Prof. Dr.-Ing. Thomas Kirste, holder of the chair Mobile Multimedia Information Systems (Institute for Visual & Analytic Computing, University of Rostock), will give a presentation on the topic
"On designing a factor library for probabilistic inference in hybrid complex dynamic systems"
as part of the VAC Colloquium.
Afterwards we look forward to discussions while enjoying coffee and cookies.
The event is open to anyone interested.
Abstract:
Estimating state from sensor data is pivotal in creating digital twins of real world entities. One option to achieve this is to use methods of probabilistic sequential state estimation. It turns out that this is not as straightforward as it sounds in the case of hybrid systems that integrate nontrivial discrete and continuous dynamics. In this talk, I will discuss some of the challenges that have to be addressed when designing a library of factor representations and associated operations facilitating the construction of Bayesian filters for such complex dynamic systems.