Stability analysis of learning feed-forward control

W.J.R. Velthuis; Theodorus J.A. de Vries; Pieter Schaak; Erik W. Gaal

doi:10.1016/S0005-1098(00)00107-2

Stability analysis of learning feed-forward control

W.J.R. Velthuis, Theodorus J.A. de Vries, Pieter Schaak, Erik W. Gaal

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

19 Citations (Scopus)

21 Downloads (Pure)

Abstract

In this paper, a learning control system is considered for motion systems that are subject to two types of disturbances; reproducible disturbances, that re-occur each run in the same way, and random disturbances. In motion systems, a large part of the disturbances appear to be reproducible. In the control system considered, the reproducible disturbances are compensated by a learning component consisting of a B-spline neural network that is operated in feed-forward. The paper presents an analysis of stability properties of the configuration in case of a linear process and second-order B-splines. The outcomes of the analysis are quantitative criteria for selection of the width of the B-splines, and of the learning rate, for which the system is guaranteed to be stable. These criteria facilitate the design of a learning feed-forward controller.

Original language	Undefined
Pages (from-to)	1889-1895
Number of pages	7
Journal	Automatica
Volume	36
Issue number	12
DOIs	https://doi.org/10.1016/S0005-1098(00)00107-2
Publication status	Published - 2000

Keywords

METIS-111793
Learning feed-forward control
Neural Networks
Adaptation
IR-74346
Stability analysis
Iterative learning control

Access to Document

10.1016/S0005-1098(00)00107-2

Cite this

@article{a306fb81116b448eaa5c62e034e3dbf7,

title = "Stability analysis of learning feed-forward control",

abstract = "In this paper, a learning control system is considered for motion systems that are subject to two types of disturbances; reproducible disturbances, that re-occur each run in the same way, and random disturbances. In motion systems, a large part of the disturbances appear to be reproducible. In the control system considered, the reproducible disturbances are compensated by a learning component consisting of a B-spline neural network that is operated in feed-forward. The paper presents an analysis of stability properties of the configuration in case of a linear process and second-order B-splines. The outcomes of the analysis are quantitative criteria for selection of the width of the B-splines, and of the learning rate, for which the system is guaranteed to be stable. These criteria facilitate the design of a learning feed-forward controller.",

keywords = "METIS-111793, Learning feed-forward control, Neural Networks, Adaptation, IR-74346, Stability analysis, Iterative learning control",

author = "W.J.R. Velthuis and {de Vries}, {Theodorus J.A.} and Pieter Schaak and Gaal, {Erik W.}",

year = "2000",

doi = "10.1016/S0005-1098(00)00107-2",

language = "Undefined",

volume = "36",

pages = "1889--1895",

journal = "Automatica",

issn = "0005-1098",

publisher = "Elsevier Ltd",

number = "12",

}

TY - JOUR

T1 - Stability analysis of learning feed-forward control

AU - Velthuis, W.J.R.

AU - de Vries, Theodorus J.A.

AU - Schaak, Pieter

AU - Gaal, Erik W.

PY - 2000

Y1 - 2000

N2 - In this paper, a learning control system is considered for motion systems that are subject to two types of disturbances; reproducible disturbances, that re-occur each run in the same way, and random disturbances. In motion systems, a large part of the disturbances appear to be reproducible. In the control system considered, the reproducible disturbances are compensated by a learning component consisting of a B-spline neural network that is operated in feed-forward. The paper presents an analysis of stability properties of the configuration in case of a linear process and second-order B-splines. The outcomes of the analysis are quantitative criteria for selection of the width of the B-splines, and of the learning rate, for which the system is guaranteed to be stable. These criteria facilitate the design of a learning feed-forward controller.

AB - In this paper, a learning control system is considered for motion systems that are subject to two types of disturbances; reproducible disturbances, that re-occur each run in the same way, and random disturbances. In motion systems, a large part of the disturbances appear to be reproducible. In the control system considered, the reproducible disturbances are compensated by a learning component consisting of a B-spline neural network that is operated in feed-forward. The paper presents an analysis of stability properties of the configuration in case of a linear process and second-order B-splines. The outcomes of the analysis are quantitative criteria for selection of the width of the B-splines, and of the learning rate, for which the system is guaranteed to be stable. These criteria facilitate the design of a learning feed-forward controller.

KW - METIS-111793

KW - Learning feed-forward control

KW - Neural Networks

KW - Adaptation

KW - IR-74346

KW - Stability analysis

KW - Iterative learning control

U2 - 10.1016/S0005-1098(00)00107-2

DO - 10.1016/S0005-1098(00)00107-2

M3 - Article

SN - 0005-1098

VL - 36

SP - 1889

EP - 1895

JO - Automatica

JF - Automatica

IS - 12

ER -