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/.

翻译：本文提出了一种形态对称等变异质图神经网络（MS-HGNN），用于机器人动力学学习。该网络将机器人运动学结构与形态对称性整合至统一的图网络中，并将这些结构先验作为约束嵌入学习架构，从而确保模型具备高泛化能力、样本效率与模型效率。所提出的MS-HGNN是一种通用且灵活的架构，适用于多种多体动力学系统及广泛的动力学学习问题。我们通过理论证明了MS-HGNN的形态对称等变特性，并利用真实世界数据与仿真数据在多个四足机器人学习任务中验证了其有效性。相关代码已公开于 https://github.com/lunarlab-gatech/MorphSym-HGNN/。