Physics-penalised Regularisation for Learning Dynamics Models with Contact

Gabriella Pizzuto, Michael Mistry
Proceedings of the 3rd Conference on Learning for Dynamics and Control, PMLR 144:611-622, 2021.

Abstract

Robotic systems, such as legged robots and manipulators, often handle states which involve ground impact or interaction with objects present in their surroundings; both of which are physically driven by contact. Dynamics model learning tends to focus on continuous motion, yielding poor results when deployed on real systems exposed to non-smooth frictional discontinuities. Inspired by a recent promising direction in machine learning, in this work we present a novel method for learning dynamics models undergoing contact by augmenting data-driven deep models with physics-penalised regularisation. Precisely, this paper conceptually formalises a novel framework for using an impenetrability component in the physics-based loss function directly within the learning objective of neural networks. Our results demonstrate that our method shows superior performance to using normal deep models for learning non-smooth dynamics models of robotic manipulators, strengthening their potential for deployment in contact-rich environments.

Cite this Paper

BibTeX
@InProceedings{pmlr-v144-pizzuto21a, title = {Physics-penalised Regularisation for Learning Dynamics Models with Contact}, author = {Pizzuto, Gabriella and Mistry, Michael}, booktitle = {Proceedings of the 3rd Conference on Learning for Dynamics and Control}, pages = {611--622}, year = {2021}, editor = {Jadbabaie, Ali and Lygeros, John and Pappas, George J. and A. Parrilo, Pablo and Recht, Benjamin and Tomlin, Claire J. and Zeilinger, Melanie N.}, volume = {144}, series = {Proceedings of Machine Learning Research}, month = {07 -- 08 June}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v144/pizzuto21a/pizzuto21a.pdf}, url = {https://proceedings.mlr.press/v144/pizzuto21a.html}, abstract = {Robotic systems, such as legged robots and manipulators, often handle states which involve ground impact or interaction with objects present in their surroundings; both of which are physically driven by contact. Dynamics model learning tends to focus on continuous motion, yielding poor results when deployed on real systems exposed to non-smooth frictional discontinuities. Inspired by a recent promising direction in machine learning, in this work we present a novel method for learning dynamics models undergoing contact by augmenting data-driven deep models with physics-penalised regularisation. Precisely, this paper conceptually formalises a novel framework for using an impenetrability component in the physics-based loss function directly within the learning objective of neural networks. Our results demonstrate that our method shows superior performance to using normal deep models for learning non-smooth dynamics models of robotic manipulators, strengthening their potential for deployment in contact-rich environments.} }
Endnote
%0 Conference Paper %T Physics-penalised Regularisation for Learning Dynamics Models with Contact %A Gabriella Pizzuto %A Michael Mistry %B Proceedings of the 3rd Conference on Learning for Dynamics and Control %C Proceedings of Machine Learning Research %D 2021 %E Ali Jadbabaie %E John Lygeros %E George J. Pappas %E Pablo A. Parrilo %E Benjamin Recht %E Claire J. Tomlin %E Melanie N. Zeilinger %F pmlr-v144-pizzuto21a %I PMLR %P 611--622 %U https://proceedings.mlr.press/v144/pizzuto21a.html %V 144 %X Robotic systems, such as legged robots and manipulators, often handle states which involve ground impact or interaction with objects present in their surroundings; both of which are physically driven by contact. Dynamics model learning tends to focus on continuous motion, yielding poor results when deployed on real systems exposed to non-smooth frictional discontinuities. Inspired by a recent promising direction in machine learning, in this work we present a novel method for learning dynamics models undergoing contact by augmenting data-driven deep models with physics-penalised regularisation. Precisely, this paper conceptually formalises a novel framework for using an impenetrability component in the physics-based loss function directly within the learning objective of neural networks. Our results demonstrate that our method shows superior performance to using normal deep models for learning non-smooth dynamics models of robotic manipulators, strengthening their potential for deployment in contact-rich environments.
APA
Pizzuto, G. & Mistry, M.. (2021). Physics-penalised Regularisation for Learning Dynamics Models with Contact. Proceedings of the 3rd Conference on Learning for Dynamics and Control, in Proceedings of Machine Learning Research 144:611-622 Available from https://proceedings.mlr.press/v144/pizzuto21a.html.

Related Material