English

Morphological-Symmetry-Equivariant Heterogeneous Graph Neural Network for Robotic Dynamics Learning

Robotics 2026-03-11 v3 Machine Learning

Abstract

We present a morphological-symmetry-equivariant heterogeneous graph neural network, namely MS-HGNN, for robotic dynamics learning, that integrates robotic kinematic structures and morphological symmetries into a single graph network. These structural priors are embedded into the learning architecture as constraints, ensuring high generalizability, sample and model efficiency. The proposed MS-HGNN is a versatile and general architecture that is applicable to various multi-body dynamic systems and a wide range of dynamics learning problems. We formally prove the morphological-symmetry-equivariant property of our MS-HGNN and validate its effectiveness across multiple quadruped robot learning problems using both real-world and simulated data. Our code is made publicly available at https://github.com/lunarlab-gatech/MorphSym-HGNN/.

Keywords

Cite

@article{arxiv.2412.01297,
  title  = {Morphological-Symmetry-Equivariant Heterogeneous Graph Neural Network for Robotic Dynamics Learning},
  author = {Fengze Xie and Sizhe Wei and Yue Song and Yisong Yue and Lu Gan},
  journal= {arXiv preprint arXiv:2412.01297},
  year   = {2026}
}
R2 v1 2026-06-28T20:19:23.865Z