基于强跟踪滤波器的多目标跟踪方法

在诸多的多目标跟踪算法中，相互作用多模型（IMM）算法是目前公认的最为有效的算法。但到目前为止，LMM估计方法都是建立在卡尔曼滤波器（KF）和扩展卡尔曼滤波器（EKF）基础上，因而其性能不仅依赖于所采用的模型集，而且在更大程度上依赖于所采用的滤波技术。强跟踪滤波器（STF）克服了卡尔曼和扩展卡尔曼的三大缺陷，因而设计一种基于STF的IMM目标跟踪算法显然能提高其性能。仿真实验表明，基于STF的IMM算法的跟踪性能要优于基于KF和EKF的IMM算法的跟踪性能。

Method of IMM target tracking based on STF

Interactive multiple model(IMM) estimator is a one of most effective method for target tracking,in application the way is based on Kalman filter(KF) or extended Kalman filter(EKF).The performance of IMM is not only depend on the adopted model set,but also rely on the filters.It is all known that there exist three disfigurements,so a modified IMM algorithm is presented which is based on the strong tracking filter to avoid the disfigurements.The simulation results show the performance of target tracking of the IMM algorithm for target tracking based on the strong tracking filter(STF) is better than that one based on KF and EKF.