Optimization of UAV Flight Paths in Multi-UAV Networks for Efficient Data Collection

The traveling salesman problem (TSP), a challenging NP-hard problem, often necessitates the use of metaheuristic algorithms to find approximate solutions. These include techniques such as ant colony optimization (ACO), particle swarm optimization (PSO), artificial bee colony (ABC), genetic algorithm (GA), simulated annealing (SA), and Tabu search (TS). In this study, we explore a variant called the Close Enough TSP, characterized by dynamic targets within the communication ranges of cluster head (CH) nodes. This scenario requires data collection by an unmanned aerial vehicle (UAV), which need not directly overfly each node. We propose a new method that blends metaheuristic algorithms with geometric heuristics to create a near-optimal UAV flight path. Our evaluation, aimed at reducing travel distance across scenarios involving 20, 52, 100, and 150 target areas, shows that the ABC algorithm converges effectively toward optimal solutions. The results validate the efficiency of our approach, underlining its potential for UAV-driven data gathering and related areas.