Controller tuning with Bayesian optimization and its acceleration: Concept and experimental validation. Issue 3 (9th May 2022)

• Record Type: Journal Article
• Authors: Fujimoto, Yusuke; Sato, Hiroki; Nagahara, Masaaki
• Is Part Of: Asian journal of control. Volume 25:Issue 3(2023)
• Page Start: 2408
• Page End: 2414
• Publication Date: 2022-05-09
• Subjects: Bayesian optimization -- data‐driven control; Automatic control -- Periodicals; Control theory -- Periodicals; 629.805
• DOI: 10.1002/asjc.2847
• Languages: English
• ISSNs: 1561-8625

Abstract

Abstract: This work discusses a data‐driven approach to controller parameter tuning based on Bayesian optimization. In particular, we propose to design the prior mean function based on a model of the plant. By encoding the information on the model, the optimization needs a much fewer iterations than standard approaches. The effectiveness of the proposed method is demonstrated with a practical experiment.