Using of convolutional neural networks for the prediction of steering angles in autonomous vehicles

The swift development of artificial intelligence has revolutionized the world of self-driving cars by integrating intricate models and algorithms Autonomous vehicles are regarded as significant developments in computer technology and artificial intelligence. Using sophisticated and powerful algorith...

Full description

Saved in:
Bibliographic Details
Main Authors: ALKafajy, Hussam Jaafar Kadhim, Al-Amri, Amel Hussein Abbas
Format: Conference Proceeding
Language:English
Subjects:
Algorithms
Artificial intelligence
Artificial neural networks
Automobiles
Autonomous cars
Autonomous vehicles
Cloning
Computer vision
Data augmentation
Error analysis
Image processing
Steering
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The swift development of artificial intelligence has revolutionized the world of self-driving cars by integrating intricate models and algorithms Autonomous vehicles are regarded as significant developments in computer technology and artificial intelligence. Using sophisticated and powerful algorithms in these vehicles will effectively address a range of driving obstacles, especially for those with disabilities or advanced age. The objective of our study is to make self-driving automobiles travel from point A to point B on the first track without deviating from it and to move on the second track without having to undergo any training on it. The current study introduces a computer vision method that extracts information from a dataset. Study tasks involve processes including behavioral cloning, data augmentation, image processing, and the improvement of the CNN model. The proposed model attained a mean square error of 0.0250 for the validation value and 0.0281 for the training value, which was deemed an outstanding outcome.
ISSN:0094-243X
1551-7616
DOI:10.1063/5.0239624