Deep Learning Sobriety Monitoring System in Road-driven Car Driving Risk Assessment Pipeline

Abstract

In automotive field, alcohol attentional impairment occurs before reaching a Blood Alcohol Content (BAC index) of 0.08% (0,05% under the Italian legislation), thus generating a significant impact on driving safety if the drinker is a car driver. Specifically, in such driving scenario in which the road surface is significantly dangerous, the car driver sobriety monitoring shows a key role in driving risk assessment. The authors propose a full deep pipeline for the intelligent road-surface classification combined with an intelligent electronic alcohol sensing system to proper assess the physiological status of the driver. More in detail, the authors propose an intelligent sensing system that makes a common air quality sensor selective to alcohol. A downstream Deep Residual Convolutional Neural Network architecture will be able to learn specific embedded alcohol-dynamic features in the collected sensing data coming from a prototype GHT25S air-quality sensor designed by STMicroelectronics. A parallel deep ad-hoc designed architecture identifies and classifies the segmented road-surface in driving scenario. An overall risk assessment system evaluates the sobriety of the driver combined with the corresponding road-driven risk assessment. The collected preliminary results effectiveness of the proposed approach.