End-to-End Latency Optimization for Resilient Distributed Convolutional Neural Network Inference in Resource-Constrained Unmanned Aerial Vehicle Swarms

An unmanned aerial vehicle (UAV) swarm has emerged as a powerful tool for mission execution in a variety of applications supported by deep neural networks (DNNs). In the context of UAV swarms, conventional methods for efficient data processing involve transmitting data to cloud and edge servers. How...

Full description

Saved in:
Bibliographic Details
Main Authors: Jeongho Kim, Joonho Seon, Soohyun Kim, Seongwoo Lee, Jinwook Kim, Byungsun Hwang, Youngghyu Sun, Jinyoung Kim
Format: Article
Language:English
Published: MDPI AG 2024-11-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/14/23/10832
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846124520471003136
author Jeongho Kim
Joonho Seon
Soohyun Kim
Seongwoo Lee
Jinwook Kim
Byungsun Hwang
Youngghyu Sun
Jinyoung Kim
author_facet Jeongho Kim
Joonho Seon
Soohyun Kim
Seongwoo Lee
Jinwook Kim
Byungsun Hwang
Youngghyu Sun
Jinyoung Kim
author_sort Jeongho Kim
collection DOAJ
description An unmanned aerial vehicle (UAV) swarm has emerged as a powerful tool for mission execution in a variety of applications supported by deep neural networks (DNNs). In the context of UAV swarms, conventional methods for efficient data processing involve transmitting data to cloud and edge servers. However, these methods often face limitations in adapting to real-time applications due to the low latency of cloud-based approaches and weak mobility of edge-based approaches. In this paper, a new system called deep reinforcement learning-based resilient layer distribution (DRL-RLD) for distributed inference is designed to minimize end-to-end latency in UAV swarm, considering the resource constraints of UAVs. The proposed system dynamically allocates CNN layers based on UAV-to-UAV and UAV-to-ground communication links to minimize end-to-end latency. It can also enhance resilience to maintain mission continuity by reallocating layers when inoperable UAVs occur. The performance of the proposed system was verified through simulations in terms of latency compared to the comparison baselines, and its robustness was demonstrated in the presence of inoperable UAVs.
format Article
id doaj-art-8c88c6f2932a4be29a47064d252e8c72
institution Kabale University
issn 2076-3417
language English
publishDate 2024-11-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-8c88c6f2932a4be29a47064d252e8c722024-12-13T16:21:53ZengMDPI AGApplied Sciences2076-34172024-11-0114231083210.3390/app142310832End-to-End Latency Optimization for Resilient Distributed Convolutional Neural Network Inference in Resource-Constrained Unmanned Aerial Vehicle SwarmsJeongho Kim0Joonho Seon1Soohyun Kim2Seongwoo Lee3Jinwook Kim4Byungsun Hwang5Youngghyu Sun6Jinyoung Kim7Department of Electronic Convergence Engineering, Kwangwoon University, Seoul 01897, Republic of KoreaDepartment of Electronic Convergence Engineering, Kwangwoon University, Seoul 01897, Republic of KoreaDepartment of Electronic Convergence Engineering, Kwangwoon University, Seoul 01897, Republic of KoreaDepartment of Electronic Convergence Engineering, Kwangwoon University, Seoul 01897, Republic of KoreaDepartment of Electronic Convergence Engineering, Kwangwoon University, Seoul 01897, Republic of KoreaDepartment of Electronic Convergence Engineering, Kwangwoon University, Seoul 01897, Republic of KoreaResearch and Development Department, SMART EVER, Co., Ltd., Seoul 01886, Republic of KoreaDepartment of Electronic Convergence Engineering, Kwangwoon University, Seoul 01897, Republic of KoreaAn unmanned aerial vehicle (UAV) swarm has emerged as a powerful tool for mission execution in a variety of applications supported by deep neural networks (DNNs). In the context of UAV swarms, conventional methods for efficient data processing