基于改进遗传算法的多传感器航迹关联  

Multi-sensor Track Association Based on Improved Genetic Algorithm

作　　者：赵翻东 蔡益朝[1] 李浩[1] 师维克 ZHAO Fandong;CAI Yichao;LI Hao;SHI Weike(Department of Early Warning Intelligence,Air Force Early Warning Academy,Wuhan 430019,China;不详)

机构地区：[1]空军预警学院预警情报系,湖北武汉430019

出　　处：《武汉理工大学学报（信息与管理工程版）》2022年第1期105-111,共7页Journal of Wuhan University of Technology：Information & Management Engineering

基　　金：国家自然科学基金项目(61502522);军委装发部装备预研领域基金项目(JZX7Y20190253036101);军委装发部与教育部装备预研教育部联合基金项目(6141A02033703);湖北省自然科学基金项目(2019CFC897).

摘　　要：针对传统航迹关联算法存在收敛速度慢、关联正确率低、运算量随传感器节点数的增加而呈指数增加的问题,对传统遗传算法的编码、选择规则和交叉变异等环节提出了相应的改进策略,然后在加权法统计量的基础上,利用改进遗传算法对结果进行优化,将多传感器多目标航迹关联问题转化为多维分配问题求其最优解。最后对所提方法与多种航迹关联算法进行仿真对比,结果表明该方法能有效解决两局部节点和多局部节点下的航迹关联问题,且所提算法具有较高的关联准确性和较快的收敛速度,充分体现了改进策略的有效性。Aiming at the problems of slow convergence rate,low correlation accuracy and exponential increase of computation with the increase of the number of sensor nodes in traditional track association algorithm,this paper proposes the corresponding improvement strategies for the coding,selection rules and crossover mutation of the traditional genetic algorithm.Then,on the basis of the weighted method,the improved genetic algorithm is used to optimize the results,and the multi-sensor and multi-target track correlation problem is transformed into a multi-dimensional assignment problem to find the optimal solution.Finally,the proposed method is compared with other track correlation algorithms,the results show that this method can effectively solve the problem of track correlation under two local nodes and multiple local nodes.Therefore,the algorithm proposed in this paper has high correlation accuracy and fast convergence rate,which fully reflects the effectiveness of the proposed improved strategy.

关 键 词：航迹关联 关联正确率 遗传算法 多传感器 多目标 多维优化 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]