AoI驱动的空-地协同移动群智感知与计算

doi:10.19678/j.issn.1000-3428.0070553

计算机工程 ›› 2026, Vol. 52 ›› Issue (8): 260-271. doi: 10.19678/j.issn.1000-3428.0070553

• 新一代网络与边缘计算 • 上一篇下一篇

AoI驱动的空-地协同移动群智感知与计算

汤晓畅¹^,², 徐海星¹^,², 沈炳楠¹^,², 王亮¹^,²^,*(), 於志文¹^,²

1. 西北工业大学计算机学院, 陕西西安 710129
2. 人机物融合群智计算教育部重点实验室, 陕西西安 710129

收稿日期:2024-10-29 修回日期:2025-01-06 出版日期:2026-08-15 发布日期:2025-03-06
通讯作者: 王亮
作者简介:
汤晓畅(CCF学生会员), 男, 硕士研究生, 主研方向为移动群智感知
徐海星, 硕士研究生
沈炳楠, 硕士研究生
王亮(通信作者), 教授
於志文, 教授
基金资助:
国家自然科学基金(62332014); 陕西省自然科学基础研究计划(2023-JC-JQ-54); 陕西省重点研发计划(2024GX-YBXM-006)

AoI-Driven Air-Ground Collaborative Mobile Crowd Sensing and Computing

TANG Xiaochang¹^,², XU Haixing¹^,², SHEN Bingnan¹^,², WANG Liang¹^,²^,*(), YU Zhiwen¹^,²

1. School of Computer Science, Northwestern Polytechnical University, Xi'an 710129, Shannxi, China
2. Key Laboratory of Human-Cyber-Physical Ternary Crowd Computing, Ministry of Education, Xi'an 710129, Shannxi, China

Received:2024-10-29 Revised:2025-01-06 Online:2026-08-15 Published:2025-03-06
Contact: WANG Liang

摘要/Abstract

摘要：

为了克服传统的"以人为中心"移动群智感知(MCS)在感知时效性、全局性等方面的局限与不足, 提出一种人机混合下的空-地协同移动群智感知与计算框架, 低空无人机与地面参与者充分利用各自的感知、计算资源, 通过协同交互的方式实现数据感知与计算的本地化。考虑到延迟敏感型应用场景对感知服务时效性的强需求, 将信息年龄(AoI)指标引入上述框架中, 通过联合优化无人机移动轨迹, 将移动端设备计算卸载至无人机的选择变量以及数据卸载比例, 最终实现所有感知数据的AoI最小化。为了实现上述目标, 设计一种多智能体深度强化学习(DRL)算法CAMATD3, 结合多智能体双延迟深度确定性策略梯度MATD3的双Q学习机制和深度Q网络(DQN)的训练机制; 同时针对性地提出了分阶段决策机制和完整经验存储策略, 以解决优化过程中混合动作空间所带来的问题。仿真实验结果表明, 所提的CAMATD3算法在智能体训练效果和AoI最小化方面显著优于现有方法, 有效提升了数据感知的时效性。

关键词: 无人机, 移动群智感知, 信息年龄, 数据卸载, 深度强化学习

Abstract:

To address the limitations and deficiencies of traditional ″human-centered″ Mobile Crowd Sensing(MCS) in terms of timeliness and global sensing coverage, this study proposes a human-Unmanned Aerial Vehicle (UAV) hybrid air—ground collaborative MCS and computing framework. In this framework, low-altitude UAVs and ground participants fully leverage their respective sensing and computing resources through collaborative interactions to achieve localized data sensing and computation. Considering the stringent requirements for data timeliness in delay-sensitive application scenarios, the Age of Information (AoI) metric is introduced into this framework. By jointly optimizing UAV flight trajectories, selection variables for offloading computations from mobile devices to UAVs, and data offloading ratios, the framework aims to minimize the AoI of all sensed data. To achieve this objective, a multi-agent Deep Reinforcement Learning (DRL) algorithm CAMATD3 is designed, which combines the dual Q-learning mechanism of multi-agent twin delayed deep deterministic policy gradient MATD3 with the training mechanism of Deep Q-Networks (DQN). Additionally, a phased decision-making mechanism and comprehensive experience replay strategy are proposed to address the challenges posed by the hybrid action space during the optimization process. The simulation results demonstrate that the proposed CAMATD3 algorithm significantly outperforms existing methods in terms of agent training effectiveness and AoI minimization, thereby substantially improving the timeliness of data sensing.

