Coordinate ascent neural Kalman-MLE for state estimation

Bettina Hanlon, Angel Garcia Fernandez
11/3/2025
cs.LG

No official code found. Best implementation: berthanlon/CAN-Kalman-MLE (3 stars, last commit 2025-11-24).

Abstract

This paper presents a coordinate ascent algorithm to learn dynamic and measurement models in dynamic state estimation using maximum likelihood estimation in a supervised manner. In particular, the dynamic and measurement models are assumed to be Gaussian and the algorithm learns the neural network parameters that model the dynamic and measurement functions, and also the noise covariance matrices. The trained dynamic and measurement models are then used with a non-linear Kalman filter algorithm to estimate the state during the testing phase.

View on arXivView PDF

Code Implementations(1)

berthanlon/CAN-Kalman-MLECommunity65%

Repository containing code used in the two scenarios for the paper 'Coordinate Ascent Neural Kalman-MLE for state estimation'

30PythonJul 14, 202511 months ago

Similar Papers

A Novel Neural Filter to Improve Accuracy of Neural Network Models of Dynamic Systems

80% match2024

PINN-Obs: Physics-Informed Neural Network-Based Observer for Nonlinear Dynamical Systems

78% match2025

Machine-learning parameter tracking with partial state observation

77% match1 repo72023

Designing Robust Software Sensors for Nonlinear Systems via Neural Networks and Adaptive Sliding Mode Control

77% match2025

Learning Physics Informed Neural ODEs With Partial Measurements

77% match2024

Learning state and proposal dynamics in state-space models using differentiable particle filters and neural networks

76% match2024

Cite this paper

@article{hanlon2025coordinate,
  title  = {Coordinate ascent neural Kalman-MLE for state estimation},
  author = {Bettina Hanlon and Angel Garcia Fernandez},
  year   = {2025},
  eprint = {2511.01855},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2511.01855v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2511.01855.svg)](https://codeofpaper.com/paper/2511.01855)

Discussion