An improved Particle Swarm Optimization Particle Filtering algorithm

Abstract

To solve the problem in Particle Swarm Optimization Particle Filtering algorithm that it can be premature too easy to fall into local minima search results, this paper proposes the Particle Swarm Optimization based Fission Bootstrap Particle Filter algorithm. The algorithm first search the space for the particle set by Particle Swarm Optimization, it makes the particle set move to a higher likelihood region, then uses fission bootstrap process to improve the diversity of the particle set, that avoids Particle Degradation caused by premature of the Particle Swarm Optimization. In Monte Carlo simulation of tracking moving target, the tracking performance of Particle Swarm Optimization based Fission Bootstrap Particle Filter algorithm is better than Particle Filtering algorithm in root mean square error(RMSE), also better than that of Particle Swarm Optimization Particle Filtering algorithm.