一种基于模体的无人机集群智能任务规划方法  

Motif-based Intelligent Mission Planning Method for UAV Swarms

作　　者：王维平[1] 王涛[1] 李小波 刘佳杰 Wang Weiping;Wang Tao;Li Xiaobo;Liu Jiajie(School of System Engineering, National University of Defense Technology, Changsha 410073, China)

机构地区：[1]国防科技大学系统工程学院,湖南长沙410073

出　　处：《系统仿真学报》2018年第4期1211-1220,共10页Journal of System Simulation

基　　金：国家自然科学基金(71373282)

摘　　要：由于无人机集群中通信受到制约,无人机集群配系需要随着任务的改变而变化。如何在有限通信下进行无人机集群的智能任务规划成为研究的热点和难点问题。基于功能建立了无人机集群作战网络模体模型框架,提出了6类不同功能类型的基础模体模型;通过能力需求和信息需求将作战任务映射到完成这项任务需要的模体种类和数量;利用"多维动态列表调度法"进行无人机集群的任务规划策略的自动生成。使用定制的NSGA-Ⅱ算法通过改变任务执行顺序,对作战时间和连接改变量进行任务规划策略的自动优化选择。通过一个案例对本文所提出的方法进行了初步验证。Due to the restriction of communication among UAV swarms, configuration of UAV swarms needs to be changed dynamically along with the task changes. How to do mission planning under limited communication for UAV swarms has become a hot and difficult question. The UA V combat network motif model framework is built based on the function, and six different types of basic motif model are proposed. Through capability demand and communication demand, a task is mapped into types and numbers of motifs which are required to complete the task. The ‘multidimensional dynamic scheduling list＇ method is used to automatically generate mission planning schemes. The NSGA-II algorithm is used to automatically select planning schemes towards optimizing the operational time and the quantity of changed connections through changing task execution order. The proposed method is validated through a case study.

关 键 词：无人机集群 模体 多维动态列表调度 NSGA-Ⅱ算法 

分 类 号：C935[经济管理—管理学]