Heterogeneous Fleet Allocation and Route Planning Using Weighted Voronoi Diagrams

The introduction of heterogeneous fleets into modern search operations presents a critical challenge of determining the ideal route for a fleet given vehicle and sensor characteristics in order to minimize search time and maximize targets found. Traditional search patterns, such as the expanding squares and lawn mowing patterns, are not sufficient to exploit the capabilities of a hetergenous fleet, nor are they flexible to incorporate prior intelligence into mission planning. To address these problems, a parametric framework was developed to evaluate advanced routing solutions utilizing Voronoi diagrams to split a region based on vehicle and sensor parameters. Agent-based simulation was utilized in a design of experiments to model the performance of the heterogeneous fleet. A full factorial experiment varied the fleet composition, target placement, routing pattern, and fleet size to explore the differences between homogeneous and heterogeneous fleets, as well as the traditional and advanced routing solutions. To enable rapid comparisons between each case and visualize top results, an interactive environment was developed in Python including filters and summary plots enabling swift visual comparisons between cases. It is observed from the study that heterogeneous fleets are superior at exploiting advanced search patterns and enabled greater flexibility to route aircraft under various search priorities.