一种基于小生境技术的群智能粒子滤波算法

Swarm intelligence particle filtering based on niching technique

针对基本粒子滤波存在严重的退化问题和重采样技术导致粒子枯竭的问题,提出一种新型粒子滤波算法——基于小生境技术的群智能优化粒子滤波算法.通过多模寻优增强粒子的多样性和寻优能力,使得采样后的粒子向高似然区域移动,从而有效地提高了系统状态估计精度.仿真实验表明,该算法是有效而稳定的.

For the problem that serious weight degeneracy based on particle filtering and re-sampling leads to the sample impoverishment,a new particle filtering algorithm is proposed,which utilizes niching swarm intelligence algorithm for optimizing particle filtering. By multi-modal optimization,particles diversity and search ability are both enhanced. As a result,particles can move towards the high likelihood areas,which can increase the precision of system state estimation. Simulation results show that the proposed algorithm is effective and stable.