Edge AI: Artificual intelligence for industrial applications

The Baden-Württemberg Ministry of Economics, Labor and Tourism is funding the joint project on adaptive AI chips “Microelectronics for AI – Data-Oriented Realization in Industrial Applications (DoRiE)” at Institut für Mikroelektronik Stuttgart (IMS CHIPS), the FZI Research Center for Information Technology and Hahn-Schickard-Gesellschaft für angewandte Forschung e.V. with around two million Euros.
This is a consistent continuation of the successfully completed basic work in the project “Microelectronics for AI – Neuromorphic Hardware (NeMoH)”, which is also funded by the state of Baden-Württemberg, and the start of an initiative to transfer know-how from edge AI into industrial practice.
The three industry-related facilities of the Baden-Württemberg Innovation Alliance will jointly be implementing lean AI systems for a decentralized use on sensors, robots or machines in industrial applications. The three scientific facilities are supported by an innovation advisory board with representatives from industry.
Lean AI solutions (Edge AI), which are based on adaptive, neuromorphic AI chips, simplify the operation of complex plants and systems by detecting and evaluating changes in condition on site. Data does not have to be transmitted first from a very large number of sensors to a central instance and subsequently evaluated. Decentralized intelligence significantly reduces the amount of data that has to be transmitted and can, for example, make a system more robust against errors. Another positive aspect is to reduce potential, since Edge AI, for example, reduces the number of different sensors on a machine, only a few sensors are needed to obtain a sufficiently accurate status image in real time.
In order to improve the transfer of know-how from science and research to industry, the research facilities are forming a low-threshold access to an Edge AI consulting team that is accessable to companies in Baden-Württemberg. Using QuickChecks and Exploring Projects, a viable technical solution path in addition to an initial assessment and evaluation of application ideas for Edge AI in each company can be defined. Companies who are interested can apply for the free QuickChecks at at partnering research facilities.