Bio-inspired Embedded System for Intelligent Driving Assistance in the Next Generation Cars

Abstract

In the field of automotive applications, scientific effort has been focused on monitoring driver’s attention level as well as on the driving scenario risk assessment. In that context, the physiological tracking of the driver has proved to be an excellent non-invasive approach to provide a robust driving assistance. The authors propose a driving assistance system based on the use of an ad-hoc designed bio-sensor that samples the driver’s photoplethysmographic (PPG) signal by correlating it with the related attention level. A downstream deep architecture processes the driver’s PPG signal by reconstructing the corresponding attention level. Simultaneously, an external intelligent automotive-grade vision-based system will be responsible for characterizing the driving scenario risk level by using video saliency analysis techniques. The collected experiment results confirmed the effectiveness of the proposed full pipeline.