Intelligent Vehicle Computation Offloading in Vehicular Ad Hoc Networks: A Multi-Agent LSTM Approach with Deep Reinforcement Learning
As distributed computing evolves, edge computing has become increasingly important. It decentralizes resources like computation, storage, and bandwidth, making them more accessible to users, particularly in dynamic Telematics environments. However, these environments are marked by high levels of dyn...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2024-01-01
|
Series: | Mathematics |
Subjects: | |
Online Access: | https://www.mdpi.com/2227-7390/12/3/424 |
_version_ | 1797318438359662592 |
---|---|
author | Dingmi Sun Yimin Chen Hao Li |
author_facet | Dingmi Sun Yimin Chen Hao Li |
author_sort | Dingmi Sun |
collection | DOAJ |
description | As distributed computing evolves, edge computing has become increasingly important. It decentralizes resources like computation, storage, and bandwidth, making them more accessible to users, particularly in dynamic Telematics environments. However, these environments are marked by high levels of dynamic uncertainty due to frequent changes in vehicle location, network status, and edge server workload. This complexity poses substantial challenges in rapidly and accurately handling computation offloading, resource allocation, and delivering low-latency services in such a variable environment. To address these challenges, this paper introduces a “Cloud–Edge–End” collaborative model for Telematics edge computing. Building upon this model, we develop a novel distributed service offloading method, LSTM Muti-Agent Deep Reinforcement Learning (L-MADRL), which integrates deep learning with deep reinforcement learning. This method includes a predictive model capable of forecasting the future demands on intelligent vehicles and edge servers. Furthermore, we conceptualize the computational offloading problem as a Markov decision process and employ the Multi-Agent Deep Deterministic Policy Gradient (MADDPG) approach for autonomous, distributed offloading decision-making. Our empirical results demonstrate that the L-MADRL algorithm substantially reduces service latency and energy consumption by 5–20%, compared to existing algorithms, while also maintaining a balanced load across edge servers in diverse Telematics edge computing scenarios. |
first_indexed | 2024-03-08T03:52:25Z |
format | Article |
id | doaj.art-1b25220f3a134ec8858c6246cbdd38b7 |
institution | Directory Open Access Journal |
issn | 2227-7390 |
language | English |
last_indexed | 2024-03-08T03:52:25Z |
publishDate | 2024-01-01 |
publisher | MDPI AG |
record_format | Article |
series | Mathematics |
spelling | doaj.art-1b25220f3a134ec8858c6246cbdd38b72024-02-09T15:18:17ZengMDPI AGMathematics2227-73902024-01-0112342410.3390/math12030424Intelligent Vehicle Computation Offloading in Vehicular Ad Hoc Networks: A Multi-Agent LSTM Approach with Deep Reinforcement LearningDingmi Sun0Yimin Chen1Hao Li2The School of Information Science and Engineering, Yunnan University, Kunming 650504, ChinaThe School of Information Science and Engineering, Yunnan University, Kunming 650504, ChinaThe School of Information Science and Engineering, Yunnan University, Kunming 650504, ChinaAs distributed computing evolves, edge computing has become increasingly important. It decentralizes resources like computation, storage, and bandwidth, making them more accessible to users, particularly in dynamic Telematics environments. However, these environments are marked by high levels of dynamic uncertainty due to frequent changes in vehicle location, network status, and edge server workload. This complexity poses substantial challenges in rapidly and accurately handling computation offloading, resource allocation, and delivering low-latency services in such a variable environment. To address these challenges, this paper introduces a “Cloud–Edge–End” collaborative model for Telematics edge computing. Building upon this model, we develop a novel distributed service offloading method, LSTM Muti-Agent Deep Reinforcement Learning (L-MADRL), which integrates deep learning with deep reinforcement learning. This method includes a predictive model capable of forecasting the future demands on intelligent vehicles and edge servers. Furthermore, we conceptualize the computational offloading problem as a Markov decision process and employ the Multi-Agent Deep Deterministic Policy Gradient (MADDPG) approach for autonomous, distributed offloading decision-making. Our empirical results demonstrate that the L-MADRL algorithm substantially reduces service latency and energy consumption by 5–20%, compared to existing algorithms, while also maintaining a balanced load across edge servers in diverse Telematics edge computing scenarios.https://www.mdpi.com/2227-7390/12/3/424edge computingcomputation offloadingresource allocationdeep reinforcement learning |
spellingShingle | Dingmi Sun Yimin Chen Hao Li Intelligent Vehicle Computation Offloading in Vehicular Ad Hoc Networks: A Multi-Agent LSTM Approach with Deep Reinforcement Learning Mathematics edge computing computation offloading resource allocation deep reinforcement learning |
title | Intelligent Vehicle Computation Offloading in Vehicular Ad Hoc Networks: A Multi-Agent LSTM Approach with Deep Reinforcement Learning |
title_full | Intelligent Vehicle Computation Offloading in Vehicular Ad Hoc Networks: A Multi-Agent LSTM Approach with Deep Reinforcement Learning |
title_fullStr | Intelligent Vehicle Computation Offloading in Vehicular Ad Hoc Networks: A Multi-Agent LSTM Approach with Deep Reinforcement Learning |
title_full_unstemmed | Intelligent Vehicle Computation Offloading in Vehicular Ad Hoc Networks: A Multi-Agent LSTM Approach with Deep Reinforcement Learning |
title_short | Intelligent Vehicle Computation Offloading in Vehicular Ad Hoc Networks: A Multi-Agent LSTM Approach with Deep Reinforcement Learning |
title_sort | intelligent vehicle computation offloading in vehicular ad hoc networks a multi agent lstm approach with deep reinforcement learning |
topic | edge computing computation offloading resource allocation deep reinforcement learning |
url | https://www.mdpi.com/2227-7390/12/3/424 |
work_keys_str_mv | AT dingmisun intelligentvehiclecomputationoffloadinginvehicularadhocnetworksamultiagentlstmapproachwithdeepreinforcementlearning AT yiminchen intelligentvehiclecomputationoffloadinginvehicularadhocnetworksamultiagentlstmapproachwithdeepreinforcementlearning AT haoli intelligentvehiclecomputationoffloadinginvehicularadhocnetworksamultiagentlstmapproachwithdeepreinforcementlearning |