v Kernel manifold regression for the coupled electric drives dataset · Mirko Mazzoleni

Kernel manifold regression for the coupled electric drives dataset

M. Mazzoleni, M. Scandella, F. Previdi

Abstract

The aim of this work is to introduce the concept of manifold regularization to the identification of dynamic systems. The method has been tested on the coupled electric drives problem, using a purely black box approach in the framework of the Reproducing Kernel Hilbert Spaces (RKHS) [Presentation, Code]

Reference

M. Mazzoleni, G. Maroni, Y. Maccarana, S. Formentin and F. Previdi, "Fault detection in airliner electro-mechanical actuators via hybrid particle filtering", 3rd Nonlinear System Identification Benchmarks Workshop, Liege, Belgium. , Liege, Belgium, 2018.

Bibtex

Please cite the following papers:

Share this:

RELATED ARTICLES

Kernel-based system identification with manifold regularization: A Bayesian perspective
M. Mazzoleni, M. Scandella, F. Previdi
### This paper presents a nonparametric Bayesian interpretation of kernel-based function learning with manifold regularization. We show that manifold regularization corresponds to an additional likelihood term derived from noisy observations of the function gradient along the regressors graph. The hyperparameters of the method are estimated by ...
2022-04-30
Journal   
Piecewise nonlinear regression with data augmentation
M. Mazzoleni, M. Scandella, F. Previdi
### Piecewise regression represents a powerful tool to derive accurate yet modular models describing complex phenomena or physical systems. This paper presents an approach for learning PieceWise NonLinear (PWNL) functions in both a supervised and semi-supervised setting. We further equip the proposed technique with a method for the automatic ge...
2021-07-13
Conference   
KBERG: A MatLab toolbox for nonlinear kernel-based regularization and system identification
M. Mazzoleni, M. Scandella, F. Previdi
### We present KBERG, a MatLab package for nonlinear Kernel-BasEd ReGularization and system identification. The toolbox provides a complete environment for running experiments on simulated and experimental data from both static and dynamical systems. The whole identification procedure is supported: (i) data generation, (ii) excitation signals d...
2020-07-20
Conference