UAV Delivery Planning Based on K-Means++ Clustering and Genetic Algorithm

Abstract

This paper considers the optimized schedule for UAV fleets in the scenario of delivery. The target is constructing a plan including the mission allocation for each drone, the routing schedule and the number of drones. Additional factors been considered are the change of the unit energy consumption with the payload of the drone. To realize the goal, K-means++ clustering are applied to designate the mission area for each drone and genetic algorithm are adopted to make the flight path. Moreover, the number of drones are determined by two considerations respectively which are the minimum number of drones and the minimum energy consumption.