Abstract
Continuous monitoring of human cognitive and physiological state is a key challenge in digital health, particularly in safety-critical environments such as driving. This paper presents a multimodal driver-state monitoring framework integrating electroencephalography (EEG), wearable physiological sensing, and posture-based vision analysis. The proposed system enables synchronised, window-based analysis of cognitive workload and fatigue-related indicators across varying driving scenarios. We report preliminary results from a simulator-based pilot study, showing that the sensing modalities capture complementary aspects of driver state: EEG is sensitive to rapid cognitive changes, wearables reflect slower physiological trends related to fatigue and stress, and posture features capture behavioural adaptation. The study demonstrates the feasibility of multimodal integration while highlighting practical challenges, including sensor intrusiveness, movement-related noise, and inter-individual variability. Rather than proposing deployable algorithms, this work provides methodological insights to inform future real-vehicle implementations and illustrates how human-state awareness can be integrated into software-defined vehicle architectures to support health-aware, human-in-the-loop adaptation in future intelligent vehicles.

This publication has 17 references indexed in Scilit: