To address the challenges of low offloading success rates and inefficient data transmission in the internet of vehicles (IoV), this paper proposes a multi-layer distributed dynamic offloading strategy for edge computing tasks in IoV based on multi-agent deep reinforcement learning. Firstly, a multi-layer distributed internet of vehicles edge computing system model is designed by integrating software defined network and mobile edge computing. The system model can realize collaborative scheduling optimization at different levels, which can better meet the needs of dynamic allocation of mobile vehicle resources and real-time processing of tasks. Then, considering the success rate of offloading and data transmission rate of vehicle computing tasks, a multi-agent deep reinforcement learning algorithm framework is proposed. The algorithm framework uses collaborative learning of multi-agent systems to enable the vehicle edge system to independently select the optimal task offloading decision. At the same time, the optimization of the action space search and the priority experience replay mechanism were introduced to further improve the effective search of the action space and the stability and accuracy of the task offloading decision. Finally, based on the above algorithm framework and optimization mechanism, a multi-layer distributed vehicle task offloading decision optimization algorithm is proposed. The algorithm can ensure that the vehicle can complete the computing task offloading with the minimum task transmission time and effective offloading success rate according to the current network status and task size. Simulation results show that, compared with the existing offloading methods, the proposed method improves the success rate of computing task offloading by 5%~20% and the efficiency of data transmission by 17.8% on average.
随着自动驾驶和辅助驾驶等技术融入智能交通系统,车联网应用变得越来越多样化.在相关的车载边缘计算场景中,由于车辆的高速移动性使得车载边缘系统网络拓扑快速变化,对资源分配造成了很大的不确定性.为了解决此类问题,目前已有研究通过利用软件定义网络(software defined network,SDN)和移动边缘计算(mobile edge computing,MEC)来提升车联网的性能.SDN技术通过解耦控制层和数据层,提高了对车辆网络的管理和扩展,而MEC则将计算资源和存储资源推向车辆附近的边缘,减少了数据传输的时延,可以更好地满足车联网应用对低延迟和高计算能力的要求[1-2].
YUANT T, ROCHA NETO WDA, ROTHENBERGC E,et al .Dynamic controller assignment in software defined internet of vehicles through multi-agent deep reinforcement learning[J].IEEE Transactions on Network and Service Management, 2021, 18(1): 585-596.
LIG Y, XUEX, LIUY, et al. ImprovedTD3 edge computing offloading strategy for software defined networking internet of vehicles[J].Computer Integrated Manufacturing Systems, 2023,29(5): 1627-1634.(in Chinese)
[4]
HANS D, CHENY Q, CHENG H,et al. Multi-step reinforcement learning-based offloading for vehicle edge computing[C]//2023 15th International Conference on Advanced Computational Intelligence (ICACI). Seoul,Korea, IEEE, 2023:1-8.
[5]
ALAMM Z, JAMALIPOURA .Multi-agent DRL-based Hungarian algorithm (MADRLHA) for task offloading in multi-access edge computing internet of vehicles (IoVs)[J].IEEE Transactions on Wireless Communications, 2022, 21(9): 7641-7652.
[6]
LIUH Q, ZHAOH B, GENGL W,et al .A distributed dependency-aware offloading scheme for vehicular edge computing based on policy gradient[C]//2021 8th IEEE International Conference on Cyber Security and Cloud Computing (CSCloud)/2021 7th IEEE International Conference on Edge Computing and Scalable Cloud (EdgeCom). Washington,D C,USA. IEEE, 2021: 176-181.
[7]
SUNY L, WUZ Y, SHID Y, et al .Task offloading method of internet of vehicles based on cloud-edge computing[C]//2022 IEEE International Conference on Services Computing (SCC). Barcelona, Spain. IEEE, 2022: 315-320.
[8]
WANGS S, XINN, LUOZ Y, et al. An efficient computation offloading strategy based on cloud-edge collaboration in vehicular edge computing[C]//2022 International Conference on Computing,Communication,Perception and Quantum Technology (CCPQT). Xiamen,China. IEEE,2022:193-197.
