密集杂波下的双门限核化聚类JPDA算法OA
The Dual-threshold Kernelized Clustering Joint Probabilistic Data Association Algorithm for Dense Clutter Scenarios
针对联合概率数据关联算法的"关联爆炸"现象,提出一种双门限核化聚类JPDA算法.基于椭圆跟踪门结合目标速度约束引入速度跟踪门,减少落入跟踪门内的量测数量.利用核函数将数据映射至高维空间,并且放宽对隶属度的约束条件.引入公共量测修正因子,修正公共量测的关联概率.仿真结果表明,该算法在运行效率和跟踪精度方面均有提升.
Aiming at the"association explosion"problem of the Joint Probabilistic Data Association(JPDA)algorithm,a Dual-Threshold Kernelized Clustering-based JPDA(DTKC-JPDA)algorithm is proposed.Firstly,a velocity tracking gate is introduced on the basis of the elliptical tracking gate combined with the target velocity constraints to reduce the number of measurements falling within the tracking gate.Secondly,the kernel functions are utilized to map the data into a high-dimensional space,and the constraints on the membership degree are relaxed.Finally,a common-measurement correction factor is introduced to adjust the association probability of the common measurements.The simulation results demonstrate that the algorithm achieves improvements in both operational efficiency and tracking accuracy.
常金瑞;张安琳;黄子奇;韩继辉;黄道颖
郑州轻工业大学计算机科学与技术学院,郑州 450001郑州轻工业大学工程训练中心,郑州 450001北方信息控制研究院集团有限公司,南京 211153郑州轻工业大学计算机科学与技术学院,郑州 450001郑州轻工业大学计算机科学与技术学院,郑州 450001
信息技术与安全科学
数据关联目标跟踪联合概率数据关联算法核函数模糊聚类
data associationtarget trackingJPDA algorithmkernel functionsfuzzy clustering
《火力与指挥控制》 2026 (1)
42-48,7
国家科技支撑计划资助项目(2006BAK01A38)
评论