DiffAqua: A Differentiable Computational Design Pipeline for Soft Underwater Swimmers with Shape Interpolation

Pingchuan Ma, Tao Du, John Z. Zhang, Kui Wu, Andrew Spielberg, Robert K. Katzschmann, and Wojciech Matusik

SIGGRAPH 2021 [Project Page] [Paper] [Video]

@article{ma2021diffaqua,
  title={DiffAqua: A Differentiable Computational Design Pipeline for Soft Underwater Swimmers with Shape Interpolation},
  author={Ma, Pingchuan and Du, Tao and Zhang, John Z and Wu, Kui and Spielberg, Andrew and Katzschmann, Robert K and Matusik, Wojciech},
  journal={ACM Transactions on Graphics (TOG)},
  volume={40},
  number={4},
  pages={132},
  year={2021},
  publisher={ACM New York, NY, USA}
}

Get Started

Prerequistes

• Ubuntu 18.04 LTS

• CUDA 10.2

• Anaconda 2020.11

  wget -P /tmp https://repo.anaconda.com/archive/Anaconda3-2020.11-Linux-x86_64.sh
  sh /tmp/Anaconda3-2020.11-Linux-x86_64.sh

• Suitesparse

  # probably need `sudo`
  apt-get install libsuitesparse-dev

Installation

• Restore Anaconda environment

  conda env create -f environment.yml
  conda activate diffaqua

• Install diffpd

  cd /path/to/root/external/diffpd/
  pip install -e . -v

• Install diffaqua

  cd /path/to/root/
  pip install -e . -v

Run Experiment

• Open-loop co-optimization

  cd /path/to/root/example/
  python openloop.py

  More to come!

• Visualize the results using tensorboard

  cd /path/to/root/example/experiments/
  tensorboard --logdir . --port 8888

Contact

If you have any questions about the paper or the codebase, please feel free to contact pcma@csail.mit.edu.