transactions on Industrial Electronics. 2026;73(8):11775-11786. doi:10.1109/tie.2026.3663691

BibTeX
@article{Zhang2026,
  author = {Zhang, Zhelin and Ding, Liang and Xu, Peng and Liu, Tie and Zhang, Ao and Yang, Huaiguang and Gao, Haibo and Deng, Zongquan},
  title = {Legs-as-Arms: A Multitask Reinforcement Learning Framework for Simultaneous Loco-Manipulation in Hexapod Robots},
  journal = {IEEE Transactions on Industrial Electronics},
  year = {2026},
  volume = {73},
  number = {8},
  pages = {11775--11786},
  doi = {10.1109/tie.2026.3663691}
}
Kostenlos starten →