From perception to action: Building a robust AI system for safe and adaptive autonomous driving

Abstract

This study explores the development of a self-driving car using a combination of deep learning (DL), machine learning (ML), computer vision (CV), and convolutional neural networks (CNN). The proposed system aims to simulate human-like decision making in response to external conditions encountered during autonomous driving. The approach involves real-time on-road testing and a self-training mechanism to enable the car to continuously learn and adapt. Furthermore, the text suggests an investigation into the fundamental principles of artificial intelligence (AI) and their role in the autonomous car's functionality.