Open AccessFirefly optimized particle filter algorithm based on adaptive differential evolution
Author(s) -
Ziwei Meng,
Mei Jin-jie,
Chengjun Yu
Publication year - 2022
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/2187/1/012055
Subject(s) - differential evolution , crossover , particle swarm optimization , firefly algorithm , mathematical optimization , algorithm , process (computing) , local optimum , fitness function , position (finance) , multi swarm optimization , particle filter , computer science , filter (signal processing) , selection (genetic algorithm) , mathematics , genetic algorithm , artificial intelligence , economics , computer vision , operating system , finance
The improved particle filter algorithm based on fireflies reduces the particle diversity in the later iterations, which is easy to fall into the problem of local optimization. To solve the problem, after firefly position updated, we can integrate the differential evolution algorithm into it. The process can be described like this: firstly, the mutation process is guided by the particle weight adaptively; the crossover process is selecting individuals randomly according to the crossover probability; finally, the selection process takes the observation probability density function as the fitness value, and retains individuals with high fitness. After adding differential evolution, the diversity of individuals has increased and jumped out the process of local optimization. The overall quality of the particle swarm has been improved. Experiments show that the tracking accuracy of the improved algorithm has been improved, as well as the global optimization capabilities.