Achtung:

Sie haben Javascript deaktiviert!
Sie haben versucht eine Funktion zu nutzen, die nur mit Javascript möglich ist. Um sämtliche Funktionalitäten unserer Internetseite zu nutzen, aktivieren Sie bitte Javascript in Ihrem Browser.

CRC 901 – On-The-Fly Computing (OTF Computing) Show image information

CRC 901 – On-The-Fly Computing (OTF Computing)

|

Seventh SFB 901 Seminar in winter semester 2020/2021

On February 3, 2020, the 7th SFB 901 seminar in the winter semester will take place.

4:00 - 5:00 T1

Speakers: Prof. Dr. Marco Platzner (Paderborn University) and Dr. Alexander Boschmann (Weidmüller Interface)
Title: Flexible Realization of Industrial Analytics Functions on Reconfigurable Systems-on-Chip

Abstract: 

CRC.T1 transfers results from the area of heterogeneous execution environments (CRC.C2) to the application domain industrial analytics. In this presentation we first emphasize the main research challenges of CRC.C2, heterogeneity and dynamics, and show how our approach of multithreaded hardware/software programming for reconfigurable systems-on-chip addresses them. Then, we introduce to the application partner Weidmüller Interface and its products and services, in particular in the domain of industrial analytics. Finally, we give an overview over the goals we want to achieve with this transfer project.  



Title: Embedded Machine Learning on Reconfigurable Systems-on-Chip
Speakers: Lennart Clausing and Dr. Hassan Ghasemzadeh Mohammadi

Abstract: 

The first part of this talk focuses on the architecture and tool flow we develop for programming reconfigurable systems-on-chip. We detail ReconOS^64 that allows for flexible hardware/software realizations and supports transmodal migration through partial hardware reconfiguration. The second part of this talk presents DeepWind, an application case study for predictive maintenance of wind turbines. We elaborate on the problem, the developed sensor data processing chain, and the applied multi-channel convolutional neural network. We discuss the mapping of the inference step for our machine learning model to a reconfigurable system-on-chip and present the results achieved. Finally, we give an outlook over future activities planned in this project.

The talks will be held via BigBlueButton.

The University for the Information Society