Your Partner for Tailored Solutions

Mission and Vision

We envisage a world full of small smart devices that improve our everyday life and the environment: better hearing aids and prostheses, ubiquitous cheap and distributed devices for predictive maintenance, observe plants and optimize their individual supply of water and fertilizer, support and guard the self-determined lives of elderly and handicapped persons in their own homes, …

For these applications, we want to foster R&D for energy-efficient and low-cost “smartness”, and we are convinced that Deep Learning and Neuromorphic Computing will be the enablers. For the individual use cases, different types of Artificial Intelligence are best suited – we consider both Deep Neural Networks (DNNs – sometimes also termed Artificial Neural Networks – ANNs) and the more bio-inspired Spiking Neural Networks (SNN), and we develop AI acceleration chips and the associated tool eco-system for both.

Fraunhofer Institute for Integrated Circuits IIS

After 40 years of chip design, we know how to tweak the circuits for maximum energy efficiency. As the Home of MP3, we are moreover very strong in algorithms – and today this means also Machine Learning. The tools like hardware-aware training, compiler and virtual prototypes are needed to run the algorithms on the chips – and we have built up these tools in many projects over the last years. As the German institution for applied research, we know how the industry works and what they expect from us. And we take our mission very serious to serve industrial customers.

Friedrich-Alexander-Universität Erlangen-Nürnberg

The following FAU chairs from the departments of computer science and electrical engineering are members of the lab:

We conduct research on design automation of and tool flows for AI accelerators (ASIC-, ASIP- or FPGA-based as well as microcontrollers and programmable SoCs) for embedded systems in automotive, medical, and industrial sectors. Our design flows aim at optimizing energy, performance, and resources through techniques like model compression and hardware-aware neural architecture search.

We apply AI in areas such as radar-based gesture recognition, power amplifier linearization, and sensor signal processing. Neural networks are used to enhance the linearity of A/D and D/A converters.

We work on algorithms to facilitate the efficient calculation of AI tasks by using mathematical transformations of the operations, such as decomposing a vector-matrix-multiplication into several series of ultra-sparse matrix factors.

We focus on novel architecture design and integration of state-of-the-art technologies like Resistive RAMs and High-Bandwidth-Memory. We optimize and adapt systems for applications ranging from server-based batch processing down to embedded applications.

Technische Hochschule
Nürnberg / Nuremberg Institute of Technology

The Technische Hochschule Nürnberg Georg Simon Ohm (the Ohm) has grown from one of the first polytechnics in the German-speaking world. Its members have made significant contributions to the industrialization of Europe. Ohm’s Law and the first German railway are as much a part of this as the first German patent. The Ohm is one of the largest universities of applied sciences of its kind in Germany and particularly distinguished by the wide range of applied research and development (R&D) projects. Artificial Intelligence expertise in this topic is bundled at the university, particularly in the interdisciplinary Center for Artificial Intelligence (Zentrum für Künstliche Intelligenz, KIZ). Here the Sensorimotor Processing, Intelligence, and Control in Edge Computing Systems (SPICES) Laboratory leads the research in cognitive neurocomputing and neuromorphic systems. The SPICES lab focuses on the end-to-end design, development, and deployment of closed-loop cognitive neurocomputing systems. Algorithmically, we specialize in neural network learning algorithms for multi-sensory fusion, cognitive computing, and intelligent control.

Leistungszentrum Elektroniksysteme

The Leistungszentrum Elektroniksysteme (LZE) is a joint initiative of the Fraunhofer-Gesellschaft, Fraunhofer IIS, Fraunhofer IISB, and Friedrich-Alexander University Erlangen-Nürnberg (FAU), in addition to other non-university research institutes and various industry partners. The center grew out of many years of intense cooperation between the Fraunhofer institutes in Erlangen and FAU, in addition to the unique concentration of electronic systems research activities and industries based in Erlangen – Nuremberg Metropolitan Region.

We’re waiting to help you

Get in touch with us today and let’s start making our world smart.

Contact Us