ZuSE-KI-Mobil (ZuKIMo) is a nationally funded research project, currently in its intermediate stage. The goal of the ZuKIMo project is to develop a new System-on-Chip (SoC) platform and corresponding ecosystem to enable efficient Artificial Intelligence (AI) applications with specific requirements. With ZuKIMo, we specifically target applications from the mobility domain, i.e. autonomous vehicles and drones. The initial ecosystem is built by a consortium consisting of seven partners from German academia and industry. We develop the SoC platform and its ecosystem around a novel AI accelerator design. The customizable accelerator is conceived from scratch to fulfill the functional and non-functional requirements derived from the ambitious use cases. A tape-out in 22 nm FDX-technology is planned in 2023. Apart from the System-on-Chip hardware design itself, the ZuKIMo ecosystem has the objective of providing software tooling for easy deployment of new use cases and hardware-CNN co-design. Furthermore, AI accelerators in safety-critical applications like our mobility use cases, necessitate the fulfillment of safety requirements. Therefore, we investigate new design methodologies for fault analysis of Deep Neural Networks (DNNs) and introduce our new redundancy mechanism for AI accelerators.
2023 Design, Automation and Test in Europe Conference and Exhibition, DATE 2023 - Proceedings
year
2023
publisher
Institute of Electrical and Electronics Engineers Inc.
comment
Funding Information: ACKNOWLEDGMENT This work was funded by the German Federal Ministry of Education and Research (BMBF) under grant number 16ME0096 (ZuSE-KI-mobil). The responsibility for the content of this publication lies with the authors. Furthermore, we thank GlobalFoundries, Synopsys, Cadence, Arteris, and ARM for supporting this research.
10.23919/DATE56975.2023.10137257
%0 Generic
%1 kempf2023zusekimobil
%A Kempf, Fabian
%A Hoefer, Julian
%A Harbaum, Tanja
%A Becker, Juergen
%A Fasfous, Nael
%A Frickenstein, Alexander
%A Voegel, Hans Joerg
%A Friedrich, Simon
%A Wittig, Robert
%A Matúš, Emil
%A Fettweis, Gerhard
%A Lueders, Matthias
%A Blume, Holger
%A Benndorf, Jens
%A Grantz, Darius
%A Zeller, Martin
%A Engelke, Dietmar
%A Eickel, Karl Heinz
%B 2023 Design, Automation and Test in Europe Conference and Exhibition, DATE 2023 - Proceedings
%D 2023
%I Institute of Electrical and Electronics Engineers Inc.
%K AI Accelerator Development Fault Functional Methodology Safety Simulation System-on-Chip myown
%R 10.23919/DATE56975.2023.10137257
%T The ZuSE-KI-Mobil AI Accelerator SoC
%U http://www.scopus.com/inward/record.url?scp=85162703402&partnerID=8YFLogxK
%X ZuSE-KI-Mobil (ZuKIMo) is a nationally funded research project, currently in its intermediate stage. The goal of the ZuKIMo project is to develop a new System-on-Chip (SoC) platform and corresponding ecosystem to enable efficient Artificial Intelligence (AI) applications with specific requirements. With ZuKIMo, we specifically target applications from the mobility domain, i.e. autonomous vehicles and drones. The initial ecosystem is built by a consortium consisting of seven partners from German academia and industry. We develop the SoC platform and its ecosystem around a novel AI accelerator design. The customizable accelerator is conceived from scratch to fulfill the functional and non-functional requirements derived from the ambitious use cases. A tape-out in 22 nm FDX-technology is planned in 2023. Apart from the System-on-Chip hardware design itself, the ZuKIMo ecosystem has the objective of providing software tooling for easy deployment of new use cases and hardware-CNN co-design. Furthermore, AI accelerators in safety-critical applications like our mobility use cases, necessitate the fulfillment of safety requirements. Therefore, we investigate new design methodologies for fault analysis of Deep Neural Networks (DNNs) and introduce our new redundancy mechanism for AI accelerators.
@conference{kempf2023zusekimobil,
abstract = {ZuSE-KI-Mobil (ZuKIMo) is a nationally funded research project, currently in its intermediate stage. The goal of the ZuKIMo project is to develop a new System-on-Chip (SoC) platform and corresponding ecosystem to enable efficient Artificial Intelligence (AI) applications with specific requirements. With ZuKIMo, we specifically target applications from the mobility domain, i.e. autonomous vehicles and drones. The initial ecosystem is built by a consortium consisting of seven partners from German academia and industry. We develop the SoC platform and its ecosystem around a novel AI accelerator design. The customizable accelerator is conceived from scratch to fulfill the functional and non-functional requirements derived from the ambitious use cases. A tape-out in 22 nm FDX-technology is planned in 2023. Apart from the System-on-Chip hardware design itself, the ZuKIMo ecosystem has the objective of providing software tooling for easy deployment of new use cases and hardware-CNN co-design. Furthermore, AI accelerators in safety-critical applications like our mobility use cases, necessitate the fulfillment of safety requirements. Therefore, we investigate new design methodologies for fault analysis of Deep Neural Networks (DNNs) and introduce our new redundancy mechanism for AI accelerators.},
added-at = {2024-02-05T16:21:48.000+0100},
author = {Kempf, Fabian and Hoefer, Julian and Harbaum, Tanja and Becker, Juergen and Fasfous, Nael and Frickenstein, Alexander and Voegel, Hans Joerg and Friedrich, Simon and Wittig, Robert and Matúš, Emil and Fettweis, Gerhard and Lueders, Matthias and Blume, Holger and Benndorf, Jens and Grantz, Darius and Zeller, Martin and Engelke, Dietmar and Eickel, Karl Heinz},
biburl = {https://www.bibsonomy.org/bibtex/291dbe0d5383a989118f702e98d36ed11/fabcho},
booktitle = {2023 Design, Automation and Test in Europe Conference and Exhibition, DATE 2023 - Proceedings},
comment = {Funding Information: ACKNOWLEDGMENT This work was funded by the German Federal Ministry of Education and Research (BMBF) under grant number 16ME0096 (ZuSE-KI-mobil). The responsibility for the content of this publication lies with the authors. Furthermore, we thank GlobalFoundries, Synopsys, Cadence, Arteris, and ARM for supporting this research.
10.23919/DATE56975.2023.10137257},
doi = {10.23919/DATE56975.2023.10137257},
interhash = {9dac614df7cf66f222f8a2b7329a31a1},
intrahash = {91dbe0d5383a989118f702e98d36ed11},
keywords = {AI Accelerator Development Fault Functional Methodology Safety Simulation System-on-Chip myown},
publisher = {Institute of Electrical and Electronics Engineers Inc.},
timestamp = {2024-03-05T15:37:51.000+0100},
title = {The ZuSE-KI-Mobil AI Accelerator SoC},
url = {http://www.scopus.com/inward/record.url?scp=85162703402&partnerID=8YFLogxK},
year = 2023
}