基于空地协同的动态车载边缘任务卸载方法

doi:10.19678/j.issn.1000-3428.0069836

摘要/Abstract

摘要：

为了进一步优化车载服务的服务质量(QoS)，移动边缘计算(MEC)被深度整合于车联网(IoV)中，旨在为车辆提供地理位置较近的计算资源，降低任务处理延迟和能耗。然而，传统的MEC服务器部署主要依赖于地面基站(BS)，这不仅导致高昂的部署成本，而且限制其覆盖范围，难以确保为所有车辆提供无间断服务。为了应对上述挑战，空地协同IoV作为一种新兴的技术方案应运而生。无人机(UAV)能够借助其视距链路的灵活性动态地协助路边单元(RSU)，为车辆用户提供更为灵活的计算资源，进而保障车载服务的连续性和高效性。提出一种基于空地协同的动态车载边缘任务卸载方法(DVETOM)。该方法采用车-路-空架构，构建了车辆到RSU(V2R)链路和车辆到UAV(V2U)链路。针对车辆任务的本地执行、卸载至RSU执行和卸载至UAV执行3种模式分别构建传输模型和计算模型，并以最小化系统时延和能耗作为联合优化目标构建目标函数。DVETOM将任务卸载问题转化为马尔可夫决策过程(MDP)，基于深度强化学习(DRL)的分布式深度确定性策略梯度(D4PG)算法优化任务卸载策略。与5种基准方法进行对比，实验结果表明，DVETOM在提升车辆用户QoS的同时，在降低系统时延方面优于现有方法3.45%~23.7%，在降低系统能耗方面优于现有方法5.8%~23.47%。综上所述，DVETOM有效地优化了IoV中的车载边缘任务卸载，为IoV用户提供了更高效、更节能的服务解决方案，展现了其在智能交通系统领域的广泛应用潜力。

关键词: 车联网, 边缘计算, 空地协同, 任务卸载, 深度强化学习

Abstract:

To optimize Quality of Service (QoS), Mobile Edge Computing (MEC) has been deeply integrated into the Internet of Vehicle (IoV) to provide geographically proximal computing resources for vehicles, thereby reducing task processing latency and energy consumption. However, traditional MEC server deployment relies primarily on terrestrial Base Stations (BSs), resulting in high deployment costs and limited coverage, making it difficult to ensure uninterrupted services for all vehicles. Air-ground collaborative IoV technology has emerged as a solution to these challenges. Unmanned Aerial Vehicles (UAVs) can dynamically assist Road-Side Units (RSUs) using their flexibility in line-of-sight links, providing more flexible computing resources for vehicular users, thereby ensuring the continuity and efficiency of in-vehicle services. Therefore, this study proposes a Dynamic Vehicular Edge Task Offloading Method (DVETOM) based on air-ground collaboration. This method adopts a vehicle-road-air architecture, establishing Vehicle-to-RSU (V2R) and Vehicle-to-UAV (V2U) links. Transmission and computation models are constructed for three modes: local execution of vehicular tasks, offloading tasks to the RSU, and offloading tasks to the UAV. An objective function is established with the joint optimization goal of minimizing system latency and energy consumption. DVETOM transforms the task offloading problem into a Markov Decision Process (MDP) and optimizes the task offloading strategy by using the Distributed Deep Deterministic Policy Gradient (D4PG) algorithm based on Deep Reinforcement Learning (DRL). Compared with 5 benchmark methods, experimental results show that DVETOM outperforms existing methods by 3.45%—23.7% in terms of reducing system latency and 5.8%—23.47% in terms of reducing system energy consumption while improving QoS for vehicular users. In conclusion, DVETOM enhances the offloading of vehicular edge computing tasks within the IoV effectively. It offers IoV users a more efficient and energy-conserving solution, showcasing its extensive potential for application in intelligent transportation systems.

Key words: Internet of Vehicle (IoV), edge computing, air-ground collaboration, task offloading, Deep Reinforcement Learning (DRL)

崔萌萌, 施静燕, 项昊龙. 基于空地协同的动态车载边缘任务卸载方法[J]. 计算机工程, 2025, 51(9): 25-37.

