Learning Data-Efficient Rigid-Body Contact Models: Case Study of Planar Impact

Nima Fazeli, Samuel Zapolsky, Evan Drumwright, Alberto Rodriguez
Proceedings of the 1st Annual Conference on Robot Learning, PMLR 78:388-397, 2017.

Abstract

In this paper we demonstrate the limitations of common rigid-body contact models used in the robotics community by comparing them to a collection of data-driven and data-reinforced models that exploit underlying structure inspired by the rigid contact paradigm. We evaluate and compare the analytical and data-driven contact models on an empirical planar impact data-set, and show that the learned models are able to outperform their analytical counterparts with a small training set.

Cite this Paper

BibTeX
@InProceedings{pmlr-v78-fazeli17a, title = {Learning Data-Efficient Rigid-Body Contact Models: Case Study of Planar Impact}, author = {Fazeli, Nima and Zapolsky, Samuel and Drumwright, Evan and Rodriguez, Alberto}, booktitle = {Proceedings of the 1st Annual Conference on Robot Learning}, pages = {388--397}, year = {2017}, editor = {Levine, Sergey and Vanhoucke, Vincent and Goldberg, Ken}, volume = {78}, series = {Proceedings of Machine Learning Research}, month = {13--15 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v78/fazeli17a/fazeli17a.pdf}, url = {https://proceedings.mlr.press/v78/fazeli17a.html}, abstract = {In this paper we demonstrate the limitations of common rigid-body contact models used in the robotics community by comparing them to a collection of data-driven and data-reinforced models that exploit underlying structure inspired by the rigid contact paradigm. We evaluate and compare the analytical and data-driven contact models on an empirical planar impact data-set, and show that the learned models are able to outperform their analytical counterparts with a small training set.} }
Endnote
%0 Conference Paper %T Learning Data-Efficient Rigid-Body Contact Models: Case Study of Planar Impact %A Nima Fazeli %A Samuel Zapolsky %A Evan Drumwright %A Alberto Rodriguez %B Proceedings of the 1st Annual Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2017 %E Sergey Levine %E Vincent Vanhoucke %E Ken Goldberg %F pmlr-v78-fazeli17a %I PMLR %P 388--397 %U https://proceedings.mlr.press/v78/fazeli17a.html %V 78 %X In this paper we demonstrate the limitations of common rigid-body contact models used in the robotics community by comparing them to a collection of data-driven and data-reinforced models that exploit underlying structure inspired by the rigid contact paradigm. We evaluate and compare the analytical and data-driven contact models on an empirical planar impact data-set, and show that the learned models are able to outperform their analytical counterparts with a small training set.
APA
Fazeli, N., Zapolsky, S., Drumwright, E. & Rodriguez, A.. (2017). Learning Data-Efficient Rigid-Body Contact Models: Case Study of Planar Impact. Proceedings of the 1st Annual Conference on Robot Learning, in Proceedings of Machine Learning Research 78:388-397 Available from https://proceedings.mlr.press/v78/fazeli17a.html.

Related Material