Genetic Evolution Approach for Target Movement Prediction

This paper presents a genetic evolution system, for target movement prediction, which includes functions inferring opponents’ strategic movements and displaying such predicted movements in an interactive 3D visualization space. To speed up the analysts’ ability to access and integrate information, the prediction approach generates new movements based on past behaviors and application of an inheritance mechanism. It applies Genetic Algorithms (GAs) learning techniques to evolve new individuals in the population of movements in order to converge the evolution process toward optimal movements. The approach is implemented into the GEM (Genetic Evolution of Movement) system and its performance has been experimentally evaluated.