Learning Stable Nonlinear Dynamical Systems With Gaussian Mixture Models

被引：567

作者：

Khansari-Zadeh, S. Mohammad ^{[1
]}

Billard, Aude ^{[1
]}

机构：

[1] Ecole Polytech Fed Lausanne, Sch Engn, CH-1015 Lausanne, Switzerland

来源：

IEEE TRANSACTIONS ON ROBOTICS | 2011年 / 27卷 / 05期

关键词：

Dynamical systems (DS); Gaussian mixture model; imitation learning; point-to-point motions; stability analysis;

D O I：

10.1109/TRO.2011.2159412

中图分类号：

TP24 [机器人技术];

学科分类号：

080202 ; 1405 ;

摘要：

This paper presents a method to learn discrete robot motions from a set of demonstrations. We model a motion as a nonlinear autonomous (i.e., time-invariant) dynamical system (DS) and define sufficient conditions to ensure global asymptotic stability at the target. We propose a learning method, which is called Stable Estimator of Dynamical Systems (SEDS), to learn the parameters of the DS to ensure that all motions closely follow the demonstrations while ultimately reaching and stopping at the target. Time-invariance and global asymptotic stability at the target ensures that the system can respond immediately and appropriately to perturbations that are encountered during the motion. The method is evaluated through a set of robot experiments and on a library of human handwriting motions.

引用

页码：943 / 957

页数：15

共 50 条

[11] Gaussian Mixture Models with Rare Events [J].

Li, Xuetong ;

Zhou, Jing ;

Wang, Hansheng .

JOURNAL OF MACHINE LEARNING RESEARCH, 2024, 25 :1-40

[12] A Versatile Pilot Design Scheme for FDD Systems Utilizing Gaussian Mixture Models [J].

Turan, Nurettin ;

Boeck, Benedikt ;

Fesl, Benedikt ;

Joham, Michael ;

Gunduz, Deniz ;

Utschick, Wolfgang .

IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, 2025, 24 (05) :4115-4130

[13] Learning robot motions with stable dynamical systems under diffeomorphic transformations [J].

Neumann, Klaus ;

Steil, Jochen J. .

ROBOTICS AND AUTONOMOUS SYSTEMS, 2015, 70 :1-15

[14] Learning Lyapunov-Stable Polynomial Dynamical Systems Through Imitation [J].

Abyaneh, Amin ;

Lin, Hsiu-Chin .

CONFERENCE ON ROBOT LEARNING, VOL 229, 2023, 229

[15] Fast diffeomorphic matching to learn globally asymptotically stable nonlinear dynamical systems [J].

Perrin, Nicolas ;

Schlehuber-Caissier, Philipp .

SYSTEMS & CONTROL LETTERS, 2016, 96 :51-59

[16] The supervised learning Gaussian mixture model [J].

Jiyong Ma ;

Wen Gao .

Journal of Computer Science and Technology, 1998, 13 (5) :471-474

[17] The Supervised Learning Gaussian Mixture Model [J].

马继涌 ;

高文 .

Journal of Computer Science and Technology, 1998, (05) :471-474

[18] Fault detection and diagnosis for nonlinear systems: A new adaptive Gaussian mixture modeling approach [J].

Karami, Majid ;

Wang, Liping .

ENERGY AND BUILDINGS, 2018, 166 :477-488

[19] Machine Learning evaluation of microscopy image segmentation methods: The case of Gaussian Mixture Models [J].

Chatzigeorgiou, Manolis ;

Vrigkas, Michalis ;

Beazi-Katsioti, Margarita ;

Katsiotis, Marios ;

Boukos, Nikos ;

Constantoudis, Vassilios .

PROCEEDINGS OF THE 12TH HELLENIC CONFERENCE ON ARTIFICIAL INTELLIGENCE, SETN 2022, 2022,

[20] Machine Learning for Subgrade Moisture in Cold Regions: Random Forest and Gaussian Mixture Models [J].

Ahmed, Asif ;

Karimpour, Abolfazl .

GEOTECHNICAL AND GEOLOGICAL ENGINEERING, 2025, 43 (02)

← 1 2 3 4 5 →