Key words: Unmanned Aerial Vehicles (UAV), Mobile Crowd Sensing(MCS), Age of Information (AoI), data offloading, Deep Reinforcement Learning (DRL)

汤晓畅, 徐海星, 沈炳楠, 王亮, 於志文. AoI驱动的空-地协同移动群智感知与计算[J]. 计算机工程, 2026, 52(8): 260-271.

TANG Xiaochang, XU Haixing, SHEN Bingnan, WANG Liang, YU Zhiwen. AoI-Driven Air-Ground Collaborative Mobile Crowd Sensing and Computing[J]. Computer Engineering, 2026, 52(8): 260-271.

https://www.ecice06.com/CN/Y2026/V52/I8/260

图/表 13

图1 系统模型

Fig.1 System model

图2 AoI与时间的关系

Fig.2 The relationship between AoI and time

图3 MATD3算法

Fig.3 MATD3 algorithm

图4 CAMATD3动作选择过程

Fig.4 CAMATD3 action selection process

图5 CAMATD3在不同学习率下的收敛性能

Fig.5 Convergence performance of CAMATD3 under different learning rates

图6 CAMATD3在不同奖励折扣因子下的收敛性能

Fig.6 Convergence performance of CAMATD3 under different reward discounts factors

图7 不同算法的收敛性能对比

Fig.7 Comparison of convergence performance among different algorithms

图8 不同算法下的轨迹优化结果

Fig.8 Trajectory optimization results under different algorithms

图9 不同任务量对系统性能的影响

Fig.9 The impact of different task volumes on system performance

图10 不同地面参与者数量对系统性能的影响

Fig.10 Impact of different numbers of ground participants on system performance

图11 不同无人机数量对系统性能的影响

Fig.11 Impact of different numbers of UAVs on system performance

参考文献 28

1	郭斌, 翟书颖, 於志文, 等. 群智大数据: 感知, 优选与理解. 大数据, 2017, 3 (5): 57- 69.
	GUO B , ZHAI S Y , YU Z W , et al. Crowdsensing big data: sensing, optimization, and understanding. Big Data Research, 2017, 3 (5): 57- 69.
2	郭斌, 刘思聪, 刘琰, 等. 智能物联网: 概念、体系架构与关键技术. 计算机学报, 2023, 46 (11): 2259- 2278.
	GUO B , LIU S C , LIU Y , et al. AIoT: the concept, architecture and key techniques. Chinese Journal of Computers, 2023, 46 (11): 2259- 2278.
3	GUO B , WANG Z , YU Z W , et al. Mobile crowd sensing and computing: the review of an emerging human-powered sensing paradigm. ACM computing surveys (CSUR), 2016, 48 (1): 1- 31.
4	SAMIR M , SHARAFEDDINE S , ASSI C M , et al. UAV trajectory planning for data collection from time-constrained IoT devices. IEEE Transactions on Wireless Communications, 2019, 19 (1): 34- 46.
5	LI J X , ZHAO H T , WANG H J , et al. Joint optimization on trajectory, altitude, velocity, and link scheduling for minimum mission time in UAV-aided data collection. IEEE Internet of Things Journal, 2019, 7 (2): 1464- 1475.
6	LIU C H , MA X X , GAO X D , et al. Distributed energy-efficient multi-UAV navigation for long-term communication coverage by deep reinforcement learning. IEEE Transactions on Mobile Computing, 2019, 19 (6): 1274- 1285.
7	WANG Y , RU Z Y , WANG K Z , et al. Joint deployment and task scheduling optimization for large-scale mobile users in multi-UAV-enabled mobile edge computing. IEEE Transactions on Cybernetics, 2019, 50 (9): 3984- 3997.
8	KAUL S, GRUTESER M, RAI V, et al. Minimizing age of information in vehicular networks[C]//Proceedings of the 8th Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks. Washington D.C., USA: IEEE Press, 2011: 350-358.
9	FUJIMOTO S, HOOF H V, MEGER D. Addressing function approximation error in actor-critic methods[EB/OL]. [2024-09-20]. https://arxiv.org/pdf/1802.09477.
10	MNIH V , KAVUKCUOGLU K , SILVER D , et al. Human-level control through deep reinforcement learning. Nature, 2015, 518 (7540): 529- 533. doi: 10.1038/nature14236
11	杨佳珠, 余芳, 杨勇生. 考虑能耗的海铁联运集装箱码头多设备协同调度. 计算机工程, 2024, 50 (10): 393- 404. doi: 10.19678/j.issn.1000-3428.0068369
	YANG J Z , YU F , YANG Y S . Multi-equipment coordinated scheduling considering energy consumption in sea-rail intermodal container terminal. Computer Engineering, 2024, 50 (10): 393- 404. doi: 10.19678/j.issn.1000-3428.0068369
