摘要
粒子滤波(Particle Filter,PF)相对于其他滤波算法在处理非线性非高斯类型的系统中具有明显的优势,但是标准粒子滤波本身并不完美,因此提出了一种改进的人工鱼群算法去优化粒子滤波算法。先改进人工鱼群算法中鱼群移动的步长,再将人工鱼群算法的觅食行为和聚群行为引入粒子滤波,使其驱动粒子向高似然区域移动,进而改善粒子分布,有利于解决粒子退化和粒子贫化的问题。最后,将改进的人工鱼群粒子滤波算法与粒子滤波法以及扩展卡尔曼滤波器(Extended Kalman Filter,EKF)算法进行仿真实验对比,结果表明改进的人工鱼群粒子滤波算法在各方面均优于其他两种算法。
Compared with other filtering algorithms,Particle Filter(PF) has obvious advantages in dealing with nonlinear and non Gaussian systems,but standard particle filter itself is not perfect,so an improved artificial fish swarm algorithm is proposed to optimize particle filter algorithm.Firstly,improve the step size of fish swarm movement in the artificial fish swarm algorithm,and then introduce the foraging behavior and clustering behavior of the artificial fish swarm algorithm into particle filtering,so that it can drive particles to move to the high likelihood region,thereby improving the particle distribution,which is conducive to solving the problem of particle degradation and particle dilution.Finally,the improved artificial fish swarm particle filter algorithm is compared with particle filter algorithm and Extended Kalman Filter(EKF) algorithm through simulation experiments.The results show that the improved artificial fish swarm particle filter algorithm is superior to the other two algorithms in all aspects.
作者
郭延鹏
熊林正
顾恩到
GUO Yanpeng;XIONG Linzheng;GU Endao(School of Electrical Engineering,North China University of Water Conservancy and Hydropower,Zhengzhou Henan 450045,China)
出处
《信息与电脑》
2022年第18期67-69,95,共4页
Information & Computer
关键词
人工鱼群
粒子滤波(PF)
状态估计
artificial fish swarm
Particle Filter(PF)
state estimation
