Distributed Time-Varying Gaussian Regression via Kalman Filtering

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Taddei, Nicola, Maggioni, Riccardo, Eising, Jaap, De Pasquale, Giulia, Dorfler, Florian
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910915415244800
author Taddei, Nicola
Maggioni, Riccardo
Eising, Jaap
De Pasquale, Giulia
Dorfler, Florian
author_facet Taddei, Nicola
Maggioni, Riccardo
Eising, Jaap
De Pasquale, Giulia
Dorfler, Florian
contents We consider the problem of learning time-varying functions in a distributed fashion, where agents collect local information to collaboratively achieve a shared estimate. This task is particularly relevant in control applications, whenever real-time and robust estimation of dynamic cost/reward functions in safety critical settings has to be performed. In this paper, we,adopt a finite-dimensional approximation of a Gaussian Process, corresponding to a Bayesian linear regression in an appropriate feature space, and propose a new algorithm, DistKP, to track the time-varying coefficients via a distributed Kalman filter. The proposed method works for arbitrary kernels and under weaker assumptions on the time-evolution of the function to learn compared to the literature. We validate our results using a simulation example in which a fleet of Unmanned Aerial Vehicles (UAVs) learns a dynamically changing wind field.
format Preprint
id arxiv_https___arxiv_org_abs_2504_14900
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Distributed Time-Varying Gaussian Regression via Kalman Filtering
Taddei, Nicola
Maggioni, Riccardo
Eising, Jaap
De Pasquale, Giulia
Dorfler, Florian
Systems and Control
We consider the problem of learning time-varying functions in a distributed fashion, where agents collect local information to collaboratively achieve a shared estimate. This task is particularly relevant in control applications, whenever real-time and robust estimation of dynamic cost/reward functions in safety critical settings has to be performed. In this paper, we,adopt a finite-dimensional approximation of a Gaussian Process, corresponding to a Bayesian linear regression in an appropriate feature space, and propose a new algorithm, DistKP, to track the time-varying coefficients via a distributed Kalman filter. The proposed method works for arbitrary kernels and under weaker assumptions on the time-evolution of the function to learn compared to the literature. We validate our results using a simulation example in which a fleet of Unmanned Aerial Vehicles (UAVs) learns a dynamically changing wind field.
title Distributed Time-Varying Gaussian Regression via Kalman Filtering
topic Systems and Control
url https://arxiv.org/abs/2504.14900