Icon KAMeri – Cognition-based workplace safety for human-machine interaction

Practical example and current challenges

  • In the future, machines will work closely with people in more and more sectors in terms of space and function. This can lead to new mental stress states, which must be avoided as a preventive measure. The most common cause of accidents at work is human behavioural errors due to carelessness, stress or hectic. In addition to consequences for the affected persons themselves, this is also an additional cost factor for the company.
  • The physically close cooperation between man and machine and the mental stress states of workers resulting from this cooperation require new, adapted and reliable occupational safety concepts and preventive measures.
  • The aim of the "KAMeri" project is to improve human-machine interaction. The use of new technologies, such as the continuous recording of EEG brain waves and subsequent evaluation in cloud-based solutions, reduces occupational accidents and promotes the physical and mental health of employees. In this way, a decisive contribution is made to improving occupational safety and working conditions.
  • For the implementation and training of the models, extensive amounts of data are required. Further challenges lie in ensuring legal requirements for handling sensitive personal data.
KAMeri – Cognition-based workplace safety for human-machine interaction

What added value does the "GAIA-X project" offer?

  • GAIA-X enables the transmission of recorded EEG data using proven and above all secure protocols via a wireless connection.
  • GAIA-X provides access to innovative methods and algorithms. Machine learning and artificial intelligence allow the automatic evaluation of EEG data, the classification of features and patterns and statements about the cognitive state of the worker.
  • In addition, new types of robot controllers are being developed which are controlled or adapted in their mode of operation via a central cloud instance. The use of Big Data applications enables the data to be evaluated in higher quality.
  • An innovative device authentication concept, which is provided via the cloud infrastructure, ensures the security of sensitive and personal data. The project will develop a secure and GDPR-compliant handling of EEG data. User acceptance will be increased through anonymisation and pseudonymisation.

Use Case Team

  • Dr. Elsa A. Kirchner – German Research Center for Artificial Intelligence (DFKI)
  • Dr. Ralf Hauffe – eemagine Medical Imaging Solutions
  • Dr. Dirk Werth – August-Wilhelm Scheer Institut (AWSi)
  • Marcus Frei – NEXT. robotics