Closed-loop Model Selection for Kernel-based Models using Bayesian Optimization

Thomas Beckers, Somil Bansal, Claire J. Tomlin, Sandra Hirche

Publikation: Beitrag in Buch/Bericht/KonferenzbandKonferenzbeitragBegutachtung

4 Zitate (Scopus)

Abstract

Kernel-based nonparametric models have become very attractive for model-based control approaches for nonlinear systems. However, the selection of the kernel and its hyperparameters strongly influences the quality of the learned model. Classically, these hyperparameters are optimized to minimize the prediction error of the model but this process totally neglects its later usage in the control loop. In this work, we present a framework to optimize the kernel and hyperparameters of a kernel-based model directly with respect to the closed-loop performance of the model. Our framework uses Bayesian optimization to iteratively refine the kernel-based model using the observed performance on the actual system until a desired performance is achieved. We demonstrate the proposed approach in a simulation and on a 3-DoF robotic arm.

OriginalspracheEnglisch
Titel2019 IEEE 58th Conference on Decision and Control, CDC 2019
Herausgeber (Verlag)Institute of Electrical and Electronics Engineers Inc.
Seiten828-834
Seitenumfang7
ISBN (elektronisch)9781728113982
DOIs
PublikationsstatusVeröffentlicht - Dez. 2019
Veranstaltung58th IEEE Conference on Decision and Control, CDC 2019 - Nice, Frankreich
Dauer: 11 Dez. 201913 Dez. 2019

Publikationsreihe

NameProceedings of the IEEE Conference on Decision and Control
Band2019-December
ISSN (Print)0743-1546
ISSN (elektronisch)2576-2370

Konferenz

Konferenz58th IEEE Conference on Decision and Control, CDC 2019
Land/GebietFrankreich
OrtNice
Zeitraum11/12/1913/12/19

Fingerprint

Untersuchen Sie die Forschungsthemen von „Closed-loop Model Selection for Kernel-based Models using Bayesian Optimization“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren