Learning Stable Nonlinear Dynamical Systems With Gaussian Mixture Models

被引:571
作者
Khansari-Zadeh, S. Mohammad [1 ]
Billard, Aude [1 ]
机构
[1] Ecole Polytech Fed Lausanne, Sch Engn, CH-1015 Lausanne, Switzerland
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
收藏
页码:943 / 957
页数:15
相关论文
共 50 条
[41]   GPGPU Implementation of Variational Bayesian Gaussian Mixture Models [J].
Nishimoto, Hiroki ;
Nakada, Takashi ;
Nakashima, Yasuhiko .
2019 SEVENTH INTERNATIONAL SYMPOSIUM ON COMPUTING AND NETWORKING (CANDAR 2019), 2019, :185-190
[42]   A Gaussian mixture filter with adaptive refinement for nonlinear state estimation [J].
Zhang, Bin ;
Shin, Yung C. .
SIGNAL PROCESSING, 2022, 201
[43]   A Greedy Merge Learning Algorithm for Gaussian Mixture Model [J].
Li, Yan ;
Li, Lei .
2009 THIRD INTERNATIONAL SYMPOSIUM ON INTELLIGENT INFORMATION TECHNOLOGY APPLICATION, VOL 2, PROCEEDINGS, 2009, :506-+
[44]   Gaussian Mixture Model Learning for Multipath Assisted Positioning [J].
Karasek, Rostislav ;
Gentner, Christian .
2021 15TH EUROPEAN CONFERENCE ON ANTENNAS AND PROPAGATION (EUCAP), 2021,
[45]   Model Error Modeling for a Class of Multivariable Systems Utilizing Stochastic Embedding Approach with Gaussian Mixture Models [J].
Orellana, Rafael ;
Coronel, Maria ;
Carvajal, Rodrigo ;
Escarate, Pedro ;
Aguero, Juan C. .
SYMMETRY-BASEL, 2025, 17 (03)
[46]   On the Uncertainty Modelling for Linear Continuous-Time Systems Utilising Sampled Data and Gaussian Mixture Models [J].
Orellana, Rafael ;
Coronel, Maria ;
Carvajal, Rodrigo ;
Delgado, Ramon A. ;
Escarate, Pedro ;
Aguero, Juan C. .
IFAC PAPERSONLINE, 2021, 54 (07) :589-594
[47]   Gaussian-input Gaussian mixture model for representing density maps and atomic models [J].
Kawabata, Takeshi .
JOURNAL OF STRUCTURAL BIOLOGY, 2018, 203 (01) :1-16
[48]   Encrypted traffic classification based on Gaussian mixture models and Hidden Markov Models [J].
Yao, Zhongjiang ;
Ge, Jingguo ;
Wu, Yulei ;
Lin, Xiaosheng ;
He, Runkang ;
Ma, Yuxiang .
JOURNAL OF NETWORK AND COMPUTER APPLICATIONS, 2020, 166
[49]   A survey of feature selection methods for Gaussian mixture models and hidden Markov models [J].
Stephen Adams ;
Peter A. Beling .
Artificial Intelligence Review, 2019, 52 :1739-1779
[50]   A survey of feature selection methods for Gaussian mixture models and hidden Markov models [J].
Adams, Stephen ;
Beling, Peter A. .
ARTIFICIAL INTELLIGENCE REVIEW, 2019, 52 (03) :1739-1779