Development of real time data processing algorithms for UAV traffic optimization

• Published in: Sučasnij stan naukovih doslìdženʹ ta tehnologìj v promislovostì (Online) no. 1(31); pp. 49 - 60
• Main Author: Yena, Maksym
• Format: Journal Article
• Language: English
• Published: 31.03.2025
• ISSN: 2522-9818; 2524-2296
• DOI: 10.30837/2522-9818.2025.1.049

Subject matter: UAV traffic management processes, including algorithms for processing large data streams in real time to ensure safety, efficiency, and optimal flight routing. Goal: To development and implementation of real-time data processing algorithms to ensure safe, efficient and automated UAV traffic management in urban and rural environments. Tasks: To analyze existing approaches to UAV traffic management and real-time data processing technologies; to develop a mathematical model that takes into account the specifics of UAV routing, including collision avoidance and route optimization; to create an algorithm for processing input data in real time that integrates dynamic traffic changes, weather conditions, and airspace conditions; to implement and test the proposed algorithm in a simulation environment; to conduct a comparative analysis of UAV simulations with and without the proposed algorithm. Methods: To apply nonlinear optimization methods to construct routes that minimize energy consumption, flight time, and collision risk; to use graph-theoretic models to represent airspace as a network with nodes (route points) and edges (potential trajectories), which allows for effective solution of routing problems; to use genetic algorithms to find optimal solutions in complex multi-factor routing problems; to process data based on Kalman filters; creation of virtual copies of the airspace for conducting experiments and evaluating the effectiveness of algorithms in a safe environment. Results: The developed nonlinear optimization algorithms allowed to minimize the energy consumption for UAV flights and the time of task execution; the effectiveness of the approach was confirmed by testing, which showed a reduction in energy consumption and a decrease in the average flight time compared to the basic algorithms; a graph-theoretical model of the airspace was built, which allows to visualize and analyze possible routes; filtering algorithms showed high accuracy in predicting the position of the UAV even under conditions of instability of GPS signals. The introduction of Kalman filters allowed to reduce the error in determining the position of the UAV, which is critical for avoiding collisions. Conclusions: The developed methods ensure safe airspace management and significantly reduce the risks of UAV collisions, which makes them promising for integration into urban and regional management systems; The use of optimization, clustering and evolutionary algorithms allows to improve routing, reduce energy consumption and task execution time.