involve transmitting data to cloud and edge servers. However, these methods often face limitations in adapting to real-time applications due to the low latency of cloud-based approaches and weak mobility of edge-based approaches. In this paper, a new system called deep reinforcement learning-based resilient layer distribution (DRL-RLD) for distributed inference is designed to minimize end-to-end latency in UAV swarm, considering the resource constraints of UAVs. The proposed system dynamically allocates CNN layers based on UAV-to-UAV and UAV-to-ground communication links to minimize end-to-end latency. It can also enhance resilience to maintain mission continuity by reallocating layers when inoperable UAVs occur. The performance of the proposed system was verified through simulations in terms of latency compared to the comparison baselines, and its robustness was demonstrated in the presence of inoperable UAVs.https://www.mdpi.com/2076-3417/14/23/10832resource-constrained UAV swarmdistributed inferenceresilient UAV systemdeep reinforcement learningend-to-end latency optimization
spellingShingle Jeongho Kim
Joonho Seon
Soohyun Kim
Seongwoo Lee
Jinwook Kim
Byungsun Hwang
Youngghyu Sun
Jinyoung Kim
End-to-End Latency Optimization for Resilient Distributed Convolutional Neural Network Inference in Resource-Constrained Unmanned Aerial Vehicle Swarms
Applied Sciences
resource-constrained UAV swarm
distributed inference
resilient UAV system
deep reinforcement learning
end-to-end latency optimization
title End-to-End Latency Optimization for Resilient Distributed Convolutional Neural Network Inference in Resource-Constrained Unmanned Aerial Vehicle Swarms
title_full End-to-End Latency Optimization for Resilient Distributed Convolutional Neural Network Inference in Resource-Constrained Unmanned Aerial Vehicle Swarms
title_fullStr End-to-End Latency Optimization for Resilient Distributed Convolutional Neural Network Inference in Resource-Constrained Unmanned Aerial Vehicle Swarms
title_full_unstemmed End-to-End Latency Optimization for Resilient Distributed Convolutional Neural Network Inference in Resource-Constrained Unmanned Aerial Vehicle Swarms
title_short End-to-End Latency Optimization for Resilient Distributed Convolutional Neural Network Inference in Resource-Constrained Unmanned Aerial Vehicle Swarms
title_sort end to end latency optimization for resilient distributed convolutional neural network inference in resource constrained unmanned aerial vehicle swarms
topic resource-constrained UAV swarm
distributed inference
resilient UAV system
deep reinforcement learning
end-to-end latency optimization
url https://www.mdpi.com/2076-3417/14/23/10832
work_keys_str_mv AT jeonghokim endtoendlatencyoptimizationforresilientdistributedconvolutionalneuralnetworkinferenceinresourceconstrainedunmannedaerialvehicleswarms
AT joonhoseon endtoendlatencyoptimizationforresilientdistributedconvolutionalneuralnetworkinferenceinresourceconstrainedunmannedaerialvehicleswarms
AT soohyunkim endtoendlatencyoptimizationforresilientdistributedconvolutionalneuralnetworkinferenceinresourceconstrainedunmannedaerialvehicleswarms
AT seongwoolee endtoendlatencyoptimizationforresilientdistributedconvolutionalneuralnetworkinferenceinresourceconstrainedunmannedaerialvehicleswarms
AT jinwookkim endtoendlatencyoptimizationforresilientdistributedconvolutionalneuralnetworkinferenceinresourceconstrainedunmannedaerialvehicleswarms
AT byungsunhwang endtoendlatencyoptimizationforresilientdistributedconvolutionalneuralnetworkinferenceinresourceconstrainedunmannedaerialvehicleswarms
AT youngghyusun endtoendlatencyoptimizationforresilientdistributedconvolutionalneuralnetworkinferenceinresourceconstrainedunmannedaerialvehicleswarms
AT jinyoungkim endtoendlatencyoptimizationforresilientdistributedconvolutionalneuralnetworkinferenceinresourceconstrainedunmannedaerialvehicleswarms