Session:

LBP - Late Breaking Papers

Title:

Improving on the Kalman Swarm: Extracting its Essential Characteristics

Authors:

Christopher K. Monson
Kevin D. Seppi

Abstract:

The Kalman Swarm (KSwarm) is a new approach to particle motion in PSO that reduces the number of iterations required to reach good solutions. Unfortunately, it has much higher computational complexity than basic PSO. This paper addresses the runtime of KSwarm in a new algorithm called ``Linear Kalman Swarm'' (LinkSwarm) which has linear complexity and performs even better than KSwarm. Some possible reasons for the success of KSwarm are also explored.