Morphological symmetry-aware generalized policy network for deep reinforcement learning
Frontiers in Robotics and AI · 13.05.2026 · 13
DOI: 10.3389/frobt.2026.1816301 ↗Kurz gefasst
Fachartikel in der Zeitschrift Frontiers in Robotics and AI, online erschienen am 13.05.2026. Verfasst von Ryo Hakoda und 6 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Industrial Science, The University of Tokyo, Tokyo, Japan. Die Arbeit nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Exploiting the morphological symmetry of robotic systems, such as humanoid and quadruped robots, is a promising direction for improving robot learning. In deep reinforcement learning (DRL) for robot control, prior studies have leveraged such symmetry to improve learning efficiency through data augmentation, equivariant multilayer perceptrons (EMLPs), and multi-agent reinforcement learning (MARL) formulations. However, DRL training is inherently unstable, as the data distribution strongly depends on exploration, which is driven by stochasticity in the environment. To address this issue, we propose a symmetry-assisted, general-purpose DRL framework for morphologically symmetric robots that enables stable and robust learning. The framework models the environment as a symmetric Markov decision process (MDP) and constructs a full-body policy from a single-sided base policy using symmetry operators. We further propose a symmetric PPO objective with a coupled importance-sampling ratio. This objective aligns the policy optimization process with the imposed symmetry and serves as a principled alternative to MAPPO-style multi-agent formulations. Experimental results demonstrate that the proposed method outperforms existing approaches on most symmetric tasks, while still maintaining performance comparable to or better than standard PPO on asymmetric tasks, where symmetry is less directly exploitable.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Institute of Industrial Science, The University of Tokyo, Tokyo, Japan.
- Engineering Faculty, Kyoto University of Advanced Science, Kyoto, Japan.
- Microsoft, Redmond, WA, United States.
Zitieren
APA
Hakoda, R., Liu, Y., Hwang, M., Sato, Y., Takamatsu, J., Ikeuchi, K., & Oishi, T. (2026). Morphological symmetry-aware generalized policy network for deep reinforcement learning. Frontiers in Robotics and AI, 13. https://doi.org/10.3389/frobt.2026.1816301
Vancouver
Hakoda R, Liu Y, Hwang M, Sato Y, Takamatsu J, Ikeuchi K, et al. Morphological symmetry-aware generalized policy network for deep reinforcement learning. Frontiers in Robotics and AI. 2026;13. doi:10.3389/frobt.2026.1816301
BibTeX
@article{Hakoda2026,
author = {Hakoda, Ryo and Liu, Yubin and Hwang, Matthew and Sato, Yoshihiro and Takamatsu, Jun and Ikeuchi, Katsushi and Oishi, Takeshi},
title = {Morphological symmetry-aware generalized policy network for deep reinforcement learning},
journal = {Frontiers in Robotics and AI},
year = {2026},
volume = {13},
doi = {10.3389/frobt.2026.1816301}
}
EUVIMED