ARTIFICIAL NEURAL NETWORKS FOR ENHANCED UAV PERFORMANCE IN URBAN AREA INSPECTIONS

Аннотация

Urban area inspections often present unique challenges for unmanned aerial vehicles (UAVs), requiring adaptable and responsive systems to navigate complex, dynamic environments. This study explores the integration of Artificial Neural Networks (ANN) to enhance the capabilities of small UAVs specifically for urban inspections. By leveraging ANN models, UAVs can improve obstacle avoidance, optimize flight paths, and enhance image processing for real- time data analysis, all of which are critical in densely populated and infrastructure-heavy areas. We conducted a case study to evaluate the performance of ANN-enabled UAVs in typical urban scenarios, assessing improvements in operational efficiency, safety, and accuracy. The findings suggest that incorporating ANN significantly enhances UAV performance, offering a robust solution for detailed urban inspections, monitoring, and data acquisition. This research contributes to the development of autonomous UAV systems capable of addressing the demands of urban environments with high reliability and minimal human intervention. KEYWORDS Artificial Neural Networks (ANN), Unmanned Aerial Vehicles (UAVs), Urban Area Inspection, Autonomous Navigation, Real-Time Data Processing, Obstacle Avoidance, Smart Cities.