Sirine Ammar

Electroencephalography (EEG)-based Brain–Computer Interfaces offer direct insight into a driver’s mental state, relevant to autonomous-vehicle perception stacks. This work addresses vigilance/drowsiness specifically, one of several driver states relevant to Level 3 takeover readiness. Detecting drowsiness before handover in Level 3 vehicles requires balancing high-dimensional EEG features against the real-time demands of a lightweight safe-stop trigger mechanism. We present a complete pipeline: a standardized 22-feature-per-channel schema extracted across four driving-related EEG datasets, an autoencoder achieving over 95% dimensionality reduction with minimal performance loss, and a classifier driving a constant-deceleration kinematic profile used to obtain a measurable latency figure, evaluated under a fully subject-disjoint, leakage-free protocol. Leave-one-subject-out cross-validation on the vigilance dataset yields a mean macro-F1 of 0.793 ± 0.163 (0.763 ± 0.155 on an alternative run, within expected autoencoder-retraining variance). Extending compression evaluation to all four datasets shows preserved or improved performance on three, with a small cost on the fourth. The full decision chain completes in under 1 ms (mean 0.13 ms) on standard hardware; this covers the perception-to-trigger budget only, excluding solver latency that an optimization-based planner (MPC/CBF-QP) would add downstream. These findings show cognitive-state-aware safe-stop triggering is implementation-viable, while highlighting the calibration and cross-machine validation needed for deployment.

Electroencephalography (EEG)-based brain-computer interfaces (BCIs) hold significant potential for enhancing driver safety through real-time monitoring of cognitive and affective states. However, the development of reliable BCI systems for Advanced Driver Assistance Systems (ADAS) depends on the availability of high-quality, publicly accessible EEG datasets collected during driving tasks. Existing datasets lack standardized parameters and contain demographic biases, which undermine their reliability and prevent the development of robust systems. This study presents a multidimensional benchmark analysis of seven publicly available EEG driving datasets. We compare these datasets across multiple dimensions, including task design, modality integration, demographic representation, accessibility, and reported model performance. This benchmark synthesizes existing literature without conducting new experiments. Our analysis reveals critical gaps, including significant age and gender biases, overreliance on simulated environments, insufficient affective monitoring, and restricted data accessibility. These limitations hinder real-world applicability and reduce ADAS performance. To address these gaps and facilitate the development of generalizable BCI systems, this study provides a structured, quantitative benchmark analysis of publicly available driving EEG datasets, suggesting criteria and recommendations for future dataset design and use. Additionally, we emphasize the need for balanced participant distributions, standardized emotional annotation, and open data practices.