Pregled bibliografske jedinice broj: 1162919
A Novel Real-Time UAVs Disaster Management Framework
A Novel Real-Time UAVs Disaster Management Framework // 29th Tеlеcommunicаtions Forum (TELFOR 2021)
Beograd: TЕLЕCOMMUNICАTIONS SOCIЕTY – Belgrade ЕTF – School of Еlectrical Еngineering, University of Belgrade, АCАDЕMIC MIND - Belgrade, 2021. str. 1-4 (predavanje, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)
CROSBI ID: 1162919 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
A Novel Real-Time UAVs Disaster Management Framework
Autori
Lorincz, Josip ; Tahirovic, Adnan ; Risteska Stojkoska, Biljana
Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni
ISBN
978-1-6654-2584-1
Skup
29th Tеlеcommunicаtions Forum (TELFOR 2021)
Mjesto i datum
Beograd, Srbija, 23.11.2021. - 24.11.2021
Vrsta sudjelovanja
Predavanje
Vrsta recenzije
Međunarodna recenzija
Ključne riječi
holistic framework ; disaster sensing ; UAV ; deep learning ; architecture ; image processing ; CNN ; cloud ; wireless
Sažetak
The paper proposes a novel computing and networking framework that can be implemented for the realization of different disaster management applications or real-time surveillance. The framework is based on networks of unmanned aerial vehicles (UAVs) equipped with different sensors including cameras. The framework represents a holistic approach that exploits the distributed architecture of clusters of UAVs and cloud computing resources located on the ground. The proposed framework is characterized by the hierarchical organization among framework elements. In such a framework, each UAV is assumed to be fully autonomous and locally implements a state of-the-art deep learning algorithms for real-time route planning, obstacle avoidance and object detection on aerial images. The main operating modules of the proposed framework have been presented, with the emphasis on the improvements which the proposed framework can bring in terms of event detection time and accuracy, energy consumption and reliability of application in disaster management systems. The proposed framework can serve as the foundation for the development of more reliable, faster in terms of disaster event detection and energy-efficient disaster management systems based on UAV networks.
Izvorni jezik
Engleski
Znanstvena područja
Elektrotehnika, Računarstvo, Informacijske i komunikacijske znanosti