EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Vision–Language–Action Model for Agricultural Fresh Produce Handling with a Humanoid Robot

Kapalik Khanal, Biswash Khatiwada, Stephen Afrifa, Sawyer Exum, Fan Wu, Ramesh Bahadur Bist

AgriEngineering · 06.11.2026 · 8 (11) · 479

DOI: 10.3390/agriengineering8110479 ↗

Kurz gefasst

Fachartikel in der Zeitschrift AgriEngineering, online erschienen am 06.11.2026. Verfasst von Kapalik Khanal und 5 weiteren Autor:innen. Die Arbeit nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: North Carolina State University.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Reaching, grasping, and placing delicate fresh produce are among the most labor-intensive operations in agricultural production, yet most robotic solutions remain task-specific, single-crop systems that generalize poorly across the morphological diversity of fresh produce. This study investigates whether a large, pre-trained Vision–Language–Action (VLA) model can be adapted with only modest demonstration data to serve as a flexible language-conditioned produce-handling policy on a humanoid robotic platform. The NVIDIA GR00T N1.7-3B foundation model is fine-tuned for a standard 23-degree-of-freedom Unitree G1 humanoid robot fitted with two Dex3-1 three-finger dexterous hands, whose arm and hand joint streams form a 24-dimensional combined manipulation command space. This study also compared the π0.5 VLA model. Demonstrations were collected through a Meta Quest 3 extended-reality teleoperation pipeline augmented with a multi-label recording mode that embeds a natural-language goal in each episode, yielding 605 language-conditioned demonstrations (261,724 synchronized frames) across four produce categories (lemon, peach, pear, and tomato). A dual-camera design paired a wide-angle head camera for global scene context with a wrist-mounted camera for close-range grasp feedback. The results show that the fine-tuned policy achieved a 96.25% individual-trial success rate and 85% on mixed-scene trials requiring retrieval of the language-specified item. However, adapting the π0.5 VLA model to the same data achieved 68.75% individual-trial success. Furthermore, a controlled ablation isolated the role of close-range perception. Without wrist-camera input, the policy reached every target in single-item trials but failed to secure a single grasp, resulting in 0% pick-and-place success despite comparable training loss. Therefore, close-range wrist perception was a precondition for contact in this setting. Together, these results demonstrate that VLA fine-tuning can transform a pre-trained humanoid foundation model into a reliable, data-efficient policy for multi-category fresh produce handling. This work provides a practical step toward flexible humanoid robotic systems for agricultural pick-and-place, sorting, and postharvest automation.

Förderung

  • North Carolina State University

Zitieren

APA

Khanal, K., Khatiwada, B., Afrifa, S., Exum, S., Wu, F., & Bist, R. B. (2026). Vision–Language–Action Model for Agricultural Fresh Produce Handling with a Humanoid Robot. AgriEngineering, 8(11), 479. https://doi.org/10.3390/agriengineering8110479

Vancouver

Khanal K, Khatiwada B, Afrifa S, Exum S, Wu F, Bist RB. Vision–Language–Action Model for Agricultural Fresh Produce Handling with a Humanoid Robot. AgriEngineering. 2026;8(11):479. doi:10.3390/agriengineering8110479

BibTeX
@article{Khanal2026,
  author = {Khanal, Kapalik and Khatiwada, Biswash and Afrifa, Stephen and Exum, Sawyer and Wu, Fan and Bist, Ramesh Bahadur},
  title = {Vision–Language–Action Model for Agricultural Fresh Produce Handling with a Humanoid Robot},
  journal = {AgriEngineering},
  year = {2026},
  volume = {8},
  number = {11},
  pages = {479},
  doi = {10.3390/agriengineering8110479}
}
Kostenlos starten →