基于改进平方根无迹粒子滤波的RAM类目标跟踪

RAM target tracking based on improved SR-UPF algorithm

RAM类目标运动复杂,对跟踪系统提出了更高的要求。针对粒子滤波在解决RAM类目标跟踪过程中存在的粒子退化等问题,通过无迹变换计算得出后验概率密度函数充当粒子滤波的先验概率密度函数,提高后验概率密度函数拟合程度,最终提高粒子的有效性。同时,在无迹变换过程中采用qr分解取代部分Cholesky分解,使输出信号方差矩阵保持非负性。最后,针对一段RAM类目标轨迹建立动力学模型,通过Matlab进行仿真,验证了该算法的有效性。

RAM target has complex motion,which puts forward higher requirements for tracking system.Aiming at particle degradation and other problems existing in the process of particle filter solving RAM target tracking,this paper calculates the posterior probability density function as the prior probability density function of the particle filter through unscented transformation,so as to improve the fitting degree of the posterior probability density function and finally improve the particle effectiveness.At the same time,qr decomposition is used to replace part of Cholesky decomposition in the process of unscented transformation,so that the variance matrix of the output signal remains non-negative.Finally,this paper establishes a dynamic model for a reentry target trajectory,and simulates with Matlab to verify the effectiveness of the proposed algorithm.