Verifiable Error Bounds for Physics-Informed Neural KKL Observers

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Berin-Costain, Hannah, Wang, Harry, Morris, Kirsten, Liu, Jun
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908903321632768
author Berin-Costain, Hannah
Wang, Harry
Morris, Kirsten
Liu, Jun
author_facet Berin-Costain, Hannah
Wang, Harry
Morris, Kirsten
Liu, Jun
contents This paper proposes a computable state-estimation error bound for learning-based Kazantzis--Kravaris/Luenberger (KKL) observers. Recent work learns the KKL transformation map with a physics-informed neural network (PINN) and a corresponding left-inverse map with a conventional neural network. However, no computable state-estimation error bounds are currently available for this approach. We derive a state-estimation error bound that depends only on quantities that can be certified over a prescribed region using neural network verification. We further extend the result to bounded additive measurement noise and demonstrate the guarantees on nonlinear benchmark systems.
format Preprint
id arxiv_https___arxiv_org_abs_2603_20434
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Verifiable Error Bounds for Physics-Informed Neural KKL Observers
Berin-Costain, Hannah
Wang, Harry
Morris, Kirsten
Liu, Jun
Systems and Control
Machine Learning
This paper proposes a computable state-estimation error bound for learning-based Kazantzis--Kravaris/Luenberger (KKL) observers. Recent work learns the KKL transformation map with a physics-informed neural network (PINN) and a corresponding left-inverse map with a conventional neural network. However, no computable state-estimation error bounds are currently available for this approach. We derive a state-estimation error bound that depends only on quantities that can be certified over a prescribed region using neural network verification. We further extend the result to bounded additive measurement noise and demonstrate the guarantees on nonlinear benchmark systems.
title Verifiable Error Bounds for Physics-Informed Neural KKL Observers
topic Systems and Control
Machine Learning
url https://arxiv.org/abs/2603.20434