EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Language-Guided Dual-Mode Policy for Dual-Arm Manipulation

Jianghao Sun, Pengjun Mao iD, Yu Wang, Wenguang Guo

Sensors · 03.08.2026 · 26 (15) · 4883

DOI: 10.3390/s26154883 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 03.08.2026. Verfasst von Jianghao Sun und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Mechanical and Electrical Engineering, Henan University of Science and Technology, Luoyang 471000, China. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Henan Science and Technology Department, Key Science and Technology Program of Henan Province, China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Dual-arm robot manipulation tasks are highly complex, and different tasks can be categorized as dual-arm synchronous coordination or asynchronous sequential execution. Existing action-generative policies adopt a fully parameter-sharing single-network architecture in multi-task learning, overlooking the differences between these two task types in terms of temporal dependencies and action distributions, which leads to cross-mode interference. To address this limitation, we propose LGDM (Language-Guided Dual-Mode Policy), a language-guided dual-mode policy framework. Under shared language embeddings, it parallelly constructs dual-conditional branches for language routing and visual perception. The execution layer selects independent synchronous or asynchronous policy branches according to routing signals and outputs actions by combining them with perceptual features, thereby explicitly decoupling the two motion modes. Specifically, the model organizes three functional modules around language embeddings as a hub: (1) a language mode routing layer that predicts motion modes from temporal information in semantics and provides mode-selection signals to the execution layer; (2) a vision perception layer that injects semantic information into the visual backbone via FiLM, enabling task-aware dynamic feature modulation; and (3) a dual-mode execution layer that builds independent synchronous and asynchronous policy branches, selects the corresponding branch according to routing, and generates actions with the modulated features. On the RoboTwin 2 dual-arm manipulation benchmark comprising six synchronous and asynchronous tasks, LGDM outperforms existing baselines, achieving an absolute improvement of 14.5 percentage points over RDT, and maintains robust execution performance under unseen temporal-coordination instruction variants at inference time.

Schlagwörter

Stichwörter der Autor:innen

FiLMdual-arm robotdual-modelanguage-guided

Autor:innen und Institutionen

  • School of Mechanical and Electrical Engineering, Henan University of Science and Technology, Luoyang 471000, China.

ORCID-Profile: Pengjun Mao

Förderung

  • National Natural Science Foundation of China
  • Henan Science and Technology Department
  • Key Science and Technology Program of Henan Province, China
  • Postdoctoral Fellowship Program of CPSF

Zitieren

APA

Sun, J., Mao, P., Wang, Y., & Guo, W. (2026). Language-Guided Dual-Mode Policy for Dual-Arm Manipulation. Sensors, 26(15), 4883. https://doi.org/10.3390/s26154883

Vancouver

Sun J, Mao P, Wang Y, Guo W. Language-Guided Dual-Mode Policy for Dual-Arm Manipulation. Sensors. 2026;26(15):4883. doi:10.3390/s26154883

BibTeX
@article{Sun2026,
  author = {Sun, Jianghao and Mao, Pengjun and Wang, Yu and Guo, Wenguang},
  title = {Language-Guided Dual-Mode Policy for Dual-Arm Manipulation},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {15},
  pages = {4883},
  doi = {10.3390/s26154883}
}
Kostenlos starten →