Our article “Manipulation of Granular Materials by Learning Particle Interactions” was just accepted to IEEE Robotics and Automation Letters.
In this work we propose to use a Graph Neural Network (GNN) surrogate model to learn the particle interactions of granular materials. We perform planning of manipulation trajectories with the learnt surrogate model to arrange the material into a desired configuration.
For more information you can also visit the project website.
The code is also available open source.