Kernel methods and gaussian processes for system identification and control: A road map on regularized kernel-based learning for control
The commonly adopted route to control a dynamic system and make it follow the desired
behavior consists of two steps. First, a model of the system is learned from input–output data …
behavior consists of two steps. First, a model of the system is learned from input–output data …
Kernel-based methods for Volterra series identification
Volterra series approximate a broad range of nonlinear systems. Their identification is
challenging due to the curse of dimensionality: the number of model parameters grows …
challenging due to the curse of dimensionality: the number of model parameters grows …
Deep prediction networks
A Dalla Libera, G Pillonetto - Neurocomputing, 2022 - Elsevier
The challenge for next generation system identification is to build new flexible models and
estimators able to simulate complex systems. This task is especially difficult in the nonlinear …
estimators able to simulate complex systems. This task is especially difficult in the nonlinear …
A data-efficient geometrically inspired polynomial kernel for robot inverse dynamic
A Dalla Libera, R Carli - IEEE Robotics and Automation Letters, 2019 - ieeexplore.ieee.org
In this letter, we introduce a novel data-driven inverse dynamics estimator based on
Gaussian Process Regression. Driven by the fact that the inverse dynamics can be …
Gaussian Process Regression. Driven by the fact that the inverse dynamics can be …
Two-stage transfer learning-based nonparametric system identification with Gaussian process regression
S Wang, Z Xu, M Chen, J Zhao, J Fang… - Computers & Chemical …, 2024 - Elsevier
Most system identification methods ignore correlations between different identification tasks
and do not make full use of historical models when identifying a new process. In this paper …
and do not make full use of historical models when identifying a new process. In this paper …
Model-based policy search using monte carlo gradient estimation with real systems application
In this article, we present a model-based reinforcement learning (MBRL) algorithm named
Monte Carlo probabilistic inference for learning control (MC-PILCO). This algorithm relies on …
Monte Carlo probabilistic inference for learning control (MC-PILCO). This algorithm relies on …
[HTML][HTML] An efficient method for generalised Wiener series estimation of nonlinear systems using Gaussian processes
J Massingham, O Nielsen, T Butlin - Mechanical Systems and Signal …, 2024 - Elsevier
Abstract System identification of dynamical systems aims to predict the output of a system for
a given input by inferring model details from data. This is particularly challenging for …
a given input by inferring model details from data. This is particularly challenging for …
Learning nonlinear systems via Volterra series and Hilbert-Schmidt operators
This article examines the application of regularization techniques and kernel methods in
addressing the task of learning nonlinear dynamical systems from input–output data. Our …
addressing the task of learning nonlinear dynamical systems from input–output data. Our …
Learning causal estimates of linear operators from noisy data
This article studies the identification problem of linear systems from a set of noisy input–
output trajectories. The problem is formulated and solved as a least-square regularized …
output trajectories. The problem is formulated and solved as a least-square regularized …
Embedding the Physics in Black-box Inverse Dynamics Identification: a Comparison Between Gaussian Processes and Neural Networks
In recent years, black-box estimators for robot inverse dynamics have drawn the attention of
the robotics community. This paper compares two recent black-box approaches that try to …
the robotics community. This paper compares two recent black-box approaches that try to …