Task Offloading with LLM-Enhanced Multi-Agent Reinforcement Learning in UAV-Assisted Edge Computing

Unmanned aerial vehicles (UAVs) furnished with computational servers enable user equipment (UE) to offload complex computational tasks, thereby addressing the limitations of edge computing in remote or resource-constrained environments. The application of value decomposition algorithms for UAV traje...

Full description

Saved in:
Bibliographic Details
Main Authors: Feifan Zhu, Fei Huang, Yantao Yu, Guojin Liu, Tiancong Huang
Format: Article
Language:English
Published: MDPI AG 2024-12-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/1/175
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850113444623155200
author Feifan Zhu
Fei Huang
Yantao Yu
Guojin Liu
Tiancong Huang
author_facet Feifan Zhu
Fei Huang
Yantao Yu
Guojin Liu
Tiancong Huang
author_sort Feifan Zhu
collection DOAJ
description Unmanned aerial vehicles (UAVs) furnished with computational servers enable user equipment (UE) to offload complex computational tasks, thereby addressing the limitations of edge computing in remote or resource-constrained environments. The application of value decomposition algorithms for UAV trajectory planning has drawn considerable research attention. However, existing value decomposition algorithms commonly encounter obstacles in effectively associating local observations with the global state of UAV clusters, which hinders their task-solving capabilities and gives rise to reduced task completion rates and prolonged convergence times. To address these challenges, this paper introduces an innovative multi-agent deep learning framework that conceptualizes multi-UAV trajectory optimization as a decentralized partially observable Markov decision process (Dec-POMDP). This framework integrates the QTRAN algorithm with a large language model (LLM) for efficient region decomposition and employs graph convolutional networks (GCNs) combined with self-attention mechanisms to adeptly manage inter-subregion relationships. The simulation results demonstrate that the proposed method significantly outperforms existing deep reinforcement learning methods, with improvements in convergence speed and task completion rate exceeding 10%. Overall, this framework significantly advances UAV trajectory optimization and enhances the performance of multi-agent systems within UAV-assisted edge computing environments.
format Article
id doaj-art-d66e67243394467aaf661f9e6def5055
institution OA Journals
issn 1424-8220
language English
publishDate 2024-12-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-d66e67243394467aaf661f9e6def50552025-08-20T02:37:09ZengMDPI AGSensors1424-82202024-12-0125117510.3390/s25010175Task Offloading with LLM-Enhanced Multi-Agent Reinforcement Learning in UAV-Assisted Edge ComputingFeifan Zhu0Fei Huang1Yantao Yu2Guojin Liu3Tiancong Huang4School of Microelectronics and Communication Engineering, Chongqing University, Chongqing 400044, ChinaState Grid Chongqing Electric Power Company, Electric Power Research Institute, Chongqing 401123, ChinaSchool of Microelectronics and Communication Engineering, Chongqing University, Chongqing 400044, ChinaSchool of Microelectronics and Communication Engineering, Chongqing University, Chongqing 400044, ChinaSchool of Microelectronics and Communication Engineering, Chongqing University, Chongqing 400044, ChinaUnmanned aerial vehicles (UAVs) furnished with computational servers enable user equipment (UE) to offload complex computational tasks, thereby addressing the limitations of edge computing in remote or resource-constrained environments. The application of value decomposition algorithms for UAV trajectory planning has drawn considerable research attention. However, existing value decomposition algorithms commonly encounter obstacles in effectively associating local observations with the global state of UAV clusters, which hinders their task-solving capabilities and gives rise to reduced task completion rates and prolonged convergence times. To address these challenges, this paper introduces an innovative multi-agent deep learning framework that conceptualizes multi-UAV trajectory optimization as a decentralized partially observable Markov decision process (Dec-POMDP). This framework integrates the QTRAN algorithm with a large language model (LLM) for efficient region decomposition and employs graph convolutional networks (GCNs) combined with self-attention mechanisms to adeptly manage inter-subregion relationships. The simulation results demonstrate that the proposed method significantly outperforms existing deep reinforcement learning methods, with improvements in convergence speed and task completion rate exceeding 10%. Overall, this framework significantly advances UAV trajectory optimization and enhances the performance of multi-agent systems within UAV-assisted edge computing environments.https://www.mdpi.com/1424-8220/25/1/175multi-agent deep learningLLMUAVtrajectory planning
spellingShingle Feifan Zhu
Fei Huang
Yantao Yu
Guojin Liu
Tiancong Huang
Task Offloading with LLM-Enhanced Multi-Agent Reinforcement Learning in UAV-Assisted Edge Computing
Sensors
multi-agent deep learning
LLM
UAV
trajectory planning
title Task Offloading with LLM-Enhanced Multi-Agent Reinforcement Learning in UAV-Assisted Edge Computing
title_full Task Offloading with LLM-Enhanced Multi-Agent Reinforcement Learning in UAV-Assisted Edge Computing
title_fullStr Task Offloading with LLM-Enhanced Multi-Agent Reinforcement Learning in UAV-Assisted Edge Computing
title_full_unstemmed Task Offloading with LLM-Enhanced Multi-Agent Reinforcement Learning in UAV-Assisted Edge Computing
title_short Task Offloading with LLM-Enhanced Multi-Agent Reinforcement Learning in UAV-Assisted Edge Computing
title_sort task offloading with llm enhanced multi agent reinforcement learning in uav assisted edge computing
topic multi-agent deep learning
LLM
UAV
trajectory planning
url https://www.mdpi.com/1424-8220/25/1/175
work_keys_str_mv AT feifanzhu taskoffloadingwithllmenhancedmultiagentreinforcementlearninginuavassistededgecomputing
AT feihuang taskoffloadingwithllmenhancedmultiagentreinforcementlearninginuavassistededgecomputing
AT yantaoyu taskoffloadingwithllmenhancedmultiagentreinforcementlearninginuavassistededgecomputing
AT guojinliu taskoffloadingwithllmenhancedmultiagentreinforcementlearninginuavassistededgecomputing
AT tianconghuang taskoffloadingwithllmenhancedmultiagentreinforcementlearninginuavassistededgecomputing