Safe Bayesian Optimisation for Controller Design by Utilising the Parameter Space Approach

Lorenz Dörschel, David Stenger, Dirk Abel
Proceedings of the 3rd Conference on Learning for Dynamics and Control, PMLR 144:299-311, 2021.

Abstract

As control systems become more and more complex, the optimal tuning of control parameters using Bayesian Optimisation gained an increased interest of research in recent years. Safe Bayesian Optimisation, tries to prevent sampling of unsafe parametrizations and therefore allow parameter tuning in real world experiments. Usually this is achieved by approximating a safe set using probabilistic GPR-predictions. In contrast in this work, analytical knowledge about robustly stable parameter configurations is gained by the parameter space approach and then incorporated within the optimisation as constraint. Simulation results on a linear system with uncertain parameters show a significant performance gain compared to standard approaches. .

Cite this Paper

BibTeX
@InProceedings{pmlr-v144-dorschel21a, title = {Safe Bayesian Optimisation for Controller Design by Utilising the Parameter Space Approach}, author = {D\"orschel, Lorenz and Stenger, David and Abel, Dirk}, booktitle = {Proceedings of the 3rd Conference on Learning for Dynamics and Control}, pages = {299--311}, 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/dorschel21a/dorschel21a.pdf}, url = {https://proceedings.mlr.press/v144/dorschel21a.html}, abstract = {As control systems become more and more complex, the optimal tuning of control parameters using Bayesian Optimisation gained an increased interest of research in recent years. Safe Bayesian Optimisation, tries to prevent sampling of unsafe parametrizations and therefore allow parameter tuning in real world experiments. Usually this is achieved by approximating a safe set using probabilistic GPR-predictions. In contrast in this work, analytical knowledge about robustly stable parameter configurations is gained by the parameter space approach and then incorporated within the optimisation as constraint. Simulation results on a linear system with uncertain parameters show a significant performance gain compared to standard approaches. .} }
Endnote
%0 Conference Paper %T Safe Bayesian Optimisation for Controller Design by Utilising the Parameter Space Approach %A Lorenz Dörschel %A David Stenger %A Dirk Abel %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-dorschel21a %I PMLR %P 299--311 %U https://proceedings.mlr.press/v144/dorschel21a.html %V 144 %X As control systems become more and more complex, the optimal tuning of control parameters using Bayesian Optimisation gained an increased interest of research in recent years. Safe Bayesian Optimisation, tries to prevent sampling of unsafe parametrizations and therefore allow parameter tuning in real world experiments. Usually this is achieved by approximating a safe set using probabilistic GPR-predictions. In contrast in this work, analytical knowledge about robustly stable parameter configurations is gained by the parameter space approach and then incorporated within the optimisation as constraint. Simulation results on a linear system with uncertain parameters show a significant performance gain compared to standard approaches. .
APA
Dörschel, L., Stenger, D. & Abel, D.. (2021). Safe Bayesian Optimisation for Controller Design by Utilising the Parameter Space Approach. Proceedings of the 3rd Conference on Learning for Dynamics and Control, in Proceedings of Machine Learning Research 144:299-311 Available from https://proceedings.mlr.press/v144/dorschel21a.html.

Related Material