Cognitive sensor systems for non-destructive material analysis

Our comprehensive knowledge in the field of non-destructive testing is based on decades of research in this area carried out at the Fraunhofer EZRT und IZFP. From this strong position, we support the change from manual process control and monitoring to automated process monitoring. For this purpose we are developing cognitive sensor systems. In the future, these systems will evaluate extensive, multi-dimensional and highly complex data streams and on the basis of these evaluations, they will also be in a position to configure and adapt themselves independently and intelligently.

In the future our cognitive sensor systems will...

  • independently implement the measuring methods ideally suited to the particular question,
  • with the aid of "machine learning" collect only those data which are relevant to the particular question (relevant data),
  • continue learning autonomously and adapt themselves independently to new processes (from supervised learning to machine learning),
  • parameterize themselves independently,
  • monitor their own functional capability,
  • intelligently interconnect with other sensor networks (sensor fusion).
© Fraunhofer IIS
Robot-based CT systems are able to inspect large components quickly and accurately.

Non-destructive monitoring with cognitive sensor systems

Our intelligent sensor systems enable us to generate exactly the information and knowledge required to optimize processes along the entire product lifecycle. Through intelligent linking of this current information and data from earlier stages of the product lifecycle, characteristics of materials and products can be improved continuously and even adapted to suit the individual needs of users in the future.

© Fraunhofer IIS