Year: 2025
Type: conference-paper
Abstract: Ensuring safety in autonomous driving requires robust systems capable of detecting and mitigating dangerous driving behaviors in real time. This paper proposes an AI-powered approach that integrates deep learning models, specifically Convolutional Neural Networks (CNNs), with sensor fusion from vehicle-embedded sensors and onboard cameras. The system processes multimodal data to identify hazardous autonomous driving patterns and applies a Random Forest classifier to distinguish between normal and unsafe behaviors. By leveraging advanced data fusion and machine learning techniques, the proposed method enhances the reliability of autonomous driving safety mechanisms. Experimental results demonstrate the effectiveness of the approach in accurately classifying driving behaviors, offering a promising solution for proactive risk mitigation and accident prevention in autonomous vehicles.
Language: English
Subfield: Automotive Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4413158190