Implementación de un sistema de determinación vehicular mediante inteligencia artificial para la evaluación del potencial de generación de energía eólica en el Campus María Auxiliadora de la Universidad Politécnica Salesiana

Abstract

This proposal presents the implementation of an artificial intelligence system for counting and classifying vehicles in front of the María Auxiliadora campus of the Salesian Polytechnic University located on the coast road. This system was used to detect and classify cars, motorcycles, trucks, trailers and buses in real time, improving accuracy compared to other traditional methods. The developed system has a built-in ESP32-CAM camera and an ESP32 microprocessor, which captures and transfers the information to the database in the cloud for further analysis. A margin of error of less than 2% was achieved compared to manual counting. Data was collected during five days and at different times, resulting in higher congestion during the hours of 7:00 - 9:00 a.m. and 16:00 - 18:00 p.m., mainly on weekdays as opposed to weekends where there is a reduction in traffic, especially in cargo transportation. The data analysis showed that motorcycles maintain a constant presence during the week, while trucks and buses show peaks in the mornings and afternoons, coinciding with work and school schedules. The AI-based system offers an efficient and accurate solution for traffic monitoring, providing key information for urban planning and future wind potential analysis. The results can contribute to the optimization of vehicular traffic and road infrastructure, as well as to the implementation of strategies to harness the kinetic energy of wind generated by vehicle traffic.