Metody bayesovského sledování jsou klíčovou metodologií v širokém spektru aplikací včetně autonomního řízení, kontroly vzdušného provozu, zabezpečovacích systémů, logistiky, zemědělství, záchranářství a monitorování životního prostředí. Tento projekt se zabývá pokročilými metodami pro dvě specifické aplikace: medicínské zobrazování a mobilní komunikace. Tyti aplikace jsou v současnosti předmětem společného výzkumu týmů na Vysokém učení technickém v Brně a na Technické univerzitě Vídeň. Prostředky z navrhovaného projektu významně podpoří a prohloubí spolupráci a transfer vědomostí mezi těmito dvěma institucemi.

Description in English
The proposed project aims at achieving substantial improvements in medical imaging and mobile communications through the use of two advanced Bayesian methodologies, namely, RFS-based multi-object tracking and GPR. For medical imaging, general project goals are to model the structural properties of IP evolution, to learn these structural properties from measurement data, and to exploit them for improved IP tracking. More concretely, we aim to model arterial motion analysis as a multi-object tracking problem and develop suitable multi-object tracking algorithms. For mobile communications, our general goal is to build an accurate RSS map of mobile cellular users. More concretely, we aim to develop methods for RSS estimation that operate time-recursively, have moderate complexity and good scaling properties, and exploit the spatial and temporal correlations of the physical radio channel in an effective manner. At a general level, the most essential goal of the project is to strengthen and deepen the research cooperation between BUT and TUW through a frequent and extended exchange of researchers. The project has a strong focus on young researchers (at the Master's and PhD levels) and female researchers (the participants at both institutions comprise three female researchers). We will report our results in at least two joint publications in top-quality international journals and two joint publications in the proceedings of high-level international conferences. This will enhance the scientific standing and visibility of the partitioning researchers and institutions. To guarantee a continuation of our collaborative research also beyond the duration of the project, we plan to apply for research projects at the national and international levels, including EU H2020 projects.

Keywords
bayesovská analýza, víceobjektové sledování, gaussovská regrese, 5G, mobilní sítě, medicínské zobrazování

Key words in English
Bayesian analysis, multi-object tracking, Gaussian process regression, 5G, mobile networks, medical imaging