12	YANG S, HAN K Y, ZHENG Z Z, et al. Towards personalized task matching in mobile crowdsensing via fine-grained user profiling[C]//Proceedings of IEEE Conference on Computer Communications. Washington D.C., USA: IEEE Press, 2018: 2411-2419.
13	ZHU Q X, UDDIN M Y S, VENKATASUBRAMANIAN N, et al. Spatiotemporal scheduling for crowd augmented urban sensing[C]//Proceedings of IEEE Conference on Computer Communications. Washington D.C., USA: IEEE Press, 2018: 1997-2005.
14	YANG Y J , LIU W B , WANG E , et al. A prediction-based user selection framework for heterogeneous mobile crowdsensing. IEEE Transactions on Mobile Computing, 2019, 18 (11): 2460- 2473. doi: 10.1109/TMC.2018.2879098
15	XIAO M J , WU J , HUANG L S , et al. Online task assignment for crowdsensing in predictable mobile social networks. IEEE Transactions on Mobile Computing, 2017, 16 (8): 2306- 2320. doi: 10.1109/TMC.2016.2616473
16	姚玉鹏, 魏立斐, 张蕾. 一种隐私保护的抗投毒攻击联邦学习方案. 计算机工程, 2025, 51 (6): 223- 235. doi: 10.19678/j.issn.1000-3428.0069133
	YAO Y P , WEI L F , ZHANG L . A privacy-preserving federated learning scheme against poisoning attack. Computer Engineering, 2025, 51 (6): 223- 235. doi: 10.19678/j.issn.1000-3428.0069133
17	LIU D , DU Z , LIU X , et al. Task-based network reconfiguration in distributed UAV swarms: a bilateral matching approach. IEEE/ACM Transactions on Networking, 2022, 30 (6): 2688- 2700. doi: 10.1109/TNET.2022.3181036
18	YUAN Z X, LI B, LIU J. Can we improve information freshness with predictions in mobile crowd-learning?[C]//Proceedings of IEEE Conference on Computer Communications Workshops. Washington D.C., USA: IEEE Press, 2020: 702-709.
19	DAI Z, LIU C H, YE Y, et al. AoI-minimal UAV crowdsensing by model-based graph convolutional reinforcement learning[C]//Proceedings of IEEE Conference on Computer Communications. Washington D.C., USA: IEEE Press, 2022: 1029-1038.
20	DAI Z, WANG H, LIU C H, et al. Mobile crowdsensing for data freshness: a deep reinforcement learning approach[C]//Proceedings of IEEE Conference on Computer Communications. Washington D.C., USA: IEEE Press, 2021: 1-10.
21	WANG H , LIU C H , YANG H M , et al. Ensuring threshold AoI for UAV-assisted mobile crowdsensing by multi-agent deep reinforcement learning with transformer. IEEE/ACM Transactions on Networking, 2023, 32 (1): 556- 581.
22	SU S, XU H X, WANG L, et al. Age-of-Information driven mobile crowdsensing in wireless edge computing[C]//Proceedings of the 19th International Conference on Mobility, Sensing and Networking (MSN). Washington D.C., USA: IEEE Press, 2023: 40-47.
23	ZHAO H T , WANG H J , WU W Y , et al. Deployment algorithms for UAV airborne networks toward on-demand coverage. IEEE Journal on Selected Areas in Communications, 2018, 36 (9): 2015- 2031. doi: 10.1109/JSAC.2018.2864376
24	ZHANG L , CELIK A , DANG S P , et al. Energy-efficient trajectory optimization for UAV-assisted IoT networks. IEEE Transactions on Mobile Computing, 2021, 21 (12): 4323- 4337.
25	GONG Z Z , HASHHASH O , WANG Y Z , et al. UAV-aided lifelong learning for AoI and energy optimization in non-stationary IoT networks. IEEE Internet of Things Journal, 2024, 11 (24): 39206- 39224. doi: 10.1109/JIOT.2024.3406220
26	LOWE R, WU Y, TAMAR A, et al. Multi-agent actor-critic for mixed cooperative-competitive environments[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. New York, USA: ACM Press, 2017: 6382-6393.
27	ZHANG H, DU J, JIANG C, et al. MATD3-based joint user association and resource allocation in UAV networks[C]//Proceedings of IEEE Global Communications Conference. Washington D.C., USA: IEEE Press, 2023: 6892-6897.
28	CHEN L, GONG G, JIANG K, et al. DDPG-based computation offloading and service caching in mobile edge computing[C]//Proceedings of IEEE Conference on Computer Communications Workshops. Washington D.C., USA: IEEE Press, 2022: 1-6.