CUI Mengmeng, SHI Jingyan, XIANG Haolong. Dynamic Vehicle Edge Task Offloading Method Based on Air-Ground Collaboration[J]. Computer Engineering, 2025, 51(9): 25-37.

https://www.ecice06.com/CN/Y2025/V51/I9/25

图/表 16

图1 UAV辅助IoV边缘卸载系统框架

Fig.1 UAV-assisted IoV edge offloading system framework

图2 车辆将计算任务卸载到RSU示意图

Fig.2 Schematic diagram of offloading computing tasks from vehicles to RSU

图3 车辆将计算任务卸载到UAV示意图

Fig.3 Schematic diagram of offloading computing tasks from vehicles to UAV

图4 车辆边缘计算任务卸载的MDP决策流程图

Fig.4 MDP decision flowchart for vehicle edge computing task offloading

图5 基于D4PG的DVETOM实现

Fig.5 Implementation of DVETOM based on D4PG

图6 DQN、DDPG和DVETOM实行每个步骤的奖励

Fig.6 Reward per learning step achieved by DQN, DDPG and DVETOM

图7 不同车辆数量下的系统时延

Fig.7 System latency under different numbers of vehicles

图8 不同UAV数量下的系统时延

Fig.8 System latency under different numbers of UAV

图9 不同任务数据大小下的系统时延

Fig.9 System latency under different task data sizes

图10 不同车辆数量下的系统能耗

Fig.10 System energy consumption under different numbers of vehicles

图11 不同UAV数量下的系统能耗

Fig.11 System energy consumption under different numbers of UAVs

图12 不同任务数据大小下的系统能耗

Fig.12 System energy consumption under different task data sizes

图13 不同时延权重因子z_t下的系统成本

Fig.13 System cost under different latency weighting factor z_t

图14 不同车辆数量下的系统成本

Fig.14 System cost under different numbers of vehicles

图15 不同任务数据大小下的系统成本

Fig.15 System cost under different task data sizes

参考文献 26

1	PAUL A , CHILAMKURTI N , DANIEL A , et al. Intelligent vehicular networks and communications: fundamentals, architectures and solutions. [S. l.]: Elsevier, 2016.
2	XU X L , ZHOU X H , ZHOU X K , et al. Distributed edge caching for zero trust-enabled connected and automated vehicles: a multi-agent reinforcement learning approach. IEEE Wireless Communications, 2024, 31 (2): 36- 41. doi: 10.1109/MWC.001.2300414
3	ZHANG X C , ZHANG J , XIONG J , et al. Energy-efficient multi-UAV-enabled multiaccess edge computing incorporating NOMA. IEEE Internet of Things Journal, 2020, 7 (6): 5613- 5627. doi: 10.1109/JIOT.2020.2980035
4	SARWAR M M G , MURPHY C , HOU D Q , et al. Machine learning at the network edge: a survey. ACM Computing Surveys, 2022, 54 (8): 1- 37.
5	CHANG Z Q , LIU S B , XIONG X X , et al. A survey of recent advances in edge-computing-powered artificial intelligence of things. IEEE Internet of Things Journal, 2021, 8 (18): 13849- 13875. doi: 10.1109/JIOT.2021.3088875
6	XIONG K , LIU Y , ZHANG L T , et al. Joint optimization of trajectory, task offloading, and CPU control in UAV-assisted wireless powered fog computing networks. IEEE Transactions on Green Communications and Networking, 2022, 6 (3): 1833- 1845. doi: 10.1109/TGCN.2022.3157735
7	QU Y B , DAI H P , WANG H C , et al. Service provisioning for UAV-enabled mobile edge computing. IEEE Journal on Selected Areas in Communications, 2021, 39 (11): 3287- 3305. doi: 10.1109/JSAC.2021.3088660
8	SUN L , WAN L T , WANG X P . Learning-based resource allocation strategy for industrial IoT in UAV-enabled MEC systems. IEEE Transactions on Industrial Informatics, 2021, 17 (7): 5031- 5040. doi: 10.1109/TII.2020.3024170
9	WANG Y G, WANG H, WEI X L. Energy-efficient UAV deployment and task scheduling in multi-UAV edge computing[C]//Proceedings of the International Conference on Wireless Communications and Signal Processing (WCSP). Nanjing, China: IEEE Press, 2020: 1147-1152.
10	HUANG N , DOU C L , WU Y , et al. Unmanned-aerial-vehicle-aided integrated sensing and computation with mobile-edge computing. IEEE Internet of Things Journal, 2023, 10 (19): 16830- 16844. doi: 10.1109/JIOT.2023.3270332
11	DENG X H , LIU Y J , ZHU C X , et al. Air-ground surveillance sensor network based on edge computing for target tracking. Computer Communications, 2021, 166, 254- 261. doi: 10.1016/j.comcom.2020.10.012
