Creation of an embedded sensor for collecting respiration in a car

Internship, HumanTech Institute, 2019

  • 1 student to supervise.
  • The student worked full-time for 4 months.

Context of the project

More and more cars now incorporate advanced driver assistance systems. The car will be increasingly capable of driving autonomously and the role of the driver is bound to change in the years to come. Indeed, the driver will soon only need to supervise the car’s environment in order to react if a critical situation arises.

In the AdVitam (Adaptive Driver-Vehicle InTerAction to Make future driving safer) project, led by the HumanTech Institute in collaboration with the University of Fribourg and He-Arc, we are seeking to use the driver’s psychophysiological state to make driving a semi-autonomous vehicle safer. One of the goals of this project is to investigate which physiological signals are the most representative of the driver’s state (stress, fatigue, distraction…) in order to adapt and optimize the human-machine interaction in the vehicle. This will allow the vehicle to warn the driver adequately in case of critical situations. However, drivers will not constantly wear sensors such as respiration belt. If we want to use physiological signals to evaluate driver’s state, we need to use embedded and non-intrusive sensors.

In this context, the objective of the internship is to implement a sensor in the car in order to be able to collect respiration data from the driver in real time, without this being intrusive. The sensor will be tested on a driving simulator.

Tasks done by the student

  • Litterature review of techniques for measuring respiration
  • Computer-aided design of a material parts (belt + sensor)
  • Printing of the part that holds the sensor
  • Electrical wiring : sensors and Arduino
  • Implementation of data reception and storage on the computer in Python
  • User testing with 5 participants
  • Redaction of final report
  • Presentation of results to the institute