Efficient Differentiable Simulation of Articulated Bodies

Yi-Ling Qiao, Junbang Liang, Vladlen Koltun, Ming C Lin
Proceedings of the 38th International Conference on Machine Learning, PMLR 139:8661-8671, 2021.

Abstract

We present a method for efficient differentiable simulation of articulated bodies. This enables integration of articulated body dynamics into deep learning frameworks, and gradient-based optimization of neural networks that operate on articulated bodies. We derive the gradients of the contact solver using spatial algebra and the adjoint method. Our approach is an order of magnitude faster than autodiff tools. By only saving the initial states throughout the simulation process, our method reduces memory requirements by two orders of magnitude. We demonstrate the utility of efficient differentiable dynamics for articulated bodies in a variety of applications. We show that reinforcement learning with articulated systems can be accelerated using gradients provided by our method. In applications to control and inverse problems, gradient-based optimization enabled by our work accelerates convergence by more than an order of magnitude.

Cite this Paper

BibTeX
@InProceedings{pmlr-v139-qiao21a, title = {Efficient Differentiable Simulation of Articulated Bodies}, author = {Qiao, Yi-Ling and Liang, Junbang and Koltun, Vladlen and Lin, Ming C}, booktitle = {Proceedings of the 38th International Conference on Machine Learning}, pages = {8661--8671}, year = {2021}, editor = {Meila, Marina and Zhang, Tong}, volume = {139}, series = {Proceedings of Machine Learning Research}, month = {18--24 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v139/qiao21a/qiao21a.pdf}, url = {https://proceedings.mlr.press/v139/qiao21a.html}, abstract = {We present a method for efficient differentiable simulation of articulated bodies. This enables integration of articulated body dynamics into deep learning frameworks, and gradient-based optimization of neural networks that operate on articulated bodies. We derive the gradients of the contact solver using spatial algebra and the adjoint method. Our approach is an order of magnitude faster than autodiff tools. By only saving the initial states throughout the simulation process, our method reduces memory requirements by two orders of magnitude. We demonstrate the utility of efficient differentiable dynamics for articulated bodies in a variety of applications. We show that reinforcement learning with articulated systems can be accelerated using gradients provided by our method. In applications to control and inverse problems, gradient-based optimization enabled by our work accelerates convergence by more than an order of magnitude.} }
Endnote
%0 Conference Paper %T Efficient Differentiable Simulation of Articulated Bodies %A Yi-Ling Qiao %A Junbang Liang %A Vladlen Koltun %A Ming C Lin %B Proceedings of the 38th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2021 %E Marina Meila %E Tong Zhang %F pmlr-v139-qiao21a %I PMLR %P 8661--8671 %U https://proceedings.mlr.press/v139/qiao21a.html %V 139 %X We present a method for efficient differentiable simulation of articulated bodies. This enables integration of articulated body dynamics into deep learning frameworks, and gradient-based optimization of neural networks that operate on articulated bodies. We derive the gradients of the contact solver using spatial algebra and the adjoint method. Our approach is an order of magnitude faster than autodiff tools. By only saving the initial states throughout the simulation process, our method reduces memory requirements by two orders of magnitude. We demonstrate the utility of efficient differentiable dynamics for articulated bodies in a variety of applications. We show that reinforcement learning with articulated systems can be accelerated using gradients provided by our method. In applications to control and inverse problems, gradient-based optimization enabled by our work accelerates convergence by more than an order of magnitude.
APA
Qiao, Y., Liang, J., Koltun, V. & Lin, M.C.. (2021). Efficient Differentiable Simulation of Articulated Bodies. Proceedings of the 38th International Conference on Machine Learning, in Proceedings of Machine Learning Research 139:8661-8671 Available from https://proceedings.mlr.press/v139/qiao21a.html.

Related Material