Area Partitioning by Intelligent UAVs for effective path planning using Evolutionary algorithms

Unmanned Aerial Vehicles (UAVs) have emerged as one of the most researched topics in recent times. They are being deployed in ample applications without any human intervention. The use of autonomous UAVs has given rise to many new challenges. Path planning has always been one of the most critical problems when it comes to the deployment of UAVs in real-time applications. The current work focuses on the area partitioning algorithm considering a few significant parameters such as static obstacles and altitude of UAVs. For any given geographic area, the proposed area partitioning algorithm finds a division of the entire area of interest by effectively partitioning into rectangles. The algorithm merges two or more equal-sized rectangles based on the obstacle position, resulting in a rectilinear partitioning of the area. Two mid-points are computed for each partition and are further used for the computation of an optimal path. An optimal path for the UAV is found by constructing a graph from the mid-points, posing the problem as Travelling Salesperson Problem(TSP), and finding a solution using Firefly Algorithm(FA) and Particle Swarm Optimization(PSO). Effective comparison is done to find the optimal path using both techniques. The obtained results show that FA outperforms PSO.