对于纯方位目标跟踪问题,传统的线性算法已经不能满足非线性非高斯和实时性目标跟踪的要求,由于在纯方位目标跟踪中传统的粒子滤波收敛速度慢并且容易发散,文中提出了一种基于改进的采样-重要性-重采样滤波(SIRF)算法的纯方位跟踪算法。具体的改进方法就是去掉归一化步骤直接使用非归一化权值,该算法在保持高精度估计能力的同时,具有较强的鲁棒性,是解决非线性系统状态估计问题的一种有效方法。最后通过实验验证改进的SIRF算法跟踪效果明显优于高斯粒子滤波(GPF)算法。

In view of bearings-only target tracking problem, traditional linearization algorithm can not meet the requirements of nonlinear, non-Gaussian and real-time target tracking. An improved sampling-importancere-sampling filter(SIRF) algorithm was proposed for highly non-linear bearing-only tracking system where the common particle filters often fail to catch and keep tracking of the emitter. Specific improvements are to remove normalization step and use non-normalized weights directly. While maintaining high accuracy estimation ability at the same time, the algorithm has strong robustness, is an effective solution to state estimation of nonlinear system. Finally, experiments verify that the tracking effect of the improved SIRF algorithm is better than the Gaussian particle filter (GPF) algorithm.