Tracking multiple targets with adaptive swarm optimization

Abstract

This paper mainly concentrates on the problem of tracking multiple targets in the noisy environment. To better recognize the eccentric target in a specific environment, one proposed objective function gets the target's shape in the subgraph. Inspired by particle swarm optimization, the proposed algorithm of tracking multiple targets adaptively modifies the covered radii of each subgroup in terms of the minimum distances among the subgroups, and successfully tracks the conflicting targets. The theoretic results as well as the experiments on tracking multiple ants indicate that this efficient method has successfully been applied to the complex and changing practical systems. © 2011 Springer-Verlag.