Saved in:
Bibliographic Details
Main Authors: Anurag, Kumar, Azizi, Kasra, Sorrentino, Francesco, Wan, Wenbin
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.04985
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913978784940032
author Anurag, Kumar
Azizi, Kasra
Sorrentino, Francesco
Wan, Wenbin
author_facet Anurag, Kumar
Azizi, Kasra
Sorrentino, Francesco
Wan, Wenbin
contents Accurate modeling is crucial in many engineering and scientific applications, yet obtaining a reliable process model for complex systems is often challenging. To address this challenge, we propose a novel framework, reservoir computing with unscented Kalman filtering (RCUKF), which integrates data-driven modeling via reservoir computing (RC) with Bayesian estimation through the unscented Kalman filter (UKF). The RC component learns the nonlinear system dynamics directly from data, serving as a surrogate process model in the UKF prediction step to generate state estimates in high-dimensional or chaotic regimes where nominal mathematical models may fail. Meanwhile, the UKF measurement update integrates real-time sensor data to correct potential drift in the data-driven model. We demonstrate RCUKF effectiveness on well-known benchmark problems and a real-time vehicle trajectory estimation task in a high-fidelity simulation environment.
format Preprint
id arxiv_https___arxiv_org_abs_2508_04985
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle RCUKF: Data-Driven Modeling Meets Bayesian Estimation
Anurag, Kumar
Azizi, Kasra
Sorrentino, Francesco
Wan, Wenbin
Machine Learning
Systems and Control
93E11, 68T07
I.2.6; I.5.1
Accurate modeling is crucial in many engineering and scientific applications, yet obtaining a reliable process model for complex systems is often challenging. To address this challenge, we propose a novel framework, reservoir computing with unscented Kalman filtering (RCUKF), which integrates data-driven modeling via reservoir computing (RC) with Bayesian estimation through the unscented Kalman filter (UKF). The RC component learns the nonlinear system dynamics directly from data, serving as a surrogate process model in the UKF prediction step to generate state estimates in high-dimensional or chaotic regimes where nominal mathematical models may fail. Meanwhile, the UKF measurement update integrates real-time sensor data to correct potential drift in the data-driven model. We demonstrate RCUKF effectiveness on well-known benchmark problems and a real-time vehicle trajectory estimation task in a high-fidelity simulation environment.
title RCUKF: Data-Driven Modeling Meets Bayesian Estimation
topic Machine Learning
Systems and Control
93E11, 68T07
I.2.6; I.5.1
url https://arxiv.org/abs/2508.04985