基于双层无人机网络的节能计算研究  

Research on Energy Saving Calculation Based on Double Layer Unmanned Aerial Vehicle Networks

作　　者：李盼 黄军锋 樊韩文 马忠贵[1] LI Pan;HUANG Junfeng;FAN Hanwen;MA Zhonggui(School of Computer and Communication Engineering,University of Science and Technology Beijing,Beijing 100083,China)

机构地区：[1]北京科技大学计算机与通信工程学院,北京100083

出　　处：《北方工业大学学报》2024年第1期20-28,共9页Journal of North China University of Technology

基　　金：中央高校基本科研业务费专项资金资助项目(FRF-DF-20-12,FRF-GF-18-017B)。

摘　　要：本文提出一种双层无人机辅助计算和供能系统,该系统将移动边缘计算(Mobile Edge Computing,MEC)的基站迁移至低层无人机上为地面用户设备提供服务,而高层无人机利用太阳能和激光充电技术,在保障自身续航的同时为低层无人机补充能量。同时联合优化双层无人机飞行轨迹、用户设备卸载决策及低层无人机充电决策,以最小化双层无人机与用户设备能耗。参考多智能体深度确定性策略梯度(Multi-Agent Deep Deterministic Policy Gradient,MADDPG)算法,设计了一种改进的双层MADDPG算法对问题进行求解。仿真结果表明,该算法可以有效地降低双层无人机剩余能量的衰减速率,从而间接地提升无人机的续航时间。We propose a double⁃layer Multiple Unmanned Aerial Vehicles(Multi⁃UAV)aided computing unloading and energy supply system,which migrates mobile edge computing(MEC)base stations to low⁃layer Unmanned Aerial Vehicles(UAVs)to provide services for ground user equipments,while high⁃layer UAVs use solar energy and laser charging technology to supplement energy for low⁃layer UAVs while ensuring their own endurance.Simultaneously we optimize the flight trajectory of the double⁃layer UAVs,user equipment unloading decisions,and low⁃layer UAVs charging decisions to minimize energy consumption between the double⁃layer UAVs and user equipments.Referring to the Multi⁃Agent Deep Deterministic Policy Gradient(MADDPG)algorithm,a dual layer MADDPG algorithm is designed to solve the problem.Simulation results show that this algorithm can effectively reduce the decay rate of residual energy in the dual layer UAVs,thereby it can indirectly improve endurance time of the UAVs.

关 键 词：无人机 节能计算 太阳能 多智能体深度确定性策略梯度(MADDPG) 激光充电 

分 类 号：TP393.1[自动化与计算机技术—计算机应用技术]