Detección y seguimiento de objetos mediante aprendizaje profundo para conducción autónoma
Loading...
Date
Authors
Flores-Gómez, Fernando
Journal Title
Journal ISSN
Volume Title
Publisher
Instituto Tecnológico de Costa Rica
Abstract
El laboratorio del ICAI de la Universidad de Málaga, se dedica a realizar soluciones con implementación de Inteligencia Artificial, para distintos sectores. El siguiente documento se detalla la metodología de diseño, aplicando métodos ingenieriles, el cual permita crear un sistema de frenado asistido, donde su accionamiento dependerá de la detección y seguimiento de objetos por medio de imágenes procesadas por medio de un modelo de aprendizaje profundo. Este modelo generara una salida la cual permitirá que un sistema de control analice los datos y a partir de estos actúa acorde a la situación. Además, se realizan pruebas de funcionamiento en distintos ambientes, los cuales permitan verificar el funcionamiento del sistema en ambientes no controlados. Validando la implementación del sistema a un posible mecanismo de asistencia de frenado en un automóvil autónomo.
The ICAI laboratory at the University of Malaga is dedicated to developing solutions with the implementation of Artificial Intelligence for various sectors. The following document details the design methodology, applying engineering methods, to create an assisted braking system where activation depends on the detection and tracking of objects through images processed by a deep learning model. This model will generate an output that allows a control system to analyze the data and act accordingly to the situation. In addition, operational tests are conducted in different environments to verify the system’s performance in uncontrolled settings, validating the implementation of the system as a potential braking assistance mechanism in an autonomous vehicle.
The ICAI laboratory at the University of Malaga is dedicated to developing solutions with the implementation of Artificial Intelligence for various sectors. The following document details the design methodology, applying engineering methods, to create an assisted braking system where activation depends on the detection and tracking of objects through images processed by a deep learning model. This model will generate an output that allows a control system to analyze the data and act accordingly to the situation. In addition, operational tests are conducted in different environments to verify the system’s performance in uncontrolled settings, validating the implementation of the system as a potential braking assistance mechanism in an autonomous vehicle.
Description
Proyecto de Graduación (Licenciatura en Ingeniería Mecatrónica) Instituto Tecnológico de Costa Rica, Área Académica de Ingeniería Mecatrónica, 2024
Keywords
Detección -- Objetos, Aprendizaje profundo (Aprendizaje automático), Inteligencia artificial, Sistemas -- LiDAR, Automóviles, Sistemas de control, Ambientes no controlados, Sistemas -- Visión artificial, Redes neuronales (Computadores), Detection -- Objects, Deep learning (Machine Learning), Artificial intelligence, Systems -- LiDAR, Control systems, Uncontrolled environments, Systems -- Artificial vision, Neural networks (Computers), Research Subject Categories::TECHNOLOGY
Citation
Collections
Endorsement
Review
Supplemented By
Referenced By
Creative Commons license
Except where otherwised noted, this item's license is described as acceso abierto
