German Research Center for Artificial Intelligence

Articles by the Partners

The Intelligent Networks research group at DFKI is working on innovative concepts for the implementation of particularly sophisticated communication solutions, especially for the industrial environment, e.g. for manufacturing automation in cyber-physical production systems. For this purpose, modern communication networks must have intelligent functionalities (e.g. self-organization) and adequate interfaces to production-related systems (e.g. MES) in order to provide autonomous, secure and reliable, wireless or wired networking of individual devices and components. Various projects are being pursued in the context of 5G:

TACNET 4.0 – Tactile Internet

14 German companies and organizations have joined forces in the project “TACNET 4.0 – Tactile Internet” to develop a uniformly system for real-time industrial communication. Using the 5G mobile communications technology as a basis, the project, which is funded by the German Federal Ministry of Education and Research (BMBF), aims to further develop the so-called “tactile Internet” for the digitalization of production and robotics. This creates a ecosystem for a variety of industrial applications. The project is coordinated by the German Research Center for Artificial Intelligence (DFKI), in cooperation with Nokia Bell Labs. The main goal of TACNET 4.0 is the development of a unified industrial 5G communication system that consistently integrates 5G networks and industrial communication networks. For this purpose, 5G concepts are extended with innovative industry-specific approaches, cross-network adaptation mechanisms are developed, and open interfaces between industrial and mobile communication systems are designed and developed.

Visit the project website for more information on TACNET.

PUBLICATIONS ON THE PROJECT

• Cloud Control AGV over Rayleigh Fading Channel – The Faster The Better
• Recurrent Neural Network-based Frequency-Domain Channel Prediction for Wideband Communications
• A Comparison of Wireless Channel Predictors: Artificial Intelligence versus Kalman Filter

FIND – „Future Industrial Network Architecture“

The goal of the FIND project is to develop an overall architecture for the future industrial Internet and a network control system that can automatically map and monitor the requirements of industrial applications to the various resources and networking options. Manual input during the configuration of a network as well as during an ongoing operation should be reduced to a minimum. Even complex, distributed control architectures can be implemented very flexibly and efficiently, whereby the underlying complexity is largely abstracted by the application developer and factory operator. Defaults of individual resources, such as a control hardware, can be compensated automatically to reduce unnecessary downtimes of plants or machines. In addition to future systems and those currently under development, such as time-sensitive networking (TSN) or 5G, it will also be possible to easily integrate the numerous existing industrial communication technologies and devices in order to enable a seamless migration. FIND is funded by the Federal Ministry of Education and Research (BMBF) with approximately € 4.3 million  over a period of three years. The consortium is coordinated by Professor Hans D. Schotten, Scientific Director at DFKI and head of the research area Intelligent Networks:

“By combining and integrating existing network technologies, we want to develop a novel and future-proof network control system that is flexible, secure and efficient to use.” – Prof. Hans D. Schotten

Here you will find more information concering FIND project.