An Efficient Connected Swarm Deployment via Deep Learning

Kiril Danilchenko; Michael Segal

doi:10.15439/2021F001

An Efficient Connected Swarm Deployment via Deep Learning

Kiril Danilchenko, Michael Segal

Department of Communication Systems Engineering

Research output: Contribution to journal › Article › peer-review

Abstract

In this paper, an unmanned aerial vehicles (UAVs) deployment framework based on machine learning is studied. It aims to maximize the sum of the weights of the ground users covered by UAVs while UAVs forming a connected communication graph. We focus on the case where the number of UAVs is not necessarily enough to cover all ground users. We develop an UAV Deployment Deep Neural network (UD-DNNet) as a UAV's deployment deep network method. Simulation results demonstrate that UDDNNet can serve as a computationally inexpensive replacement for traditionally expensive optimization algorithms in real-time tasks and outperform the state-of-the-art traditional algorithms.

Original language	English
Pages (from-to)	1-7
Number of pages	7
Journal	Annals of Computer Science and Information Systems
Volume	25
DOIs	https://doi.org/10.15439/2021F001
State	Published - 2 Sep 2021
Event	16th Conference on Computer Science and Intelligence Systems, FedCSIS 2021 - Virtual, Sofia, Bulgaria Duration: 2 Sep 2021 → 5 Sep 2021

ASJC Scopus subject areas

Artificial Intelligence
Computer Networks and Communications
Computer Science Applications
Hardware and Architecture
Information Systems
Information Systems and Management
Control and Optimization

Access to Document

10.15439/2021F001

https://dblp.org/db/conf/fedcsis/fedcsis2021.html#DanilchenkoS21

Cite this

@article{7865ca279642487581eca869d4b65d3b,

title = "An Efficient Connected Swarm Deployment via Deep Learning",

abstract = "In this paper, an unmanned aerial vehicles (UAVs) deployment framework based on machine learning is studied. It aims to maximize the sum of the weights of the ground users covered by UAVs while UAVs forming a connected communication graph. We focus on the case where the number of UAVs is not necessarily enough to cover all ground users. We develop an UAV Deployment Deep Neural network (UD-DNNet) as a UAV's deployment deep network method. Simulation results demonstrate that UDDNNet can serve as a computationally inexpensive replacement for traditionally expensive optimization algorithms in real-time tasks and outperform the state-of-the-art traditional algorithms. ",

author = "Kiril Danilchenko and Michael Segal",

note = "Funding Information: ACKNOWLEDGMENTS This research has been supported by a grant from Pazy Foundation. Publisher Copyright: {\textcopyright} 2021 Polish Information Processing Society.; 16th Conference on Computer Science and Intelligence Systems, FedCSIS 2021 ; Conference date: 02-09-2021 Through 05-09-2021",

year = "2021",

month = sep,

day = "2",

doi = "10.15439/2021F001",

language = "English",

volume = "25",

pages = "1--7",

journal = "Annals of Computer Science and Information Systems",

issn = "2300-5963",

publisher = "Polskie Towarzystwo Informatyczne",

}

TY - JOUR

T1 - An Efficient Connected Swarm Deployment via Deep Learning

AU - Danilchenko, Kiril

AU - Segal, Michael

PY - 2021/9/2

Y1 - 2021/9/2

N2 - In this paper, an unmanned aerial vehicles (UAVs) deployment framework based on machine learning is studied. It aims to maximize the sum of the weights of the ground users covered by UAVs while UAVs forming a connected communication graph. We focus on the case where the number of UAVs is not necessarily enough to cover all ground users. We develop an UAV Deployment Deep Neural network (UD-DNNet) as a UAV's deployment deep network method. Simulation results demonstrate that UDDNNet can serve as a computationally inexpensive replacement for traditionally expensive optimization algorithms in real-time tasks and outperform the state-of-the-art traditional algorithms.

AB - In this paper, an unmanned aerial vehicles (UAVs) deployment framework based on machine learning is studied. It aims to maximize the sum of the weights of the ground users covered by UAVs while UAVs forming a connected communication graph. We focus on the case where the number of UAVs is not necessarily enough to cover all ground users. We develop an UAV Deployment Deep Neural network (UD-DNNet) as a UAV's deployment deep network method. Simulation results demonstrate that UDDNNet can serve as a computationally inexpensive replacement for traditionally expensive optimization algorithms in real-time tasks and outperform the state-of-the-art traditional algorithms.

UR - http://www.scopus.com/inward/record.url?scp=85117796912&partnerID=8YFLogxK

U2 - 10.15439/2021F001

DO - 10.15439/2021F001

M3 - Article

SN - 2300-5963

VL - 25

SP - 1

EP - 7

JO - Annals of Computer Science and Information Systems

JF - Annals of Computer Science and Information Systems

T2 - 16th Conference on Computer Science and Intelligence Systems, FedCSIS 2021

Y2 - 2 September 2021 through 5 September 2021

ER -

An Efficient Connected Swarm Deployment via Deep Learning

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this