Car steering prediction: Implementation of two interest areas
Abstract
Steering angle prediction in intelligent vehicles is a topic widely studied in the literature. Nevertheless, user intervention is required to operate the vehicle, limiting the degree of autonomy. This work proposes algorithms based on artificial neural networks to determinate the steering angle in a moving vehicle, with real-time applications in scaled vehicles on a two-lane track. Using a front camera centered on the vehicle, two areas of interest are determined: the original image and a rectified version by changing the perspective. Subsequently, the tuple of images is used to create two labelled datasets with the desired steering angle to train two convolutional neural networks. Once trained, both models are implemented in real-time to guide a scaled vehicle on an obstacle free road, allowing autonomous navigation in the designated lane while observing the behaviour in each case.
Downloads
References
AutoNOMOS. (2017). AutomodelCar. Obtenido de https://github.com/AutoModelCar/AutoModelCarWiki/wiki/Hardware
Bojarski, M., Del Testa, D., Dworakowski, D., Firner, B., Flepp, B., Goyal, P., Zhang, J. (2016). End to end learning for self-driving cars. arXiv:1604.07316.
Bravo, C. (2018). Navegación autónoma de un robot tipo automóvil en pista de carreras con obstáculos. Ciudad de México: Tesis de Maestría en Ciencias e Ingeniería de Cómputo, Instituto Politécnico Nacional.
Broggi, A., Cerri, P., Debattisti, S., Chiara, M., Panciroli, M., y Prioletti, A. (2015). Proud–public road urban driverless-car test. IEEE In Transactions on intelligent transportation systems, volumen 16, 3508-3519.
Brown, B. (2017). The social life of autonomous cars. IEEE In Computer, volumen 50, 92-96.
Calvo, J., Guzmán, M., y Ramos, D. (2018). Machine learning, una pieza clave en la transformación de los modelos de negocio. Management Solutions, 20-22.
Diaz, E. (2020). Generación del mapa métrico de una pista de competencias para la navegación del robot móvil autonomos mini V2. Huajuapan de León: Tesis de maestría en Robótica, Universidad Tecnologica de la Mixteca.
Du, K., Dung, K., y Hai, T. (2020). Vision based steering angle estimation for autonomous vehicles. International Conference on Advanced Technologies for Communications (ATC) (pp. 187-192). IEEE.
Gonzales, R., y Woods, R. (2010). Procesamiento Digital de imágenes, 3ed. Pearson.
Hernández, R., Fernández, C., y Baptista, P. (2014). Metodología de la investigación, 6ta Edición. Mc Graw-Hill.
Martínez, J. (2021). Error cuadrático medio para regresión. Obtenido de https://www.iartificial.net/error-cuadratico-medio-para-regresion/
Moya, R. (2016). jarroba. Obtenido de https://jarroba.com/que-es-el-clustering/
Sandoval, L. (2018). Algoritmos de aprendizaje automático para análisis y predicción de datos. Revista Tecnológica, volumen 11, 36–40.
Stöckle, C., Herrmann, S., Dirndorfer, T., y Utschick, W. (2020). Automated Vehicular Safety Systems: Robust function and sensor design. IEEE Signal Processing: Autonomous driving, 24-32.
Thieng, M., Dong, T., y Ha, M. (2018). Navigating Self-Driving Vehicles Using Convolutional Neural Network. 4th International Conference on Green Technology and Sustainable Development (GTSD) (pp. 607-610). IEEE.
Copyright (c) 2024 Filiberto Eduardo Martínez-Hernández, Jose Anibal Arias-Aguilar, Edgar Macías-García, Oscar David Ramírez-Cárdenas
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
