An improved particle filter algorithm based on swarm intelligence optimization

Abstract

Due to the problem of particle degeneracy and loss of particle diversity in particle filter algorithm, this paper proposes an improved particle filter algorithm referring to some ideas of optimized algorithm like particle swarm and firefly group. The proposed algorithm utilizes the firefly algorithm to optimize particle filter and avoid re-sampling process; makes the particles move towards the location of better weight particles and prevents the small weight particles from disappearing after several iterations. Meanwhile, this algorithm sets a transition threshold and iteration times in order to improve the real-time property of the algorithm. Experimental results show that the improved algorithm possesses higher estimation accuracy and keeps good diversity of particle.