EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Kuramoto Object-Centric Reinforcement Learning for Robotic Manipulation Tasks

Leonid Ugadiarov, Aleksandr Panov

Technologies · 28.04.2026 · 14 (5) · 266

DOI: 10.3390/technologies14050266 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Technologies, online erschienen am 28.04.2026. Verfasst von Leonid Ugadiarov und 1 weiteren Autor:innen. Die Arbeit nennt 95 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministry of Economic Development of the Russian Federation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Model-based reinforcement learning (MBRL) is a promising approach for achieving high sample efficiency in learning control policies. The existing world models in MBRL typically represent the environment’s state as a single global latent vector. However, such representations limit the model’s ability to capture object interactions and reason about individual objects—capabilities that are critical for visual object-oriented tasks—and may lead to lower sample efficiency. To address this limitation, we propose Kuramoto Object-Centric Reinforcement Learning (KORL), a model-based agent that learns an object-centric world model. Our approach introduces a novel Kuramoto Slot Attention for Video (KSAVi) model that integrates Kuramoto oscillatory neurons with the Slot Attention module to robustly extract object representations. We design a world model that leverages these structured object-centric latents and predicts dynamics using graph neural networks, thereby incorporating an inductive bias for modeling object interactions. We evaluate KORL on a suite of visually diverse object-oriented robotic manipulation tasks and demonstrate that our method outperforms object-centric model-free and model-based approaches.

Förderung

  • Ministry of Economic Development of the Russian Federation

Zitieren

APA

Ugadiarov, L., & Panov, A. (2026). Kuramoto Object-Centric Reinforcement Learning for Robotic Manipulation Tasks. Technologies, 14(5), 266. https://doi.org/10.3390/technologies14050266

Vancouver

Ugadiarov L, Panov A. Kuramoto Object-Centric Reinforcement Learning for Robotic Manipulation Tasks. Technologies. 2026;14(5):266. doi:10.3390/technologies14050266

BibTeX
@article{Ugadiarov2026,
  author = {Ugadiarov, Leonid and Panov, Aleksandr},
  title = {Kuramoto Object-Centric Reinforcement Learning for Robotic Manipulation Tasks},
  journal = {Technologies},
  year = {2026},
  volume = {14},
  number = {5},
  pages = {266},
  doi = {10.3390/technologies14050266}
}
Kostenlos starten →