Genetic algorithms for positioning and utilizing sensors in synthetically generated landscapes

Abstract

Positioning multiple sensors for acquisition of a a given environment is one of the fundamental research areas in various fields, such as military scouting, computer vision and robotics. In this paper, we propose a framework for locating an configuring a set of given sensors in a synthetically generated terrain with multiple objectives of maximization of visibility of the terrain, maximization of stealth of the sensors and minimization of cost of the sensors. Because of their utility-independent nature, these complementary and conflicting objectives are represented by a multiplicative global utility function based on multi-attribute utility theory. In addition to theoretic foundations, we also present how a Genetic Algorithms can be applied to maximize the global utility function for a given terrain.