EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Safe Reinforcement Learning for Arm Manipulation with Constrained Markov Decision Process

Patrick Adjei, Norman Tasfi, Santiago Gomez-Rosero, Miriam A. M. Capretz

Robotics · 18.04.2024 · 13 (4) · 63

DOI: 10.3390/robotics13040063 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Robotics, online erschienen am 18.04.2024. Verfasst von Patrick Adjei und 3 weiteren Autor:innen. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Sciences and Engineering Research Council.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In the world of human–robot coexistence, ensuring safe interactions is crucial. Traditional logic-based methods often lack the intuition required for robots, particularly in complex environments where these methods fail to account for all possible scenarios. Reinforcement learning has shown promise in robotics due to its superior adaptability over traditional logic. However, the exploratory nature of reinforcement learning can jeopardize safety. This paper addresses the challenges in planning trajectories for robotic arm manipulators in dynamic environments. In addition, this paper highlights the pitfalls of multiple reward compositions that are susceptible to reward hacking. A novel method with a simplified reward and constraint formulation is proposed. This enables the robot arm to avoid a nonstationary obstacle that never resets, enhancing operational safety. The proposed approach combines scalarized expected returns with a constrained Markov decision process through a Lagrange multiplier, resulting in better performance. The scalarization component uses the indicator cost function value, directly sampled from the replay buffer, as an additional scaling factor. This method is particularly effective in dynamic environments where conditions change continually, as opposed to approaches relying solely on the expected cost scaled by a Lagrange multiplier.

Förderung

  • Natural Sciences and Engineering Research Council

Zitieren

APA

Adjei, P., Tasfi, N., Gomez-Rosero, S., & Capretz, M. A. M. (2024). Safe Reinforcement Learning for Arm Manipulation with Constrained Markov Decision Process. Robotics, 13(4), 63. https://doi.org/10.3390/robotics13040063

Vancouver

Adjei P, Tasfi N, Gomez-Rosero S, Capretz MAM. Safe Reinforcement Learning for Arm Manipulation with Constrained Markov Decision Process. Robotics. 2024;13(4):63. doi:10.3390/robotics13040063

BibTeX
@article{Adjei2024,
  author = {Adjei, Patrick and Tasfi, Norman and Gomez-Rosero, Santiago and Capretz, Miriam A. M.},
  title = {Safe Reinforcement Learning for Arm Manipulation with Constrained Markov Decision Process},
  journal = {Robotics},
  year = {2024},
  volume = {13},
  number = {4},
  pages = {63},
  doi = {10.3390/robotics13040063}
}
Kostenlos starten →