摘要
提出了一种元启发式数据关联的多目标跟踪方法。首先,该方法根据跟踪门逻辑确定目标的有效量测。然后,利用滤波信息的似然函数描述量测点与目标之间的关联关系,并建立约束条件下多目标数据关联模型。最后,对蚁群优化算法进行改进设计,引入量测剔除策略,将求解问题转化为无约束的组合优化形式,从而利用蚁群优化算法在离散空间的启发式机制搜索量测与目标的最佳关联。仿真结果表明,该方法可以有效实现多目标数据关联且计算量较小,具有一定的工程实用价值。
A multiple target tracking method based on metaheuristic data association is proposed.Firstly,the tracking gate logic is used to confirm the effective measurements.Then,the association relation between measurements and targets is described by the likelihood function of filter innovation,and the multiple targets data association model is estabilished under constraints.Basis of which,a measurement the elimination way is adopted,and the ant colony optimization(ACO) algorithm is designed to translate the data association problem to be solved into the form of combination optimization problem without constraints.After that,the metaheuristically searching ability of the ACO alorithm is used to search the optimal association in discrete area.The simulation results prove that the metaheuristic method,which has less computation,is reasonable and effective for data association and has the engineering application value.
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2010年第10期2176-2180,共5页
Systems Engineering and Electronics
基金
"十一五"国防预研基金(KJ-050402011)
航空科学基金(20085196011)资助课题
关键词
元启发式
数据关联
多目标跟踪
蚁群优化
metaheuristic
data association
multiple target tracking
ant colony optimization(ACO)