一种集群智能粒子滤波算法

Swarm intelligence algorithm for particle filtering

将集群智能思想引入粒子滤波,提出一种新颖的基于人工鱼群算法的粒子滤波器.该算法利用人工鱼群算法中觅食行为和聚群行为的交替,使得先验粒子不断向高似然域移动,从而改善粒子分布,提高估计精度.此外,利用Kullback信息描述聚群行为产生的粒子分布与似然分布的差别,通过迭代发现Kullback信息是递减的,从而证明该算法是合理的.仿真实验证明,这种算法是一种有效的粒子滤波算法,其滤波性能优于扩展卡尔曼滤波和常规粒子滤波.

By bringing the thought of swarm intelligence into particle filtering, a novel particle filter based on the artificial fish school algorithm is proposed. This algorithm makes prior particles move towards the high likelihood region by use of the alternation of behaviors of preying and swarming in the artificial fish school algorithm. So particle distribution and filtering accuracy are improved. Moreover, the difference between the particle distribution produced by behavior of swarming and the likelihood distribution is described by Kullback information. Kullback information decreases with the increasing iteration degree, which proves that this algorithm is rational. Finally, simulation results show that this swarm intelligence algorithm for particle filtering is effective, and has a better filtering performance than the EKF and the common PF.