Local Bayesian optimization for controller tuning with crash constraints
Abstract: Controller tuning is crucial for closed-loop performance but often involves manual adjustments. Although Bayesian optimization (BO) has been established as a data-efficient method for automated tuning, applying it to large and high-dimensional search spaces remains challenging. We extend a recently proposed local variant of BO to include crash constraints, where the controller can only be successfully evaluated in an a-priori unknown feasible region. We demonstrate the efficiency of the proposed method through simulations and hardware experiments. Our findings showcase the potential of local BO to enhance controller performance and reduce the time and resources necessary for tuning.
Eine korrekte Reglerparametrierung ist entscheidend für die Güte des geschlossenen Regelkreis, erfordert jedoch häufig manuelle Anpassungen. Obwohl sich Bayes’sche Optimierung (BO) als dateneffiziente Methode für die automatische Einstellung etabliert hat, bleibt ihre Anwendbarkeit auf große und hochdimensionale Suchräume eine Herausforderung. Wir erweitern eine kürzlich vorgeschlagene lokale Variante von BO um Absturzbeschränkungen, bei denen der Regler nur in einem a-priori unbekannten Bereich erfolgreich evaluiert werden kann. Wir demonstrieren die Effizienz der vorgeschlagenen Methode durch Simulationen und Hardwareexperimente. Unsere Ergebnisse zeigen das Potenzial von lokaler BO, die Regelgüte zu verbessern und dabei sowohl Zeit als auch Ressourcen zu sparen.
- Weitere Titel
-
Lokale Bayes'sche Optimierung für die Reglereinstellung mit Absturzbeschränkungen
- Standort
-
Deutsche Nationalbibliothek Frankfurt am Main
- Umfang
-
Online-Ressource
- Sprache
-
Englisch
- Erschienen in
-
Local Bayesian optimization for controller tuning with crash constraints ; volume:72 ; number:4 ; year:2024 ; pages:281-292 ; extent:12
Automatisierungstechnik ; 72, Heft 4 (2024), 281-292 (gesamt 12)
- DOI
-
10.1515/auto-2023-0181
- URN
-
urn:nbn:de:101:1-2024040915371637464972
- Rechteinformation
-
Open Access; Der Zugriff auf das Objekt ist unbeschränkt möglich.
- Letzte Aktualisierung
-
14.08.2025, 10:50 MESZ
Datenpartner
Deutsche Nationalbibliothek. Bei Fragen zum Objekt wenden Sie sich bitte an den Datenpartner.
Ähnliche Objekte (12)
Local Bayesian optimization for controller tuning with crash constraints
Bayesian Optimization in Robot Learning - Automatic Controller Tuning and Sample-Efficient Methods
Bayesian optimization with safety constraints: safe and automatic parameter tuning in robotics
Basic controller tuning for large offshore wind turbines
Bayesian optimization for tuning chaotic systems
Parallelized Bayesian Optimization for Expensive Robot Controller Evolution
Fine-Tuning the Odds in Bayesian Networks
Synthesis Conditions for LPV Controller with Input Covariance Constraints
Bayesian coarsening: rapid tuning of polymer model parameters
Observer Based Controller Design for Linear Systems with Input Constraints
"Crash, Baby, Crash"
Optimization of floating offshore wind turbine platforms with a self-tuning controller
Local Bayesian optimization for controller tuning with crash constraints
Bayesian Optimization in Robot Learning - Automatic Controller Tuning and Sample-Efficient Methods
Bayesian optimization with safety constraints: safe and automatic parameter tuning in robotics
Basic controller tuning for large offshore wind turbines
Bayesian optimization for tuning chaotic systems
Parallelized Bayesian Optimization for Expensive Robot Controller Evolution
Fine-Tuning the Odds in Bayesian Networks
Synthesis Conditions for LPV Controller with Input Covariance Constraints
Bayesian coarsening: rapid tuning of polymer model parameters
Observer Based Controller Design for Linear Systems with Input Constraints
"Crash, Baby, Crash"
Optimization of floating offshore wind turbine platforms with a self-tuning controller
Local Bayesian optimization for controller tuning with crash constraints
Bayesian Optimization in Robot Learning - Automatic Controller Tuning and Sample-Efficient Methods
Bayesian optimization with safety constraints: safe and automatic parameter tuning in robotics
Basic controller tuning for large offshore wind turbines
Bayesian optimization for tuning chaotic systems
Parallelized Bayesian Optimization for Expensive Robot Controller Evolution
Fine-Tuning the Odds in Bayesian Networks
Synthesis Conditions for LPV Controller with Input Covariance Constraints
Bayesian coarsening: rapid tuning of polymer model parameters
Observer Based Controller Design for Linear Systems with Input Constraints
"Crash, Baby, Crash"