Optimal Deployment Strategy of Sensing Platform Based on Multi-Objective Genetic Algorithm

Without a methodology or procedure to assist in determining an effective deployment strategy of sensing platforms such as unmanned submarine or sonar matrix loaded with all sorts of sensors, naval units will not achieve the highest level of situational awareness and understanding. This paper addresses the problem of designing objective functions for autonomous surveillance based on multi-objective genetic algorithm (MOGA). The objective functions such as detection probability, survivability and recognition rate can be thought of as different and most often conflicting objectives of our deployment problem and is treated as the basic input to the genetic algorithm. Different fitness objectives and parameters depending on the problem's characteristics were tested using a design of experiment approach. The proposed methodology generates several non-dominated Pareto optimal solutions. Some decision support techniques such as Analytical Hierarchical Process can be used to select one of these solutions.