Tuning Legged Locomotion Controllers via Safe Bayesian Optimization

Daniel Widmer, Dongho Kang, Bhavya Sukhija, Jonas Hübotter, Andreas Krause, Stelian Coros
Proceedings of The 7th Conference on Robot Learning, PMLR 229:2444-2464, 2023.

Abstract

This paper presents a data-driven strategy to streamline the deployment of model-based controllers in legged robotic hardware platforms. Our approach leverages a model-free safe learning algorithm to automate the tuning of control gains, addressing the mismatch between the simplified model used in the control formulation and the real system. This method substantially mitigates the risk of hazardous interactions with the robot by sample-efficiently optimizing parameters within a probably safe region. Additionally, we extend the applicability of our approach to incorporate the different gait parameters as contexts, leading to a safe, sample-efficient exploration algorithm capable of tuning a motion controller for diverse gait patterns. We validate our method through simulation and hardware experiments, where we demonstrate that the algorithm obtains superior performance on tuning a model-based motion controller for multiple gaits safely.

Cite this Paper

BibTeX
@InProceedings{pmlr-v229-widmer23a, title = {Tuning Legged Locomotion Controllers via Safe Bayesian Optimization}, author = {Widmer, Daniel and Kang, Dongho and Sukhija, Bhavya and H\"{u}botter, Jonas and Krause, Andreas and Coros, Stelian}, booktitle = {Proceedings of The 7th Conference on Robot Learning}, pages = {2444--2464}, year = {2023}, editor = {Tan, Jie and Toussaint, Marc and Darvish, Kourosh}, volume = {229}, series = {Proceedings of Machine Learning Research}, month = {06--09 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v229/widmer23a/widmer23a.pdf}, url = {https://proceedings.mlr.press/v229/widmer23a.html}, abstract = {This paper presents a data-driven strategy to streamline the deployment of model-based controllers in legged robotic hardware platforms. Our approach leverages a model-free safe learning algorithm to automate the tuning of control gains, addressing the mismatch between the simplified model used in the control formulation and the real system. This method substantially mitigates the risk of hazardous interactions with the robot by sample-efficiently optimizing parameters within a probably safe region. Additionally, we extend the applicability of our approach to incorporate the different gait parameters as contexts, leading to a safe, sample-efficient exploration algorithm capable of tuning a motion controller for diverse gait patterns. We validate our method through simulation and hardware experiments, where we demonstrate that the algorithm obtains superior performance on tuning a model-based motion controller for multiple gaits safely.} }
Endnote
%0 Conference Paper %T Tuning Legged Locomotion Controllers via Safe Bayesian Optimization %A Daniel Widmer %A Dongho Kang %A Bhavya Sukhija %A Jonas Hübotter %A Andreas Krause %A Stelian Coros %B Proceedings of The 7th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2023 %E Jie Tan %E Marc Toussaint %E Kourosh Darvish %F pmlr-v229-widmer23a %I PMLR %P 2444--2464 %U https://proceedings.mlr.press/v229/widmer23a.html %V 229 %X This paper presents a data-driven strategy to streamline the deployment of model-based controllers in legged robotic hardware platforms. Our approach leverages a model-free safe learning algorithm to automate the tuning of control gains, addressing the mismatch between the simplified model used in the control formulation and the real system. This method substantially mitigates the risk of hazardous interactions with the robot by sample-efficiently optimizing parameters within a probably safe region. Additionally, we extend the applicability of our approach to incorporate the different gait parameters as contexts, leading to a safe, sample-efficient exploration algorithm capable of tuning a motion controller for diverse gait patterns. We validate our method through simulation and hardware experiments, where we demonstrate that the algorithm obtains superior performance on tuning a model-based motion controller for multiple gaits safely.
APA
Widmer, D., Kang, D., Sukhija, B., Hübotter, J., Krause, A. & Coros, S.. (2023). Tuning Legged Locomotion Controllers via Safe Bayesian Optimization. Proceedings of The 7th Conference on Robot Learning, in Proceedings of Machine Learning Research 229:2444-2464 Available from https://proceedings.mlr.press/v229/widmer23a.html.

Related Material