Randomized iterative feedback tuning for fast MIMO feedback design of a mechatronic system

Leontine Aarnoudse; Peter den Toom; Tom Oomen

doi:10.1016/j.conengprac.2024.106152

Randomized iterative feedback tuning for fast MIMO feedback design of a mechatronic system

Leontine Aarnoudse^*, Peter den Toom, Tom Oomen

^*Corresponding author for this work

Team Jan-Willem van Wingerden

Research output: Contribution to journal › Article › Scientific › peer-review

49 Downloads (Pure)

Abstract

Iterative feedback tuning (IFT) enables the tuning of feedback controllers using only measured data to obtain the gradient of a cost criterion. The aim of this paper is to reduce the required number of experiments for MIMO IFT. It is shown that, through a randomization technique, an unbiased gradient estimate can be obtained from a single dedicated experiment, regardless of the size of the MIMO system. The gradient estimate is used in a stochastic gradient descent algorithm. The approach is experimentally validated on a mechatronic system, showing a significantly reduced number of experiments compared to standard IFT.

Original language	English
Article number	106152
Number of pages	10
Journal	Control Engineering Practice
Volume	154
DOIs	https://doi.org/10.1016/j.conengprac.2024.106152
Publication status	Published - 2025

Keywords

Data-driven control
Gradient estimation
Iterative feedback tuning
MIMO systems

Access to Document

10.1016/j.conengprac.2024.106152

1-s2.0-S0967066124003113-mainFinal published version, 1.31 MBLicence: CC BY-NC-ND

Cite this

@article{777ba863a12a4fe8b542050ebc0e7c47,

title = "Randomized iterative feedback tuning for fast MIMO feedback design of a mechatronic system",

abstract = "Iterative feedback tuning (IFT) enables the tuning of feedback controllers using only measured data to obtain the gradient of a cost criterion. The aim of this paper is to reduce the required number of experiments for MIMO IFT. It is shown that, through a randomization technique, an unbiased gradient estimate can be obtained from a single dedicated experiment, regardless of the size of the MIMO system. The gradient estimate is used in a stochastic gradient descent algorithm. The approach is experimentally validated on a mechatronic system, showing a significantly reduced number of experiments compared to standard IFT.",

keywords = "Data-driven control, Gradient estimation, Iterative feedback tuning, MIMO systems",

author = "Leontine Aarnoudse and {den Toom}, Peter and Tom Oomen",

year = "2025",

doi = "10.1016/j.conengprac.2024.106152",

language = "English",

volume = "154",

journal = "Control Engineering Practice",

issn = "0967-0661",

publisher = "Elsevier",

}

TY - JOUR

T1 - Randomized iterative feedback tuning for fast MIMO feedback design of a mechatronic system

AU - Aarnoudse, Leontine

AU - den Toom, Peter

AU - Oomen, Tom

PY - 2025

Y1 - 2025

N2 - Iterative feedback tuning (IFT) enables the tuning of feedback controllers using only measured data to obtain the gradient of a cost criterion. The aim of this paper is to reduce the required number of experiments for MIMO IFT. It is shown that, through a randomization technique, an unbiased gradient estimate can be obtained from a single dedicated experiment, regardless of the size of the MIMO system. The gradient estimate is used in a stochastic gradient descent algorithm. The approach is experimentally validated on a mechatronic system, showing a significantly reduced number of experiments compared to standard IFT.

AB - Iterative feedback tuning (IFT) enables the tuning of feedback controllers using only measured data to obtain the gradient of a cost criterion. The aim of this paper is to reduce the required number of experiments for MIMO IFT. It is shown that, through a randomization technique, an unbiased gradient estimate can be obtained from a single dedicated experiment, regardless of the size of the MIMO system. The gradient estimate is used in a stochastic gradient descent algorithm. The approach is experimentally validated on a mechatronic system, showing a significantly reduced number of experiments compared to standard IFT.

KW - Data-driven control

KW - Gradient estimation

KW - Iterative feedback tuning

KW - MIMO systems

UR - http://www.scopus.com/inward/record.url?scp=85208178063&partnerID=8YFLogxK

U2 - 10.1016/j.conengprac.2024.106152

DO - 10.1016/j.conengprac.2024.106152

M3 - Article

AN - SCOPUS:85208178063

SN - 0967-0661

VL - 154

JO - Control Engineering Practice

JF - Control Engineering Practice

M1 - 106152

ER -

Randomized iterative feedback tuning for fast MIMO feedback design of a mechatronic system

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this