Home > Research > Publications & Outputs > Joint Communication Scheduling and Velocity Con...

Research

Links

Text available via DOI:

View graph of relations

Joint Communication Scheduling and Velocity Control in Multi-UAV-Assisted Sensor Networks: A Deep Reinforcement Learning Approach

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Published

Standard

Joint Communication Scheduling and Velocity Control in Multi-UAV-Assisted Sensor Networks: A Deep Reinforcement Learning Approach. / Emami, Yousef; Wei, Bo; Li, Kai et al.
In: IEEE Transactions on Vehicular Technology, Vol. 70, No. 10, 31.10.2021, p. 10986-10998.

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Harvard

Emami, Y, Wei, B, Li, K, Ni, W & Tovar, E 2021, 'Joint Communication Scheduling and Velocity Control in Multi-UAV-Assisted Sensor Networks: A Deep Reinforcement Learning Approach', IEEE Transactions on Vehicular Technology, vol. 70, no. 10, pp. 10986-10998. https://doi.org/10.1109/TVT.2021.3110801

APA

Emami, Y., Wei, B., Li, K., Ni, W., & Tovar, E. (2021). Joint Communication Scheduling and Velocity Control in Multi-UAV-Assisted Sensor Networks: A Deep Reinforcement Learning Approach. IEEE Transactions on Vehicular Technology, 70(10), 10986-10998. https://doi.org/10.1109/TVT.2021.3110801

Vancouver

Emami Y, Wei B, Li K, Ni W, Tovar E. Joint Communication Scheduling and Velocity Control in Multi-UAV-Assisted Sensor Networks: A Deep Reinforcement Learning Approach. IEEE Transactions on Vehicular Technology. 2021 Oct 31;70(10):10986-10998. Epub 2021 Sept 8. doi: 10.1109/TVT.2021.3110801

Author

Emami, Yousef ; Wei, Bo ; Li, Kai et al. / Joint Communication Scheduling and Velocity Control in Multi-UAV-Assisted Sensor Networks : A Deep Reinforcement Learning Approach. In: IEEE Transactions on Vehicular Technology. 2021 ; Vol. 70, No. 10. pp. 10986-10998.

Bibtex

@article{6f2264b3e4bb40baa290577d3b3a77b2,
title = "Joint Communication Scheduling and Velocity Control in Multi-UAV-Assisted Sensor Networks: A Deep Reinforcement Learning Approach",
abstract = "Recently, Unmanned Aerial Vehicle (UAV) swarm has been increasingly studied to collect data from ground sensors in remote and hostile areas. A key challenge is the joint design of the velocities and data collection schedules of the UAVs, as inadequate velocities and schedules would lead to failed transmissions and buffer overflows of sensors and, in turn, significant packet losses. In this paper, we optimize jointly the velocity controls and data collection schedules of multiple UAVs to minimize data losses, adapting to the battery levels, queue lengths and channel conditions of the ground sensors, and the trajectories of the UAVs. In the absence of the up-to-date knowledge of the ground sensors' states, a Multi-UAV Deep Reinforcement Learning based Scheduling Algorithm (MADRL-SA) is proposed to allow the UAVs to asymptotically minimize the data loss of the system under the outdated knowledge of the network states at individual UAVs. Numerical results demonstrate that the proposed MADRL-SA reduces the packet loss by up to 54% and 46% in the considered simulation setting, as compared to an existing DRL solution with single-UAV and non-learning greedy heuristic, respectively.",
author = "Yousef Emami and Bo Wei and Kai Li and Wei Ni and Eduardo Tovar",
year = "2021",
month = oct,
day = "31",
doi = "10.1109/TVT.2021.3110801",
language = "English",
volume = "70",
pages = "10986--10998",
journal = "IEEE Transactions on Vehicular Technology",
issn = "0018-9545",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "10",

}

RIS

TY - JOUR

T1 - Joint Communication Scheduling and Velocity Control in Multi-UAV-Assisted Sensor Networks

T2 - A Deep Reinforcement Learning Approach

AU - Emami, Yousef

AU - Wei, Bo

AU - Li, Kai

AU - Ni, Wei

AU - Tovar, Eduardo

PY - 2021/10/31

Y1 - 2021/10/31

N2 - Recently, Unmanned Aerial Vehicle (UAV) swarm has been increasingly studied to collect data from ground sensors in remote and hostile areas. A key challenge is the joint design of the velocities and data collection schedules of the UAVs, as inadequate velocities and schedules would lead to failed transmissions and buffer overflows of sensors and, in turn, significant packet losses. In this paper, we optimize jointly the velocity controls and data collection schedules of multiple UAVs to minimize data losses, adapting to the battery levels, queue lengths and channel conditions of the ground sensors, and the trajectories of the UAVs. In the absence of the up-to-date knowledge of the ground sensors' states, a Multi-UAV Deep Reinforcement Learning based Scheduling Algorithm (MADRL-SA) is proposed to allow the UAVs to asymptotically minimize the data loss of the system under the outdated knowledge of the network states at individual UAVs. Numerical results demonstrate that the proposed MADRL-SA reduces the packet loss by up to 54% and 46% in the considered simulation setting, as compared to an existing DRL solution with single-UAV and non-learning greedy heuristic, respectively.

AB - Recently, Unmanned Aerial Vehicle (UAV) swarm has been increasingly studied to collect data from ground sensors in remote and hostile areas. A key challenge is the joint design of the velocities and data collection schedules of the UAVs, as inadequate velocities and schedules would lead to failed transmissions and buffer overflows of sensors and, in turn, significant packet losses. In this paper, we optimize jointly the velocity controls and data collection schedules of multiple UAVs to minimize data losses, adapting to the battery levels, queue lengths and channel conditions of the ground sensors, and the trajectories of the UAVs. In the absence of the up-to-date knowledge of the ground sensors' states, a Multi-UAV Deep Reinforcement Learning based Scheduling Algorithm (MADRL-SA) is proposed to allow the UAVs to asymptotically minimize the data loss of the system under the outdated knowledge of the network states at individual UAVs. Numerical results demonstrate that the proposed MADRL-SA reduces the packet loss by up to 54% and 46% in the considered simulation setting, as compared to an existing DRL solution with single-UAV and non-learning greedy heuristic, respectively.

U2 - 10.1109/TVT.2021.3110801

DO - 10.1109/TVT.2021.3110801

M3 - Journal article

VL - 70

SP - 10986

EP - 10998

JO - IEEE Transactions on Vehicular Technology

JF - IEEE Transactions on Vehicular Technology

SN - 0018-9545

IS - 10

ER -