Articulated-Body Dynamics Network: Dynamics-Grounded Prior for Robot Learning

TL;DR

This paper introduces ABD-Net, a graph neural network architecture grounded in articulated body dynamics, which improves policy learning efficiency and robustness for robot locomotion by incorporating dynamics-based inductive biases.

Contribution

The paper presents ABD-Net, a novel dynamics-grounded graph neural network architecture that systematically incorporates articulated body dynamics into policy learning for robots.

Findings

Increased sample efficiency in simulated robot tasks

Enhanced generalization to dynamics shifts

Successful sim-to-real transfer on real robots

Abstract

Recent work in reinforcement learning has shown that incorporating structural priors for articulated robots, such as link connectivity, into policy networks improves learning efficiency. However, dynamics properties, despite their fundamental role in determining how forces and motion propagate through the body, remain largely underexplored as an inductive bias for policy learning. To address this gap, we present the Articulated-Body Dynamics Network (ABD-Net), a novel graph neural network architecture grounded in the computational structure of forward dynamics. Specifically, we adapt the inertia propagation mechanism from the Articulated Body Algorithm, systematically aggregating inertial quantities from child to parent links in a tree-structured manner, while replacing physical quantities with learnable parameters. Embedding ABD-NET into the policy actor enables dynamics-informed representations that capture how actions propagate through the body, leading to efficient and robust policy learning. Through experiments with simulated humanoid, quadruped, and hopper robots, our approach demonstrates increased sample efficiency and generalization to dynamics shifts compared to transformer-based and GNN baselines. We further validate the learned policy on real Unitree G1 and Go2 robots, state-of-the-art humanoid and quadruped platforms, generating dynamic, versatile and robust locomotion behaviors through sim-to-real transfer with real-time inference.