gradsim

gradSim: Differentiable simulation for system identification and visuomotor control

gradSim is a unified differentiable rendering and multiphysics framework that allows solving a range of control and parameter estimation tasks (rigid bodies, deformable solids, and cloth) directly from images/video. Our unified computation graph — spanning from the dynamics and through the rendering process — enables learning in challenging visuomotor control tasks, without relying on state-based (3D) supervision, while obtaining performance competitive to or better than techniques that rely on precise 3D labels.

This repo is currently a stub. To be notified when code is released, plese "subscribe" to this issue.

Citing gradSim

For attribution in academic contexts, please cite this work as

Citation

Jatavallabhula and Macklin et al., "gradSim: Differentiable simulation for system identification and visuomotor control", ICLR 2021.

BibTeX citation

@article{gradsim,
  title   = {gradSim: Differentiable simulation for system identification and visuomotor control},
  author  = {Krishna Murthy Jatavallabhula and Miles Macklin and Florian Golemo and Vikram Voleti and Linda Petrini and Martin Weiss and Breandan Considine and Jerome Parent-Levesque and Kevin Xie and Kenny Erleben and Liam Paull and Florian Shkurti and Derek Nowrouzezahrai and Sanja Fidler},
  journal = {International Conference on Learning Representations (ICLR)},
  year    = {2021},
  url     = {https://openreview.net/forum?id=c_E8kFWfhp0},
  pdf     = {https://openreview.net/pdf?id=c_E8kFWfhp0},
}