基于杂波强度在线估计的多目标跟踪算法  

Multiple target tracking based on online estimation of clutter intensity

作　　者：闫小喜[1,2,3] 韩崇昭[1,2,3] 

机构地区：[1]西安交通大学智能网络与网络安全教育部重点实验室, [2]机械制造系统工程国家重点实验室 [3]综合自动化研究所,西安710049

出　　处：《控制与决策》2012年第4期507-512,共6页Control and Decision

基　　金：国家973计划项目(2007CB311006);国家自然科学基金项目(60921003)

摘　　要：针对多目标跟踪中的未知杂波强度,提出了基于熵分布的杂波强度在线估计算法.利用有限混合模型对未知杂波强度建模,将仅依赖于混合权重的熵分布作为混合参数的先验;利用拉格朗日乘子法推导了混合权重在极大后验意义下的在线估计公式;以随机近似过程为在线估计策略,推导了基于缺失数据的分量均值和方差的在线估计公式.仿真结果表明,基于熵分布的杂波强度在线估计算法改进了概率假设密度滤波器在多目标跟踪中的性能.Aiming at the unknown clutter intensity in multiple tracking, online estimation algorithm of clutter intensity based on entropy distribution is proposed. The clutter intensity is modeled by finite mixture model. The entropy distribution, which depends only on mixing weights, is adopted as the prior distribution of mixing parameters. The online estimation formulation of mixing weight is derived by Lagrange multiplier in the sense of maximum a posterior. Stochastic approximation procedure is regarded as the strategy of online estimation of component mean and covariance. The online estimation formulations of component mean and covariance are derived based on missing data. Simulation results show that the online estimation algorithm of clutter intensity based on entropy distribution improves the performance of probability hypothesis density filter in multilple target tracking.

关 键 词：多目标跟踪 概率假设密度 杂波强度 在线估计 熵分布 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]