基于序贯重要采样算法的被动单站机动目标跟踪

Sequential importance sampling method for tracking a maneuvering target with a single passive observer

将序贯重要采样(SIS)与交互多模型(IMM)算法相结合,提出了一种新的机动目标跟踪方法———IMM-SIS算法,并将其应用于被动单站跟踪系统,同高斯和粒子滤波器(IMM-GSPF)算法相比,其优点是不需要重采样步骤,也不会出现采样粒子的退化和贫乏现象.通过跟踪一个机动目标的仿真过程,对算法性能进行了检验,结果表明,该算法在计算速度和跟踪精度方面均优于IMM-GSPF算法,同经典的IMM-EFK算法相比,两种算法在鲁棒性和精度上都是优越的.

An IMM-SIS（ Interacting Multiple Model-Sequential Importance Sampling） algorithm is presented for tracking a maneuvering target by a proper combination of two approaches： IMM and SIS, and it is used in the single passive tracking system. The advantage of the presented algorithm over the IMM-GSPF（ Gaussian Sum Particle filter） is that it does not need the resampling step and avoids the particle degeneracy and impoverishment phenomenon. The performance of the IMM-SIS algorithm is verified by simulating a highly maneuvering target tracking. Results show that the tracking speed and accuracy of the IMM-SIS algorithm are better than those of the IMM-GSPF, the tracking robustness and accuracy of the SIS and GSPF algorithms are better than those of the classical EKF.