/diffsim

Scalable Differentiable Physics for Learning and Control (ICML2020)

Primary LanguageC++

Scalable Differentiable Physics for Learning and Control

Yi-Ling Qiao, Junbang Liang, Vladlen Koltun, Ming C. Lin

Differentiable programming in Taichi allows you to optimize neural network controllers efficiently with brute-force gradient descent, instead of using reinforcement learning.

Usage

  1. Create a conda virtual environment and activate it.
conda create -n diffsim python=3.6 -y
conda activate diffsim
  1. Download and build the project.
git clone git@github.com:YilingQiao/diffsim.git
cd diffsim
bash script_build.sh
cd pysim
  1. Run the examples

Learn to drag a cube using a cloth

python exp_learn_cloth.py

Learn to hold a rigid body using a parallel gripper

python exp_learn_stick.py

Bibtex

@aritical{Qiao2020Scalable,
author  = {Qiao, Yiling and Liang, Junbang and Koltun, Vladlen and Lin, Ming C.},
title  = {Scalable Differentiable Physics for Learning and Control},
booktitle = {ICML},
year  = {2020},
}