改进加权质心算法的集群协同定位理论研究  

Research on Collaborative Location Theory for Cluster Based on Improved Weighted Centroid Algorithm

作　　者：张薇薇 刘美红 陈懋霖 李艳霞[1] 王章 ZHANG Weiwei;LIU Meihong;CHEN Maolin;LI Yanxia;WANG Zhang(School of Automation and Software Engineering,Shanxi University,Taiyuan 237016,China;Institute of Electronic Engineering,China Academy of Engineering Physics,Mianyang 621999,China;14th Branch,Jiuquan City,Gansu Province,Jiuquan 735018,China)

机构地区：[1]山西大学自动化与软件学院,山西太原237016 [2]中国工程物理研究院电子工程研究所,四川绵阳621999 [3]甘肃省酒泉市十四支局,甘肃酒泉735018

出　　处：《无线电工程》2024年第7期1779-1785,共7页Radio Engineering

基　　金：山西省基础研究计划项目(202103021223025);中国工程物理研究院创新发展基金(20210047)。

摘　　要：无人机集群定位信息的获取对于路径规划、探测、制导和协同控制等均具有重要意义,当前通常使用的全球导航卫星系统(Global Navigation Satellite System, GNSS)定位易受到干扰和欺骗的影响而处于拒止状态,基于机间数据链的协同定位的用途日益凸显。目前缺少获得一致认同的在精度和计算量之间获得较好平衡的协同定位算法,针对这一问题,对研究GNSS拒止下基于机间测距的无人机集群协同定位方法进行了简要介绍。搭建集群无人机协同定位分散融合定位架构,针对GNSS拒止下集群无人机定位性能无法定量评估问题,建立集群协同定位精度定量评估模型,构建出典型任务场景下的无人机飞行轨迹和各历元的机间测距信息。通过仿真试验,利用包括加权质心算法、最小二乘法和网格法在内的多种算法解算得到无人机集群的相对位置信息,评估设定的典型场景下3种定位方法定位性能,解决GNSS拒止下集群无人机定位误差快速发散的问题。仿真结果表明,加权质心算法性价比最高,研究结论可为无人机集群定位提供一定的指导和帮助。The acquisition of drone cluster positioning information is of great significance for path planning,detection,guidance,collaborative control,etc.Currently,the commonly used Global Navigation Satellite System(GNSS)positioning is easily affected by interference and deception and is in a denied state.The use of collaborative cooperative positioning based on inter-node data link is increasingly prominent.However,at present,there is still a lack of collaborative positioning algorithm that can achieve a good balance between positioning accuracy and computational burden.To solve this problem,the UAV cluster collaborative positioning method based on inter-node ranging is studied under GNSS denied environment.Firstly,a decentralized fusion positioning architecture for clustered UAV collaborative positioning is built.Then to address the issue of the inability to quantitatively evaluate the positioning performance of cluster drones under GNSS denied environment,a quantitative evaluation model for collaborative positioning accuracy of cluster is established and the UAV flight path and inter-node ranging information of each epoch under typical mission scenarios are constructed.Finally,through the simulation test,the relative position information of UAV cluster is obtained by using various algorithms including weighted centroid algorithm,least square method and grid method.The positioning performance of three positioning methods under the set typical scenario is evaluated.The problem of rapid divergence of positioning error of UAV cluster under GNSS denied environment is solved.Simulation results show that the weighted centroid algorithm has the highest cost performance.The research conclusions can provide some guidance and help for UAV cluster positioning.

关 键 词：全球导航卫星系统拒止环境 机间测距 协同定位 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]