车联网中基于多智能体强化学习的边缘服务器选址策略  

Edge server deployment strategy based on multi-agent reinforcement learning in the internet of vehicles

作　　者：李闯 纪剑桥 胡志刚[2] 周舟 LI Chuang;JI Jianqiao;HU Zhigang;ZHOU Zhou(School of Computing,Hunan University of Technology and Business,Changsha 410205,China;School of Computer Science and Engineering,Central South University,Changsha 410075,China;School of Computer Science and Engineering,Changsha University,Changsha 410022,China)

机构地区：[1]湖南工商大学计算机学院,湖南长沙410205 [2]中南大学计算机学院,湖南长沙410075 [3]长沙学院计算机科学与工程学院,湖南长沙410022

出　　处：《中南大学学报（自然科学版）》2024年第7期2567-2577,共11页Journal of Central South University:Science and Technology

基　　金：国家自然科学基金资助项目(62172442,62002115,62372068);湘江实验室重大项目(23XJ01002,22XJ01001);湖南省教育厅青年项目(21B0779);湖南省重点研发计划项目(2021NK2020);长沙市杰出创新青年培养计划项目(kq2107020);湖南省自然科学基金资助项目(2022JJ40128)。

摘　　要：为解决车联网环境下边缘服务器选址难的问题,提出一种基于多智能体强化学习的边缘服务器部署策略(记为CKM-MAPPO),重点优化边缘服务器间的负载均衡,同时最小化边缘服务器的时延和能耗。首先,使用Canopy和K-means算法确定边缘服务器部署的数量和初始位置;然后,基于多智能体强化学习算法确定边缘服务器的最优部署位置;最后,通过一系列实验评估所提出算法的准确性和有效性。研究结果表明:与基准算法相比,本文提出的方法的负载均衡度提升了26.5%,时延和能耗分别降低了12.4%和17.9%。To solve the hard problem of edge server deployment in internet of vehicle environments,an edge server deployment strategy based on multi-agent reinforcement learning(CKM-MAPPO)was proposed.It focuses on optimizing the load balancing among edge servers and minimizing edge servers'delay and energy consumption.Firstly,the Canopy and K-means algorithms were used to determine the number and initial location of edge server deployment.Then,the multi-agent reinforcement learning algorithm was leveraged to determine the optimal deployment location of the edge server.Finally,the accuracy and effectiveness of the proposed algorithm were evaluated through a series of experiments.The results show that compared with the benchmark algorithm,the proposed method improves load balancing by 26.5%,and the time delay and energy consumption are reduced by 12.4%and 17.9%,respectively.

关 键 词：边缘计算 服务器部署 车联网 负载均衡 强化学习 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]