Ant Colony Optimization for Efficient Distance and Time Optimization in Swarm Drone Formation

Ronny Mardiyanto,Andri Suhartono,Devy Kuswidiastuti,Heri Suryoatmojo

DOI: https://doi.org/10.22219/kinetik.v9i1.1859

2024-02-02

Abstract:One of the challenges in swarm drone formation is achieving fast and effective formation with optimal distances. In this paper, we propose a swarm drone formation approach utilizing Ant Colony Optimization (ACO) for achieving it. We conducted simulations involving the formation of three or more drones, aiming to identify the best formation based on distance, acceleration, and time criteria. Simulation results demonstrate that formation time is significantly reduced when employing ACO optimization compared to non-optimized methods. Additionally, the optimized formations exhibit shorter inter-drone distances compared to non-optimized formations. By implementing this approach, swarm drone formations can be rapidly established with minimized distances, resulting in substantial battery savings. The simulation encompassed various patterns formed by 3, 5, 10, 15, 20, and 25 drones. The findings indicate that the approach can reduce formation time by varying degrees, ranging from 12% to 51%, across 66% of the conducted experiments, notably for patterns created with a substantial drone count. The degree of diversity observed among the proposed solutions reached 60%, with minimal variances of less than 1% for each.