基于观测相似性粒子滤波的纯角度目标跟踪

单站被动纯角度目标跟踪问题在理论和实际应用中一直都是被广泛研究的一个重要而困难的课题。本文提出一种重采样基于观测路径相似性的粒子滤波算法，该算法利用粒子观测路径和系统状态观测路径的相似性来修正粒子权值并使用修正后的权值进行重采样。实验结果表明当观测噪声方差小于系统噪声方差时，该算法在单站被动纯角度目标跟踪的误差均方根值和误差方差上均优于SIR粒子滤波、辅助粒子滤波和高斯粒子滤波。

A particle filter with the resampling based on the similarity of observation path for single station passive target tracking with bearing-only measurement

Single station passive target tracking with bearing-only measurement is an important and difficult problem whose theory and application have been researched widely. Over the last years, there has been significant interest of particle filter for the problem. A particle filter with the resampling based on the similarity of observation path is presented in this paper. In this proposed algorithm the weights of particles are modified with the similarity between the observation path of the particle and the observation path of the system state. When the variance of observation noise is less than the variance of system noise, the simulation results indicate that the proposed particle filter has better root mean squared error and variance of error over SIR, APF and GPF.