Enhancing Driver Monitoring and Decision-Making in Autonomous Vehicles

Abstract

This paper examines the integration of advanced Human-Machine Interfaces (HMI) and driver monitoring systems in level-3 autonomous vehicles, where a car can be selfdriven, but requires intervention in severe conditions. It highlights key research advancements in cognitive workload and driver interaction studies, emphasizing the importance of real-time assessment of driver state using multi-modal data. The study explores innovative decision-making approaches using fuzzy logic and machine learning to manage control transitions in semi-autonomous driving scenarios. The goal is to enhance safety and efficiency in autonomous vehicles by providing a nuanced understanding and prediction of driver behavior, contributing to the development of intelligent, human-centered automotive technology.