Detail výsledku
Communication Framework for 5G-Enabled Network Applications
Kapinus Michal, Ing., Ph.D., UPGM (FIT)
Dobeš Petr, Ing., UPGM (FIT)
Juránek Roman, Ing., Ph.D., UPGM (FIT)
Smrž Pavel, doc. RNDr., Ph.D., UPGM (FIT)
This paper presents the framework for easily deployable, universal network applications focusing on robotic tasks. The framework defines the unifying interface between the robot and the application. It could be deployed in three locations: on the cloud, edge device, or on the robot itself. This paper describes the interface in detail, followed by the reference implementation of such a network application for an object detection task. The reference implementation is evaluated under different conditions, locations, and connection methods. Results indicate that offloading computationally demanding tasks to the edge or cloud is feasible and significantly reduces power demands on the robot as CPU and RAM are less utilized. Moreover, a GPU, needed for acceleration of detection algorithms, might be omitted from the robot's configuration.
@inproceedings{BUT185167,
author="Zdeněk {Materna} and Michal {Kapinus} and Petr {Dobeš} and Roman {Juránek} and Pavel {Smrž}",
title="Communication Framework for 5G-Enabled Network Applications",
booktitle="EuCNC & 6G Summit Proceedings",
year="2023",
pages="1--5",
address="Gothenburg",
url="https://www.fit.vut.cz/research/publication/12998/"
}