Hlavní stránka Info portálu ZČU

Research results and applications in robotics and data processing with an emphasis on fusing information from multiple sources are the areas that the IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI) will focus on presenting in September. Topics to be covered at this year's event include probability theory, Bayesian inference, machine learning, neural networks, artificial intelligence in robotics, cognitive systems, big data, biomedical applications, autonomous vehicles (land, sea, air), and Industry 4.0.

The conference will feature presentations by several Czech and international experts. However, there will also be educational programs, which can already be registered here. Among the topics of the tutorials, those interested will find, for example, the theory and application of multisensor fusion and integration. Each tutorial should cover one topic in detail and present the current state of knowledge. This will give participants a complete understanding of the current issues, the main schools of thought, and possible application areas.

The conference, whose tradition dates back to 1994, is regularly attended by authors of papers from all over the world, academia and industry. "The thematic focus has, of course, changed over the years, and we reflect current trends such as Industry 4.0, the Internet of Things, cyber-physical systems, or artificial intelligence in robotics," says Ondřej Straka from FAS on behalf of the organizers.

In this expert world of multi-sensor fusion and intelligent systems, the Faculty of Applied Sciences has already seen success in 2016. At that time, the specialist committee selected a paper by Jiří Aigl and Ondřej Straka entitled Covariance Intersection in Track-to-Track Fusion With Memory from 107 accepted papers. The paper theoretically analyzed the effect of information feedback in probabilistic estimation fusion tasks. The paper's main result was a comparison of the quality of the fused estimates that can be guaranteed in situations where the dependencies of the fused information are unknown.