Subspace Graph Physics: Real-Time Rigid Body-Driven Granular Flow Simulation

TL;DR

This paper introduces a real-time, machine learning-based simulation method for rigid body-driven granular flows, combining continuum modeling, PCA for dimensionality reduction, and graph neural networks for efficient prediction, applicable to robotics in terrestrial and space environments.

Contribution

It develops a subspace machine learning simulation approach that significantly improves efficiency and enables real-time large-scale granular flow modeling using PCA and graph networks.

Findings

GNS predicts particle positions and forces with high accuracy.

PCA reduces training time and memory usage substantially.

Simulation runs 700 times faster than continuum methods.

Abstract

An important challenge in robotics is understanding the interactions between robots and deformable terrains that consist of granular material. Granular flows and their interactions with rigid bodies still pose several open questions. A promising direction for accurate, yet efficient, modeling is using continuum methods. Also, a new direction for real-time physics modeling is the use of deep learning. This research advances machine learning methods for modeling rigid body-driven granular flows, for application to terrestrial industrial machines as well as space robotics (where the effect of gravity is an important factor). In particular, this research considers the development of a subspace machine learning simulation approach. To generate training datasets, we utilize our high-fidelity continuum method, material point method (MPM). Principal component analysis (PCA) is used to reduce the dimensionality of data. We show that the first few principal components of our high-dimensional data keep almost the entire variance in data. A graph network simulator (GNS) is trained to learn the underlying subspace dynamics. The learned GNS is then able to predict particle positions and interaction forces with good accuracy. More importantly, PCA significantly enhances the time and memory efficiency of GNS in both training and rollout. This enables GNS to be trained using a single desktop GPU with moderate VRAM. This also makes the GNS real-time on large-scale 3D physics configurations (700x faster than our continuum method).