TechRxiv
double_Techrxiv.pdf (10.2 MB)
Download file

Variational Measurement Update for Extended Object Tracking Using Gaussian Processes

Download (10.2 MB)
preprint
posted on 20.02.2021, 18:33 by Murat KumruMurat Kumru, Hilal Köksal, Emre Özkan
We present an alternative inference framework for the Gaussian process-based extended object tracking (GPEOT) models. The method provides an approximate solution to the Bayesian filtering problem in GPEOT by relying on a new measurement update, which we derive using variational Bayes techniques. The resulting algorithm effectively computes approximate posterior densities of the kinematic and the extent states. We conduct various experiments on simulated and real data and examine the performance compared with a reference method, which employs an extended Kalman filter for inference. The proposed algorithm significantly improves the accuracy of both the kinematic and the extent estimates and proves robust against model uncertainties.

History

Email Address of Submitting Author

kumru@metu.edu.tr

ORCID of Submitting Author

https://orcid.org/0000-0003-2907-4559

Submitting Author's Institution

Middle East Technical University

Submitting Author's Country

Turkey