Learning Particle Dynamics for Manipulating Rigid Bodies, Deformable Objects, and Fluids

TL;DR

This paper introduces a learned particle-based simulator that models complex physical interactions for diverse materials, enabling robots to perform manipulation tasks involving fluids and deformable objects in simulation and real-world settings.

Contribution

The paper presents a novel approach combining learning with particle-based simulation, allowing rapid adaptation to unknown dynamics and effective manipulation of complex materials.

Findings

Robots successfully manipulated fluids and deformable objects using the learned simulator.

The approach demonstrated effective transfer from simulation to real-world tasks.

The learned model adapts quickly to new environments with minimal observations.

Abstract

Real-life control tasks involve matters of various substances---rigid or soft bodies, liquid, gas---each with distinct physical behaviors. This poses challenges to traditional rigid-body physics engines. Particle-based simulators have been developed to model the dynamics of these complex scenes; however, relying on approximation techniques, their simulation often deviates from real-world physics, especially in the long term. In this paper, we propose to learn a particle-based simulator for complex control tasks. Combining learning with particle-based systems brings in two major benefits: first, the learned simulator, just like other particle-based systems, acts widely on objects of different materials; second, the particle-based representation poses strong inductive bias for learning: particles of the same type have the same dynamics within. This enables the model to quickly adapt to new environments of unknown dynamics within a few observations. We demonstrate robots achieving complex manipulation tasks using the learned simulator, such as manipulating fluids and deformable foam, with experiments both in simulation and in the real world. Our study helps lay the foundation for robot learning of dynamic scenes with particle-based representations.