[1]	孟芸, 魏帆, 刘鑫一, 王威, 柴佳辉. 空地协同UDN中基于Graph-DRL的频谱资源分配[J]. 计算机工程, 2026, 52(8): 283-292.
[2]	郭宇新, 贾向东, 李月. 无人机辅助智能反射面系统的安全通信策略[J]. 计算机工程, 2026, 52(8): 293-301.
[3]	董良才, 岳涵词, 王维娟. 基于Actor-Critic深度强化学习的堆场分布式资源联合调度优化[J]. 计算机工程, 2026, 52(7): 421-433.
[4]	曾安, 郑嘉裕, 潘丹, 赵靖亮, 黄幸青. 基于深度强化学习的主动脉夹层中心线追踪算法[J]. 计算机工程, 2026, 52(6): 414-424.
[5]	王怡, 覃团发, 韦睿, 黄金宝. SAG-MEC网络下支持WPT的无人机动态任务卸载与资源分配[J]. 计算机工程, 2026, 52(5): 371-382.
[6]	李明明, 潘子豪. 基于改进TD3算法的移动机器人路径规划[J]. 计算机工程, 2026, 52(5): 150-159.
[7]	赵庶旭, 周宏泽, 王小龙. 基于改进DQN的最优联盟结构生成策略优化[J]. 计算机工程, 2026, 52(5): 117-128.
[8]	汤伟博, 方强, 李沛根, 艾龙金, 熊金红, 夏海廷. 基于RSD-YOLO的无人机航拍图像小目标检测[J]. 计算机工程, 2026, 52(4): 214-228.
[9]	刘义, 罗淳, 钟伟锋, 余意, 欧智清. 多无人机自适应合作任务卸载决策[J]. 计算机工程, 2026, 52(4): 339-348.
[10]	李斌, 郭毅. 面向异构多背包问题的深度强化学习算法[J]. 计算机工程, 2026, 52(4): 140-162.
[11]	江泰民, 谭泰, 李辉, 张建伟, 化晨昊, 董志强. 基于分层深度强化学习的六自由度固定翼无人机路径跟踪方法[J]. 计算机工程, 2026, 52(4): 90-102.
[12]	田海龙, 贾向东, 张兴元, 常恒. 无人机辅助无线传感器网络数据采集的信息年龄-能量权衡方案[J]. 计算机工程, 2026, 52(4): 358-365.
[13]	唐克, 魏飞鸣, 李东瀛, 郁文贤. 基于改进YOLOv8的轻量化无人机图像目标检测算法[J]. 计算机工程, 2026, 52(3): 97-106.
[14]	孔思维, 叶永金, 吴永政, 汪士, 侯杰, 倪明. 无人机群航迹规划问题的量子退火算法求解[J]. 计算机工程, 2026, 52(3): 308-317.
[15]	方仪豪, 邹丹平. 多旋翼无人机仿真平台综述[J]. 计算机工程, 2026, 52(2): 13-23.

选择文件类型/文献管理软件名称

选择包含的内容

AoI驱动的空-地协同移动群智感知与计算

AoI-Driven Air-Ground Collaborative Mobile Crowd Sensing and Computing

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 13

参考文献 28

相关文章 15

编辑推荐

Metrics

本文评价

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

AoI驱动的空-地协同移动群智感知与计算

AoI-Driven Air-Ground Collaborative Mobile Crowd Sensing and Computing

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 13

参考文献 28

相关文章 15

编辑推荐

Metrics

本文评价