UAV Swarm Trajectory Design for Wireless Networks Using Genetic Algorithm-Driven Repulsion Forces

Uncrewed Aerial Vehicle (UAV) swarms are increasingly recognized for their versatility and affordability. These swarms have enhanced various applications, including agriculture, surveillance, delivery services, and monitoring. However, fully utilizing the capabilities of UAV swarms requires addressing challenges related to trajectory design, particularly the Multiple Traveling Salesman Problem (MTSP). It involves optimizing the paths of multiple UAVs while avoiding collisions, minimizing overlap and interference, and managing the overall size of the swarm. These challenges highlight the complexities involved in developing high-performance, organized UAV swarm operations. We propose a novel approach based on repulsion force in UAV swarm trajectory design to tackle these issues. Our method utilizes a Genetic Algorithm (GA) to generate a dynamic Repulsion Force (RF) that optimizes the distance between UAVs and the size of the swarm. This approach reduces interference and overlap while effectively navigating the limitations posed by the MTSP. Our proposed solution aims to design efficient trajectories that enhance the overall performance of UAV swarms. We compared our proposed method to existing algorithms, including the MTSPGA, Particle Swarm Optimization (PSO), 2-OPT, Ant Colony (AC) Optimization, and Simulated Annealing (SA), using simulations and evaluations. The results indicate that our proposed method effectively optimizes travel distances and times, reduces interference levels and overlapping, prevents collisions between UAVs, and enhances the size of the UAV swarm. Overall, our method outperforms current approaches, demonstrating its effectiveness for UAV-based applications.