Vortex Particle Swarm Optimization

This paper presents an optimization algorithm based on self-propelled particle swarms which exploit vorticity features in order to avoid local minima; the proposed algorithm is termed Vortex Particle Swarm Optimization (VPSO). The optimization algorithm switches between translational and dispersion behavior of the swarm to enhance the exploration of the search space and to avoid getting trapped in local minima. These two types of behavior are induced by choosing the swarm as a collection of coupled, second-order oscillators where it is possible, via suitable parameter selection to switch between translational (convergence) and vortex-like movements (dispersion). This idea mimics living organism strategies such as foraging and predator avoidance. Performance of the algorithm is studied via simulation results of well-known 2D test functions.