SAFE-D: A Spatiotemporal Detection Framework for Abnormal Driving Among Parkinson's Disease-like Drivers
DOI: 10.48550/arXiv.2510.17517
URL: https://arxiv.org/abs/2510.17517
archive: indexed pipeline: discovered
Abstract
Driver health state is treated as a determinant factor in driving behavioral regulation. The SAFE-D framework analyzes Parkinson's disease motor symptoms and their connection to degraded driving performance, using multi-component vehicle data to build behavioral profiles. An attention-based network processes spatiotemporal features for anomaly detection, achieving 96.8% average accuracy in distinguishing normal from Parkinson-affected driving patterns, validated on Logitech G29 and CARLA simulator platforms.
Access
Route: arxiv_oa
Publisher: arxiv