Cognitive load state analysis of drivers using multimodal physiological data and machine learning in a VR simulator

Driver inattention is the primary cause of highway accidents. Various distractions can impair driving performance, posing a significant threat to road safety. This study uses a VR driving simulator to examine distracted driving, creating a simulated driving environment. It involves designing main and sub-driving tasks to replicate the cognitive load and distractions experienced by drivers. Data are collected using NASA-TLX scales, eye-tracking technology, and physiological signal acquisition. The impact of different distraction workloads on driver performance was assessed using statistical analysis and support vector machines (SVM). A multi-source data fusion model was developed to discriminate distracted driving behaviors. This research aims to provide theoretical insights and practical guidance for improving safe driving practices.