混合粒子滤波被动定位算法

On passive locating algorithm with hybrid particle filter method

为实现空中机动单站对运动目标的快速无源定位,提出了一种基于测向交叉定位方法的混合粒子滤波算法。针对粒子滤波初始化存在的问题,该算法从测向交叉定位方法得到建议分布,利用角度测量对状态变量的约束关系产生所需粒子,避免了粒子滤波产生粒子时的随机性,减少了高维情况下所需的粒子数目,降低了运算成本,提高了粒子滤波的实时性能。与粒子滤波的仿真对比分析表明:该算法的定位性能优于其他方法。

In order to fulfill the passive location moving targets by the aerial mobile platform, a new hybrid particle filter positioning algorithm was proposed based on the bearings crossing locating me- thod. According to the problems existing in initializing particle filter, the new algorithm obtains the proposed distribution from the bearings crossing locating method and generates the required particles by the constraints of angle measurement to the state variables. Thus, it can avoid the randomness in generating particles and reduce the number of required particles in high dimensional cases. As a result, the present algorithm reduces the computing cost and improves the real-time performance of the particle filter. The simulation analysis of particle filter show that the new algorithm is superior to the other methods in positioning.