RT info:eu-repo/semantics/masterThesis
T1 Diseño e implementación de un sistema deep learning para la detección de obstáculos y lectura de semáforos en tranvías usando Jetson Nano
A1 Díaz Acosta, Jorge Luis
A2 Máster Universitario en Ingeniería Industrial
AB This master’s degree final project presents the obstacle detection and traffic light reading systemusing deep learning Artificial Intelligence that has been designed for trams in the tram networkof Tenerife, Canary Islands, Spain.For its implementation, a state-of-the-art study of the current Artificial Intelligence objectdetection systems has been carried out, after which a YOLO model of object detection by transferlearning has been selected and trained, to subsequently implement it in a program that will runthe model in a Jeson Nano device and generate alerts for the tram driver depending on what hasbeen detected in front of the tram.AI-based systems are in vogue today, largely thanks to recent advances in modelling and thedemocratisation of user access to tools for developing AI systems.This project has demonstrated the feasibility of implementing an Artificial Intelligence systemthat detects obstacles and reads the traffic lights on the Tenerife Tramway network, with a viewto implementing an Artificial Intelligence driving assistance system for trams.
YR 2024
FD 2024
LK http://riull.ull.es/xmlui/handle/915/39356
UL http://riull.ull.es/xmlui/handle/915/39356
LA es
DS Repositorio institucional de la Universidad de La Laguna
RD 08-ene-2025