Industrial Communication Technology, Artificial Intelligence in Automation

KOARCH: Cognitive Architecture for Cyber-Physical Production Systems and Industrie 4.0

01.01.2018 bis 31.12.2022

Global competition combined with increasing product complexity resulted in a massive growth in complexity of production systems in the last years. The largest part of the development content in mechanical engineering concerns software development. This increasing workload is weighing on automation specialists, system engineers and plant manufacturers.

Industrie 4.0, cyber-physical-systems and intelligent automation-systems offer a solution for this increasing burden: The main ideas is shifting human expert knowledge into automation. As opposed to the procedural approach of classical automation, the expert just formulates objectives like a description of the final product, throughout targets or the allowed energy consumption.

The knowledge refers to objectives that are described as statements and not as procedures to achieve the objectives like it was before. This means, the knowledge is described declaratively and not procedurally. This way, intelligently programed systems can use this knowledge to solve adaption and optimisation problems. Therefore, the human effort in automation decreases, e.g. in optimisation tasks, commissioning and plant modifications.

In order to realise such intelligent systems, there is a need of new automation technologies and especially new software services. This includes e.g. machine-learning-methods, condition-monitoring- and diagnosis algorithms as well as optimisation processes.

Currently these new software services get implemented in Industrie 4.0 approaches by each partner independently. The interfaces are proprietary, so that data, models and results can not be exchanged. Besides, a kognitive architecture Is developed in this project, to enable an easy exchange of data and services in Industrie 4.0 environments. Therewith, Industrie 4.0 devices and components from different manufacturers are able collaborate, i.e. they can exchange data, information (e.g. anomalies and optimisation goals) as well as algorithms and solution strategies and process them.

This project is promoted by:
Bundesministerium für Bildung und Forschung (BMBF)
Sponsors: VDI Technologiezentrum GmbH
Funding Code: 13FH007IA6
Funding Lines: IngenieurNachwuchs 2016 – Kooperative Promotion
Stakeholders / Contacts: Dipl.-Math. Natalia Moriz, Christoph Geng, M. Sc.
Employees: Philip Priss, B. Sc., Christoph Geng, M. Sc.
Dr.-Ing. Andreas Bunte, Prof. Dr. rer. nat. Oliver Niggemann, Prof. Dr. rer. nat. Benno Stein
Integrating OWL Ontologies for Smart Services into AutomationML and OPC UA
Dr.-Ing. Andreas Bunte, Prof. Dr. rer. nat. Benno Stein, Prof. Dr. rer. nat. Oliver Niggemann
Model-Based Diagnosis for Cyber-Physical Production Systems Based on Machine Learning and Residual-Based Diagnosis Models
Dr.-Ing. Andreas Bunte, Andreas Fischbach, Jan Strohschein, Thomas Bartz-Beielstein, Heide Faeskorn-Woyke, Prof. Dr. rer. nat. Oliver Niggemann
Evaluation of Cognitive Architectures for Cyber-Physical ProductionSystems
In: arXiv e-prints, Feb 2019
Dr.-Ing. Andreas Bunte, Andreas Fischbach, Jan Strohschein, Thomas Bartz-Beielstein, Heide Faeskorn-Woyke, Prof. Dr. rer. nat. Oliver Niggemann
Evaluation of Cognitive Architectures for Cyber-Physical Production Systems
Andreas Fischbach, Jan Strohschein, Dr.-Ing. Andreas Bunte, Jörg Stork, Heide Faeskorn-Woyke, Dipl.-Math. Natalia Moriz, Thomas Bartz-Beielstein
CAAI -- A Cognitive Architecture to Introduce Artificial Intelligence in Cyber-Physical Production Systems
Dr.-Ing. Andreas Bunte, Henrik Ressler, B. Sc., Dipl.-Math. Natalia Moriz
Automated Detection of Production Cycles in Production Plants using Machine Learning
In: 25th IEEE International Conference on Emerging Technologies and Factory Automation (ETFA), Sep 2020
Dr.-Ing. Andreas Bunte, Frank Richter, Rosanna Diovisalvi
Why it is Hard to Find AI in SMEs - A Survey From the Practice and How to Promote it
In: International Conference on Agents and Artificial Intelligence (ICAART), Feb 2021
Technische Hochschule Köln
Deutsche Telekom AG, Innovations Laboratories (T-Labs)
telexiom AG
Promoted by