12	ZHENG Q H , TIAN X Y , YU Z G , et al. MobileRaT: a lightweight radio transformer method for automatic modulation classification in drone communication systems. Drones, 2023, 7 (10): 596. doi: 10.3390/drones7100596
13	何杰, 马强. 基于深度强化学习的C-V2X任务卸载研究. 计算机工程, 2024, 50 (12): 200- 212. doi: 10.19678/j.issn.1000-3428.0068425
	HE J , MA Q . Research on C-V2X task unloading based on deep reinforcement learning. Computer Engineering, 2024, 50 (12): 200- 212. doi: 10.19678/j.issn.1000-3428.0068425
14	MAO K , ZHU Q M , SONG M Z , et al. Machine-learning-based 3-D channel modeling for U2V mmWave communications. IEEE Internet of Things Journal, 2022, 9 (18): 17592- 17607. doi: 10.1109/JIOT.2022.3155773
15	SAMIR M , ASSI C , SHARAFEDDINE S , et al. Age of information aware trajectory planning of UAVs in intelligent transportation systems: a deep learning approach. IEEE Transactions on Vehicular Technology, 2020, 69 (11): 12382- 12395. doi: 10.1109/TVT.2020.3023861
16	PENG H X , SHEN X M . Multi-agent reinforcement learning based resource management in MEC- and UAV-assisted vehicular networks. IEEE Journal on Selected Areas in Communications, 2021, 39 (1): 131- 141. doi: 10.1109/JSAC.2020.3036962
17	ZHU S C , GUI L , ZHAO D M , et al. Learning-based computation offloading approaches in UAVs-assisted edge computing. IEEE Transactions on Vehicular Technology, 2021, 70 (1): 928- 944. doi: 10.1109/TVT.2020.3048938
18	WANG W , XU X L , BILAL M , et al. UAV-assisted content caching for human-centric consumer applications in IoV. IEEE Transactions on Consumer Electronics, 2024, 70 (1): 927- 938. doi: 10.1109/TCE.2023.3349079
19	JIANG H B , DAI X X , XIAO Z , et al. Joint task offloading and resource allocation for energy-constrained mobile edge computing. IEEE Transactions on Mobile Computing, 2023, 22 (7): 4000- 4015. doi: 10.1109/TMC.2022.3150432
20	MATOLAK D W , SUN R Y . Air-ground channel characterization for unmanned aircraft systems: part I: methods, measurements, and models for over-water settings. IEEE Transactions on Vehicular Technology, 2017, 66 (1): 26- 44. doi: 10.1109/TVT.2016.2530306
21	CHIU C C, TSAI A H, LIN H P, et al. Channel modeling of air-to-ground signal measurement with two-ray ground-reflection model for UAV communication systems[C]//Proceedings of the 30th Wireless and Optical Communications Conference (WOCC). Taipei, China: IEEE Press, 2021: 251-256.
22	FEUERSTEIN M J , BLACKARD K L , RAPPAPORT T S , et al. Path loss, delay spread, and outage models as functions of antenna height for microcellular system design. IEEE Transactions on Vehicular Technology, 1994, 43 (3): 487- 498. doi: 10.1109/25.312809
23	ZHENG Q H , SAPONARA S , TIAN X Y , et al. A real-time constellation image classification method of wireless communication signals based on the lightweight network MobileViT. Cognitive Neurodynamics, 2024, 18 (2): 659- 671. doi: 10.1007/s11571-023-10015-7
24	SON D B , BINH T H , VO H K , et al. Value-based reinforcement learning approaches for task offloading in delay constrained vehicular edge computing. Engineering Applications of Artificial Intelligence, 2022, 113, 104898. doi: 10.1016/j.engappai.2022.104898
25	WEI Z, HE R X, LI Y N. Deep reinforcement learning based task offloading and resource allocation for MEC-enabled IoT networks[C]//Proceedings of the IEEE/CIC International Conference on Communications in China (ICCC Workshops). Dalian, China: IEEE Press, 2023: 1-6.
26	吴学文, 廖婧贤. 云边协同系统中基于博弈论的资源分配与任务卸载方案. 系统仿真学报, 2022, 34 (7): 1468- 1481.
	WU X W , LIAO J X . Game-based resource allocation and task offloading scheme in collaborative cloud-edge computing system. Journal of System Simulation, 2022, 34 (7): 1468- 1481.

