We present an adaptive smoother for linear state-space models with unknown process and measurement noise covariances. The proposed method utilizes the variational Bayes technique to perform approximate inference. The resulting smoother is computationally efficient, easy to implement, and can be applied to high dimensional linear systems. The performance of the algorithm is illustrated on a target tracking example.
@article{arxiv.1412.5307,
title = {Approximate Bayesian Smoothing with Unknown Process and Measurement Noise Covariances},
author = {Tohid Ardeshiri and Emre Özkan and Umut Orguner and Fredrik Gustafsson},
journal= {arXiv preprint arXiv:1412.5307},
year = {2023}
}
Comments
Derivations for the smoother can found here: http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-120700