Machine Learning for the Internet of Things: Cognitive Tools for Cyberphysical Systems (COSY),
funded by BMBF - IKT 2020 (01.2018 - 03.2020)
In this project, practical trials on machine learning and data analytics are developed at the Environmental Campus and RWTH Aachen University. In the experiments, the trials are implemented in practical applications on hardware and software. The laboratories will be integrated into the respective Master's programmes in Computer Science and Information Technology and linked to a new study programme. Provided offline data sets enable the use of the learning units even without access to the hardware. The project benefits from synergies between the participating universities and the cooperation in a research project on machine learning. A total of five of the proposed experiments will be implemented on the Internet of Things (IoT) platform developed by Trier University of Applied Sciences together with the Internet of Things expert group of the Digital Summit. The experiments cover the fields of application environment, Industry 4.0 and mobility. This project is advised by partners from industry in order to implement the tests in a practical manner. In addition, there should also be the possibility of developing further experiments from industrial applications. A further goal of the project is to establish a maker platform for machine learning on which industrial partners can get in contact with the students of both universities and work together on interesting problems.
- M. Dziubany, J. Schneider, A. Schmeink, G. Dartmann, K. Gollmer, and S. Naumann: Prognose von Warmeverbrauchen: Stolpersteine und Losungen, Umweltinformatik zwischen Nachhaltigkeit und Wandel, Lecture Notes in Informatics (LNI), Gesellschaft fur Informatik, 2018, accepted for publication (in German).
- M. Dziubany, R. Machhamer, H. Laux, A. Schmeink, K. Gollmer, G. Burger and G. Dartmann. Machine Learning Based Indoor Localization Using a Representative k-Nearest-Neighbor Classifier on a Low-Cost IoT-Hardware, 26th European Signal Processing Conference (EUSIPCO), accepted for publication.
© ISEK at RWTH Aachen