Fraunhofer-Institut für Optronik, Systemtechnik und Bildauswertung IOSB

ML4CPS – Machine Learning for Cyber Physical Systems and Industry 4.0

Cyber Physical Systems are characterized by their ability to adapt and to learn: They analyze their environment and, based on observations, they learn patterns, correlations and predictive models. Typical applications are condition monitoring, predictive maintenance, image processing and diagnosis.

Nowadays machines can learn und develop self maintaining procedures, meaning that those systems can provide a broad range of substantial improvements if introduced in production. Typical applications are condition monitoring, predictive maintenance, image processing and diagnosis.

The 5th Conference on Machine Learning for Cyber Physical Systems and Industry 4.0 - ML4CPS - addresses these topics.

 

 We are looking forward to welcome you in Lemgo
the 23. - 24. October 2019!

 

Main topics of the conference are:

Machine Learning Methods – Deep Learning

  • Learning of automata and state-based systems
  • Time series prediction
  • Dimensionality reduction
  • Clustering, classification

Smart Data – Semantics and Meta Data

  • Description of Data for automatic model learning.
  • Usage of technologies like, OPCUA, AML, ontology learning,
    knowledge representation, information extraction

Machine Learning for Security

  • Intrusion Detection
  • Network Data Analysis
  • Log Analysis
  • Malware Detection
  • Cyber Attack Classification
  • Zero-Day Detection
  • Adversarial ML

Ehtics of Machine Learning

  • Legal usage of AI-based cyber physical systems
  • Planning of staff
  • Ethical questions on decisions for employees
  • Safe collaboration off humans and cyber physical systems
  • Legal developments in Germany, Europe and Worldwide.

Machine Learning in Robotics

  • Image Processing
  • Learning of new tasks
  • Collaboration, navigation and machine to robot interaction

Business Models for Machine Learning

  • Maintenance Services
  • Optimization assistance
  • New structures in development, platform services

Machine Learning on the Edge

  • Scalable Deep Learning services
  • Distributed modelling
  • Security through decentralized analysis
  • Decentralized deep learning
  • Machine learning for resource-constrained devices
  • Distributed optimization



 

The conference offers a forum to present new approaches to Machine Learning for Cyber Physical Systems, to discuss experiences and to develop visions. Therefore, the conference addresses researchers and users from different industry sectors such as production technology, automation, automotive and telecommunication.

 

Chairmen

Professor Dr.-Ing. habil. Jürgen Beyerer, Fraunhofer Institut IOSB
Professor Dr. rer. nat. Oliver Niggemann, Fraunhofer IOSB-INA

 

We are looking forward to welcome you in Lemgo

For more information and registration to the conference please contact

christian.kuehnert@iosb.fraunhofer.de

 

Fraunhofer IOSB

Fachmesse Control: Intuitive Fehlerdokumentation dank AR

Karlsruhe, 8.4.2019 - Per Laserpointer Produktionsmängel direkt am Bauteil schnell und einfach annotieren: Möglich macht das QSelect, eine auf Augmented Reality (AR) basierende Entwicklung des Fraunhofer-Instituts für Optronik, Systemtechnik und Bildauswertung IOSB. Die Karlsruher Forscher werden das System, das in unterschiedlichsten Bereichen vom Karosseriebau bis zur Platinenbestückung einsetzbar ist, erstmals auf der Qualitätssicherungsmesse Control vom 7. bis 10. Mai in Stuttgart präsentieren (Halle 8, Stand 8509-9).

mehr lesen

Tote Winkel ade: Virtual Reality macht das Cockpit transparent

Karlsruhe, 18.2.2019 - Parkassistenzsysteme sind inzwischen weit verbreitetet. Trotzdem geschehen immer noch Rangierunfälle bei PKW und LKW, die hauptsächlich auf die toten Winkel des Fahrzeugs zurückzuführen sind. Am Fraunhofer IOSB wird deshalb am »transparenten Cockpit« geforscht. Eine Studie hat gezeigt, dass diese Technologie den Fahrzeugführern tatsächlich helfen kann.

Read more