[9]
SOUAA, TOHMES .Multi-level SDN with vehicles as fog computing infrastructures: a new integrated architecture for 5G-VANETs[C]//2018 21st Conference on Innovation in Clouds,Internet and Networks and Workshops (ICIN). Paris,France. IEEE,2018:1-8.
[10]
LINC, HANG J, JIANGJ F,et al .Underwater pollution tracking based on software-defined multi-tier edge computing in 6G-based underwater wireless networks[J]. IEEE Journal on Selected Areas in Communications, 2023, 41(2): 491-503.
[11]
WANGY Q, HUX, GUOL J,et al .Research on V2I/V2V hybrid multi-hop edge computing offloading algorithm in IoV environment[C]//2020 IEEE 5th International Conference on Intelligent Transportation Engineering (ICITE). Beijing, China. IEEE, 2020: 336-340.
[12]
ZHAON, LIANGY C, NIYATOD,et al .Deep reinforcement learning for user association and resource allocation in heterogeneous cellular networks[J]. IEEE Transactions on Wireless Communications,2019,18(11):5141-5152.
[13]
SHARMAH, KUMARN, TEKCHANDANIR .Mitigating jamming attack in 5G heterogeneous networks:a federated deep reinforcement learning approach[J]. IEEE Transactions on Vehicular Technology, 2023, 72(2): 2439-2452.
[14]
LIUY B, MAOB M .On a novel content edge caching approach based on multi-agent federated reinforcement learning in internet of vehicles[C]//2023 32nd Wireless and Optical Communications Conference (WOCC). Newark, NJ, USA. IEEE, 2023: 1-5.
[15]
GONGY Q, HAOF, WANGL,et al .A socially-aware dependent tasks offloading strategy in mobile edge computing[J].IEEE Transactions on Sustainable Computing, 2023, 8(3):328-342.
ZHUS F, ZHAOM Y, CHAIZ Y. Computing offloading scheme based on particle swarm optimization algorithm in edge computing scene[J]. Journal of Jilin University (Engineering and Technology Edition), 2022, 52(11): 2698-2705.(in Chinese)
[20]
GAURK, GROVERJ. Exploring VANET using edge computing and SDN[C]//2019 Second International Conference on Advanced Computational and Communication Paradigms (ICACCP).Gangtok,India. IEEE,2019:1-4.
[21]
WANGB X, LIUL, WANGJ .Multi-agent deep reinforcement learning for task offloading in vehicle edge computing[C]//2023 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB). Beijing, China. IEEE, 2023: 1-6.
ZHUS F, HUJ M, YANGC R, et al. Optimization of priority task based unloading decision in the context of edge computing of the internet of things[J]. Journal of Jilin University (Engineering and Technology Edition):2024,54(11): 3338-3350. (in Chinese)
[28]
TANGS H, CHENB, IWENH,et al .VECFrame:a vehicular edge computing framework for connected autonomous vehicles[C]//2021 IEEE International Conference on Edge Computing (EDGE). Chicago, IL, USA. IEEE, 2021: 68-77.
[29]
CHENX W, ZHANGQ Y, JINZ G, et al. Research on intelligent vehicle infrastructure cooperative system based on 5G mobile edge computing[C]//2021 6th International Conference on Transportation Information and Safety (ICTIS). Wuhan,China. IEEE, 2021:21-27.
[30]
TAOX, HAFIDA S. DeepSensing: a novel mobile crowdsensing framework with double deep Q-network and prioritized experience replay[J]. IEEE Internet of Things Journal,2020, 7(12): 11547-11558.
FANGW W, WANGY P, ZHANGH, et al. Optimization of communication resource allocation in connected vehicles based on multi agent deep reinforcement learning[J]. Journal of Beijing Jiaotong University, 2022, 46(2): 64-72.(in Chinese)
基金资助
国家自然科学基金资助项目(61862037)
国家自然科学基金资助项目(62262038)
National Natural ScienceFoundation of China(61862037)
National Natural ScienceFoundation of China(62262038)
甘肃省科技计划项目(23CXGA0028)
Gansu Provincial Science and Technology Plan Project(23CXGA0028)
兰州市人才创新创业项目(2021-RC-40)
Lanzhou Talent Innovation and Entrepreneurship Project(2021-RC-40)