[1]	毕昌兵, 田有亮. 车联网中基于身份签名的匿名可追溯消息认证方案[J]. 计算机工程, 2025, 51(9): 158-165.
[2]	陈彦如, 刘珂良, 冉茂亮. 基于深度强化学习的外卖即时配送实时优化[J]. 计算机工程, 2025, 51(9): 328-339.
[3]	赵季红, 臧若雨, 刘振. 面向卫星车载MEC网络的协同计算卸载方法[J]. 计算机工程, 2025, 51(9): 49-58.
[4]	黄业恒, 覃团发, 苏振朗, 王素红. 6G网络下超密集无线体域网高效卸载策略[J]. 计算机工程, 2025, 51(9): 201-212.
[5]	秦敏浩, 孙未未. 基于隐状态预测的失真交通信号灯路口控制策略[J]. 计算机工程, 2025, 51(9): 1-13.
[6]	张伟航, 钟永彦, 向元柱, 丁士旵. 边云辅助下的可撤销属性加密方案[J]. 计算机工程, 2025, 51(7): 244-253.
[7]	杨莉斌, 詹成, 李婷婷, 廖婧睿. 多天线无人机视频通信中的节能资源分配策略[J]. 计算机工程, 2025, 51(6): 266-274.
[8]	亓明凯, 王迪, 张立晔. 基于分层强化学习的在线三维装箱模型[J]. 计算机工程, 2025, 51(6): 136-145.
[9]	武小丰, 袁培燕. 一种改进蛇优化算法的边缘服务器动态放置策略[J]. 计算机工程, 2025, 51(6): 255-265.
[10]	吕超峰, 徐鹏飞, 罗迪, 刘金平. 基于多智能体深度强化学习的SD-IoT控制器部署[J]. 计算机工程, 2025, 51(5): 83-92.
[11]	杜剑波, 董伟哲, 金蓉, 王军选, 康嘉文, 刘雷, 策力木格. 基于MEC的空天地一体化网络任务分割与资源分配[J]. 计算机工程, 2025, 51(5): 43-51.
[12]	吴凯峰, 刘磊, 刘晨, 梁成庆. 基于融合课程思想MADDPG的无人机编队控制[J]. 计算机工程, 2025, 51(5): 73-82.
[13]	李俊吉, 张佳琦, 高改梅, 杨莉. 基于信誉机制的车联网共识算法[J]. 计算机工程, 2025, 51(4): 217-226.
[14]	林绍福, 陈盈盈, 李硕朋. 基于深度强化学习的多无人机能量传输与边缘计算联合优化方法[J]. 计算机工程, 2025, 51(3): 144-154.
[15]	李思源, 钟兴宇, 李凯茵, 徐清振. 基于多层图关系和强化学习的策略教学研究[J]. 计算机工程, 2025, 51(3): 122-130.

选择文件类型/文献管理软件名称

选择包含的内容