VEDLIoT: Very Efficient Deep Learning in IoT
M. Kaiser1, R. Griessl1, N. Kucza1, C. Haumann1, L. Tigges1, K. Mika1, J. Hagemeyer1, F. Porrmann1, U. Rückert1, M. vor dem Berge2, S. Krupop2, M. Porrmann3, M. Tassemeier3, P. Trancoso4, F. Qararyah4, S. Zouzoula4, A. Casimiro5, A. Bessani5, J. Cecilio5, S. Andersson6, O. Brunnegard6, O. Eriksson6, R. Weiss7, F. Meierhöfer7, H. Salomonsson8, E. Malekzadeh8, D. ödman8, A. Khurshid10, P. Felber11, M. Pasin11, V. Schiavoni11, J. Ménétrey11, K. Gugala12, P. Zierhoffer12, E. Knauss9 and H. Heyn9
1Bielefeld University, Germany
2christmann informationstechnik + medien GmbH & Co. KG, Germany
3Osnabrück University, Germany
4Chalmers University of Technology, Sweden
5University of Lisbon, Portugal
6VEONEER Inc., Sweden
7Siemens AG, Germany
8EMBEDL AB, Sweden
9Göteborg University, Sweden
10Research Institutes of Sweden AB (RISE)
11University of Neuchâtel, Switzerland
12Antmicro, Poland
ABSTRACT
The VEDLIoT project targets the development of energy-efficient Deep Learning for distributed AIoT applications. A holistic approach is used to optimize algorithms while also dealing with safety and security challenges. The approach is based on a modular and scalable cognitive IoT hardware platform. Using modular microserver technology enables the user to configure the hardware to satisfy a wide range of applications. VEDLIoT offers a complete design flow for Next-Generation IoT devices required for collaboratively solving complex Deep Learning applications across distributed systems. The methods are tested on various use-cases ranging from Smart Home to Automotive and Industrial IoT appliances. VEDLIoT is an H2022 EU project which started in November 2022. It is currently in an intermediate stage with the first results available.
