Diseño de sistema inteligente para la detección de conductas de riesgo al interior de un vehículo
Cargando...
Fecha
2023
Autores
Díaz Sippli, Liliana Andrea
Martínez Redondo, José Alberto
Muñoz Mass, Brayan Elías
Maestre Mejía, Kevin Rafael
Sanguino Esmeral, Matías
Título de la revista
ISSN de la revista
Título del volumen
Editor
Ediciones Universidad Simón Bolívar
Facultad de Ingenierías
Facultad de Ingenierías
Resumen
El presente artículo de investigación tiene como objetivo principal contribuir al proceso de neutralización y reducción de accidentes de tráfico en Colombia. Estos accidentes representan la segunda causa de muerte más común en el país, únicamente superada por los homicidios. Esta realidad subraya la apremiante necesidad de implementar intervenciones efectivas destinadas a mejorar la seguridad vial. En este contexto, se introduce un sistema diseñado para detectar conductas de riesgo en vehículos utilizando modelos de aprendizaje profundo. El sistema se compone de elementos clave que incluyen un banco de videos, una interfaz de usuario y un evaluador inteligente de situaciones riesgosas en el interior del vehículo. La metodología del sistema implica el entrenamiento con un banco de videos que ilustra conductas riesgosas, tales como la somnolencia, la manipulación de objetos como celulares o botellas, así como la desviación prolongada de la vista de la carretera. Cuando el evaluador inteligente identifica cualquiera de estas conductas mediante el uso de algoritmos de aprendizaje profundo, el sistema emite alertas visuales y sonoras al usuario. Además, facilita el monitoreo y la comunicación con el usuario a través de una interfaz amigable. En conclusión, el sistema propuesto proporciona una alternativa viable para alertar y prevenir conductas riesgosas en la conducción, contribuyendo así a la disminución de la siniestralidad vial en Colombia.
The road accident rate in Colombia is increased by a series of factors, ranging from speeding to disobedience of traffic signals, failure to keep a safe distance and drunkenness of drivers, among other irresponsible behaviors of road actors. These behaviors not only endanger the lives of those who engage in them, but also threaten the safety of other road users. The problem of traffic accidents can be analyzed from multiple perspectives, including climatic, technological, human, and vehicle-related (inhuman) factors. However, this research focuses on human factors, as they represent a significant component in the occurrence of traffic accidents and, therefore, offer a key opportunity for the implementation of effective solutions. Artificial intelligence (AI) presents itself as a powerful and promising tool to address this issue. By leveraging its capabilities, advanced systems for detecting and predicting dangerous behavior within vehicles can be developed. These systems could contribute significantly to the reduction of road accidents by identifying highrisk behaviors, such as cell phone use while driving, driver fatigue or distraction. The relevance of this research lies in its ability to save lives and reduce the suffering caused by traffic accidents in Colombia. Furthermore, by proposing AI-based solutions, it also contributes to the advancement of technology and innovation in the field of road safety. With the implementation of this AI-based tool, a new path towards a safer and more sustainable driving environment in Colombia can be opened, serving as an example for other countries facing similar road safety challenges.
The road accident rate in Colombia is increased by a series of factors, ranging from speeding to disobedience of traffic signals, failure to keep a safe distance and drunkenness of drivers, among other irresponsible behaviors of road actors. These behaviors not only endanger the lives of those who engage in them, but also threaten the safety of other road users. The problem of traffic accidents can be analyzed from multiple perspectives, including climatic, technological, human, and vehicle-related (inhuman) factors. However, this research focuses on human factors, as they represent a significant component in the occurrence of traffic accidents and, therefore, offer a key opportunity for the implementation of effective solutions. Artificial intelligence (AI) presents itself as a powerful and promising tool to address this issue. By leveraging its capabilities, advanced systems for detecting and predicting dangerous behavior within vehicles can be developed. These systems could contribute significantly to the reduction of road accidents by identifying highrisk behaviors, such as cell phone use while driving, driver fatigue or distraction. The relevance of this research lies in its ability to save lives and reduce the suffering caused by traffic accidents in Colombia. Furthermore, by proposing AI-based solutions, it also contributes to the advancement of technology and innovation in the field of road safety. With the implementation of this AI-based tool, a new path towards a safer and more sustainable driving environment in Colombia can be opened, serving as an example for other countries facing similar road safety challenges.
Descripción
Palabras clave
Accidentes de tráfico, Actores en la vía, Seguridad vial, Tasa de mortalidad vial, Traffic accidents, Road actors, Road safety, Road fatality rate