一种强抗噪的连续帧无人机集群时序表征方法  

A strong anti-noise time series representation method for continuous frame UA V swarms

作　　者：肖晖 赵辰霆 王博[1] 盛庆红[1] 刘慧臻 Xiao Hui;Zhao Chenting;Wang Bo;Sheng Qinghong;Liu Huizhen(College of Astronautics,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China;School of environment science,Nanjing Xiaozhuang University,Nanjing 211171,China)

机构地区：[1]南京航空航天大学航天学院,南京210016 [2]南京晓庄学院环境科学学院,南京211171

出　　处：《空天技术》2023年第5期72-78,86,共8页Aerospace Technology

基　　金：国家自然科学基金项目(41701531)。

摘　　要：针对敌方无人机来袭时,其集群队形识别易受传感器噪声等外界干扰的情况,提出了一种提取无人机集群连续帧内部关系结构特征的时序表征方法。该方法充分利用无人机集群的方位和距离两个主要元素,对连续帧无人机集群内部关系结构特征进行描述,利用帧与帧之间的参数变化趋势对无人机集群进行时序表征。通过仿真对比实验证明,该方法对无人机集群的特征描述具有较好的抗噪能力和鲁棒性,能够有效解决传感器噪声等外界干扰影响下的队形变形问题,实现无人机集群持续稳定识别。Aiming at the situation that the cluster formation recognition is susceptible to external interference such as sensor noise when the enemy UAV comes,a time series representation method for extracting the internal relationship structure characteristics of continuous frames of UAV cluster is proposed.This method makes full use of the two main elements of the azimuth and distance of the UAV cluster to describe the internal relationship structure characteristics of the UAV cluster in continuous frames,and uses the parameter change trend between frames to characterize the time series of the UAV cluster.Through simulation and comparative experiments,it has been proven that this method has good noise resistance and robustness in describing the characteristics of unmanned aerial vehicle clusters,and can effectively solve the problem of formation deformation under external interference such as sensor noise,achieving continuous and stable recognition of unmanned aerial vehicle clusters.

关 键 词：群目标 无人机集群 内部关系结构参数 时序表征 动态检测 

分 类 号：V19[航空宇航科学与技术—人机与环境工程]