You are here : Home > INDUSTRIAL INNOVATION > DEMOS > REAL TIME STRESS MONITORING FOR WEARABLE-BASED STRESS AWARE SYSTEMS

Articles ＆ files | Focus | Article


Stress Observer

Published on 19 April 2023



​Real time stress monitoring for wearable-based stress aware systems

​  What is Stress Observer?

Stress Observer is a data fusion process, based on the user's motion and physiological signals. It analyzes the data to monitor stress levels. Users include both passengers and transportation professionals (racing drivers, truck drivers etc).


The device integrates data fusion processing that:

  • Automatically estimates each person's stress levels regardless of the activity 
  • Uses sensors typically integrated into wearables 

Automatic identification of stress levels allows:

  • Specific action when high stress situations
         are detected
  • Personalized coaching based on stress level and corresponding to specific activities

​   Applications

Non-invasive, wearable-based stress monitoring measures
the emotional state of a person. Designed for transport
and mobility, it offers:

  • Real-time journey planning specific to each traveller
  • Smart emotion tracking to improve transport and mobility comfort and safety
  • Awareness of mobility wellness for specific social groups
  • Human-centric services and applications
  • Professional driver monitoring and biofeedback during training and practice

​   What's new?

  • Well controlled multi-task multi-user experiment for database construction
  • Validation from real life experiments
  • Smartphone application estimating stress in real time

​   What's next?

  • The "Bon Voyage" cooperation project, funded by the EU Horizon 2020 research and innovation program (Grant 635867), has successfully developed traveler stress level monitoring. During the "HADRIAN" cooperation project, funded by EU Horizon 2020, CEA/LETI will develop an observer of driving ability, for better use of automated vehicles.
  • Such data fusion methodology will be used to develop similar tools forassessing driver vigilance, individual panic detection and better biofeedback for transport users.

   How does it work ?

  • Feature extraction dedicated to wearable sensors
  • Machine learning based on a multi-task multi-user experiment
  • Baseline personalization

​

Contact

PUBLICATIONS:

[1] O. Sakri et al., "A Multi-
User Multi-Task Model For
Stress Monitoring From
Wearable Sensors", in 2018
21st International Conference
on Information Fusion
(FUSION), 2018, p. 761‑766,
[2] G. Vila et al., "Real-Time
Monitoring of Passenger’s
Psychological Stress", Future
Internet, vol. 11, n°5, p. 102,
may 2019.



FEATURES: 
  • Key feature extraction by motion and physiological sensor signal processing
  • Machinelearning
  • Classification, estimation



​

Download it

(862 Ko)