无人机网络计算卸载和轨迹控制联合优化策略  

Joint Optimization Strategies for Computational Offloading and Trajectory Control in UAV Networks

作　　者：黄嘉伟 黎海涛[1] 李哲超 HUANG Jia-wei;LI Hai-tao;LI Zhe-chao(School of Information Science and Technology,Beijing University of Technology,Beijing 100124,China)

机构地区：[1]北京工业大学信息科学技术学院,北京100124

出　　处：《中国电子科学研究院学报》2024年第7期615-621,共7页Journal of China Academy of Electronics and Information Technology

基　　金：航空科学基金(2018ZC15003)。

摘　　要：无人机网络中计算卸载和轨迹控制互耦,对二者进行联合优化设计有助于提升其整体性能。为此,本文以最小化无人机系统时延和能耗为目标,研究了计算任务卸载和轨迹控制的联合优化问题。首先,将该问题分解为计算任务卸载子问题与轨迹控制子问题;然后,提出基于种群多样性粒子群优化与多评论家深度确定性策略梯度(PDPSO-MCDDPG)的求解算法,通过在DDPG框架中引入多Critic(MC)网络,可减缓单个Critic网络引起的异常波动,进而学习到最优策略。仿真实验表明,所提基于PDPSO-MCDDPG的计算卸载和轨迹控制联合优化策略能够有效降低无人机系统处理时延和能耗。Computational offloading and trajectory control are mutually coupled in UAV networks,and joint design of these two aspects can enhance their overall performance.To this end,this paper investigates the joint optimization problem of computation task offloading and trajectory control with the objective of minimizing the delay and energy consumption of UAV systems.First,the problem is decomposed into a computation task offloading subproblem and a trajectory control subproblem.Then,a solution algorithm based on population diversity particle swarm optimization with multi-critic deep deterministic policy gradient(PDPSO-MCDDPG)is proposed.The introduction of multi-critic(MC)networks in the DDPG framework mitigates the abnormal fluctuations caused by a single critic network,thus enable to achieve the optimal policy.Simulation results indicate that the proposed joint optimization strategy of computation offloading and trajectory control based on the PDPSO-MCDDPG algorithm can effectively reduce the processing delay and energy consumption of the UAV system.

关 键 词：UAV网络 计算卸载 轨迹控制 多评论家 深度确定性策略梯度 

分 类 号：TN92[电子电信—通信与信息系统]