Learning to Manipulate Amorphous Materials

TL;DR

This paper introduces a reinforcement learning approach to train controllers for manipulating amorphous materials like rice and honey in a physics simulator, enabling tasks such as spreading and flipping.

Contribution

It presents a novel method combining physics simulation and reinforcement learning to manipulate amorphous materials with neural network policies.

Findings

Successfully trained policies for spreading, gathering, and flipping materials

Demonstrated realistic manipulation animations using inverse kinematics

Applied position-based dynamics for accurate material simulation

Abstract

We present a method of training character manipulation of amorphous materials such as those often used in cooking. Common examples of amorphous materials include granular materials (salt, uncooked rice), fluids (honey), and visco-plastic materials (sticky rice, softened butter). A typical task is to spread a given material out across a flat surface using a tool such as a scraper or knife. We use reinforcement learning to train our controllers to manipulate materials in various ways. The training is performed in a physics simulator that uses position-based dynamics of particles to simulate the materials to be manipulated. The neural network control policy is given observations of the material (e.g. a low-resolution density map), and the policy outputs actions such as rotating and translating the knife. We demonstrate policies that have been successfully trained to carry out the following tasks: spreading, gathering, and flipping. We produce a final animation by using inverse kinematics to guide a character's arm and hand to match the motion of the manipulation tool such as a knife or a frying pan.