Implementation of a Driving Simulator for the Collection of Data on Human Behavior in Vehicular Traffic

Producción científica: Capítulo del libro/informe/acta de congresoContribución de conferenciarevisión exhaustiva


Traffic psychology covers not only the act of driving but also includes inherent phenomena such as accidents, violations, and the generation of emotions. This article proposes the development of a driving simulator within the video game Grand Theft Auto V, taking advantage of the artificial intelligence of characters that interact with the player, as well as other drivers and pedestrians. The implementation includes the adaptation of steering wheel type controls and the creation of a mod in C# for data collection and export. This implementation seeks to obtain useful data for the analysis of the behavior of the drivers evaluated during the driving tests carried out in three virtual circuits within a city. By performing the tests with drivers of different levels of experience, it was possible to verify the effectiveness of the tool in its use and realism. According to the individuals evaluated, the simulator achieved an average realism of 4.6 on a Likert scale from 1 to 5. Likewise, the results obtained show that the data collected by the simulator is less biased than those obtained through a post-driving survey.

Idioma originalInglés
Título de la publicación alojadaModern Management based on Big Data III - Proceedings of MMBD 2022
EditoresAntonio J. Tallon-Ballesteros
EditorialIOS Press BV
Número de páginas8
ISBN (versión digital)9781643683003
ISBN (versión impresa)9781643683003
EstadoPublicada - 10 ago. 2022
Evento3rd Conference on Modern Management Based on Big Data, MMBD 2022 - Seoul, República de Corea
Duración: 15 ago. 202218 ago. 2022

Serie de la publicación

NombreFrontiers in Artificial Intelligence and Applications


Conferencia3rd Conference on Modern Management Based on Big Data, MMBD 2022
País/TerritorioRepública de Corea

Nota bibliográfica

Publisher Copyright:
© 2022 The authors and IOS Press.

Citar esto