基于欧氏距离的无人艇载毫米波雷达点迹凝聚方法  

Plots-Centroid Method for USV-Borne Millimeter-Wave Radar Based on Euclidean Distance

作　　者：李瑞伟 李立刚[2] 金久才 刘德庆 李方旭 戴永寿[2] LI Rui-wei;LI Li-gang;JIN Jiu-cai;LIU De-qing;LI Fang-xu;DAI Yong-shou(College of Control Science and Engineering,China University of Petroleum(East China),Qingdao 266580,China;College of Ocean and Space Information,China University of Petroleum(East China),Qingdao 266580,China;Laboratory of Marine Physics and Remote Sensing,First Institute of Oceanography,Ministry of Natural Resources,Qingdao 266061,China)

机构地区：[1]中国石油大学(华东)控制科学与工程学院,山东青岛266580 [2]中国石油大学(华东)海洋与空间信息学院,山东青岛266580 [3]自然资源部第一海洋研究所,山东青岛266061

出　　处：《水下无人系统学报》2020年第6期604-610,617,共8页Journal of Unmanned Undersea Systems

基　　金：国家重点研发计划(2017YFC1405203);中央高校基本科研业务费专项资助(19CX05003A-1).

摘　　要：毫米波雷达在工作中会存在虚警和目标点迹分裂等问题。当前多数研究均通过处理雷达回波,根据波形特征降低虚警,但结果中仍存在无效点迹;通过依次比较点迹距离和角度等多维信息判断其是否属于同一目标,方法过程较为繁琐,且表示目标状态常用的质心法不够准确。因此文中提出一种基于欧氏距离的点迹凝聚方法,通过对雷达点迹数据进行处理解决了虚警和目标点迹分裂问题。首先,结合雷达回波强度和有效检测范围等先验知识,采用阈值滤波法去除无效点迹。然后,依据“属于同一个目标的点迹速度相同、距离相近”这一规律,通过欧氏距离度量点迹间的信息相似度,实现目标点迹聚类。最后,计算障碍目标的位置和截面宽度等信息,以矩形危险区域表示障碍目标所属范围,从而实现无人艇前方障碍目标的准确检测。实船试验验证了文中方法的有效性。To solve the problems of false alarm and target-plot splitting of millimeter-wave(MMW)radar used in the detection of obstacles in unmanned surface vessels(USVs)on the sea,most recent studies have reduced false alarms by processing radar echo based on waveform characteristics.These studies have also compared multi-dimensional information such as the distance and angle of plots to determine whether the plots belong to the same target,this studies are cumbersome.And the centroid methods are not sufficiently accurate to represent the target state.Therefore,this study proposes a plots-centroid method based on Euclidean distance that solves the problems of false alarm and target-plot splitting by processing radar plot data.First,using prior knowledge related to radar echo intensity and effective detection range,a threshold filtering method is used to remove invalid plots.Then,based on the rule stating that plots belonging to the same target have the same speed and close distance,the Euclidean distance is used to measure the similarity of information between plots.Finally,target-plot clustering is realized.Finally,the position and section width of the obstacle target are calculated,where the range of the obstacle target is denoted by a rectangular dangerous area to enable accurate detection of the obstacle target in front of the USV.The effectiveness of this method is verified by an actual ship test conducted.

关 键 词：无人艇 毫米波雷达 障碍目标检测 点迹凝聚 

分 类 号：TJ630[兵器科学与技术—武器系统与运用工程] U674.91[交通运输工程—船舶及航道工程]