Incorporates the physics of forward dynamics directly into a GNN architecture for articulated robot control.
March 20, 2026
Original Paper
Articulated-Body Dynamics Network: Dynamics-Grounded Prior for Robot Learning
arXiv · 2603.19078
The Takeaway
By mimicking the Articulated Body Algorithm's inertia propagation, this model provides a far more effective inductive bias than standard Transformers or GNNs for robotics. It demonstrates superior sample efficiency and robust sim-to-real transfer on advanced hardware like the Unitree G1 humanoid.
From the 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 archit