基于MEC的无人物流车任务卸载策略  

MEC Based Workflow Offloading Strategy of Driverless Logistics Distribution Vehicle

作　　者：郜文灿 齐平[1] 束红 蒋剑军[1] GAO Wen-can;QI Ping;SHU Hong;JIANG Jian-jun(Tongling University,Tongling Anhui 244061,China)

机构地区：[1]铜陵学院,安徽铜陵244061

出　　处：《铜陵学院学报》2021年第6期107-111,115,共6页Journal of Tongling University

基　　金：安徽省重点研究与开发计划项目“基于健康大数据的社区慢性病个性化干预关键技术研究”(202004a05020010);安徽省高校科学研究项目“边缘计算机环境下基于区块链的智慧医疗系统身份认证方法研究”(KJ2020A0694)。

摘　　要：移动边缘环境下的计算卸载技术将移动设备的计算任务卸载至边缘服务器或云服务器以解决终端设备在计算性能以及能效等方面的不足。然而现有计算卸载策略大多设定移动设备处于静止状态,并未考虑终端的可移动性。针对小区无人车物流配送场景下的工作流任务卸载问题,文章首先对小区无人车物流配送场景、工作流任务、移动边缘计算系统进行建模描述,再根据无人车位置、移动速率,结合边缘服务器无线信号覆盖范围构建了移动边缘环境下的无人车任务卸载模型。最后使用遗传算法设计了基于最优路径的任务调度算法和无人车最优配送路线算法以计算最优路径和最优配送路线。仿真实验结果说明通过预先规划最优配送路径和配送顺序,合理地分配计算资源,能够在用户响应时间约束下有效降低终端能耗和违约率。Computation offloading in mobile edge computing environment would transfer the resource computational tasks to edge servers or cloud servers.It can deal with the shortage of mobile terminal in computation performance and energy efficiency.However,the most existing computation offloading strategy assume that the mobile devices are not moving,and do not take into account the mobility of the mobile device.In view of the above problem,we firstly described the driverless vehicle distribution scenario,the workflow and the mobile edge computing system.According to the location and the speed of the driverless vehicle and the wireless signal covering area of the edge servers,the offloading model of driverless vehicle was constructed.The genetic algorithm was used to design the workflow offloading strategy and the task scheduling algorithm based on the optimal route.Experimental results showed that the proposed algorithm could effectively low the energy consumption and the default rate with the constraint of response time,on the premise of planning a route in advance.

关 键 词：移动边缘计算 边缘服务器 计算卸载 最优路径 能耗 遗传算法 

分 类 号：TP368.5[自动化与计算机技术—计算机系统结构]