oktoflow/IIP-Ecosphere AI-enabled Industry 4.0/IIoT platform

The oktoflow Industry 4.0/IIoT platform aims at demonstrating new approaches to vendor-independent, interoperable and open platform concepts for easy-to-use AI. In particular, it aims at uniform, distributed deployment of services to heterogeneous (edge) devices, an AI toolkit for the industrial production, consistent configuration support ranging from network to applications as well as the use and application of established and upcoming IIoT standards, e.g., asset administration shells (to be used for all platform interfaces).

For installation and first steps, see install guide.

News

Exiting: oktoflow shall become the technological core of the funded DATI Innovation Community ReGaP. Stay tuned!!!
ETFA'24: Paper on generating IDTA AAS submodel APIs accepted (preprint on arxiv).
MODELS'24: Tutorial on AAS accepted.
INFLUX DB connector for InfluxDb2
Platform core migrated to JDK 17 (see our migration story). Container creation still needs upgrade #165. Platform runs on JDK 21 with EASy-Producer/development based on Eclipse 2024-06.
BaSyX 1.5.1 support (currently as alternative retro-fit for BaSyx 1.3.1 used as platform AAS version)
MODBUS/TCP connector by Christian Nikolajew (tested for compliance with Phoenix Contact EEM-MA 370 and Siemens SENTRON PAC 3200), example re-usable MODBUS/TCP definitions for Phoenix Contact EEM-MA 370
Plugins for alternative components with conflicting classpaths. Retrofit for BaSyx 1.0, in particular in AAS connector.
Serial connector tested and working.
Auto-generation of all IDTA specs published in February 2024 (accepted for ETFA'23 conference).
AAS abstraction/integration upgrade: referenceElement, multiLanguageProperties, entity, blob, more semanticIds, IDTA 02004-1-2 Handover Documentation, IDTA 02011-1-0 Hierarchical Structures enabling Bills of Material, IDTA 2023-01-24 Draft Submodel PCF, IDTA 02008-1-1 Time Series Data, IDTA 02003-1-2 Generic Frame for Technical Data for Industrial Equipment in Manufacturing, IDTA 02002-1-0 Submodel for Contact Information, IDTA 02007-1-0 Nameplate for Software in Manufacturing related examples and XMAS AAS
Ongoing revision of the management UI: Initial UI design revision and cleanup, application implementation templates can be created and downloaded from the management UI (December 2023).

Public demonstrations

Public demonstration on EMO 2023 with magnetic identification by MIP technology, drive condition monitoring by Lenze and AI-based visual quality inspection.
Public demonstrations on 2023/06/29 in Hildesheim at the "day of computer science" as well as at Nürnberg Digital "The future of intelligent Manufacturing" on 2023/07/07.
Presentation of a federated learning case with two cobots on HM'23 in collaboration with Phoenix Contact, Bitmotec/OnLogic, UR including platform-integrated federated learning based on Flower.
Presentation of improved robot-based platform demonstrator at TddT'22, Berlin
Presentation of robot-based visual quality inspection app on HM'22 (with 3 sources, Python-based AI, AAS-based Angular application).

The demonstrations and further experiments confirmed compatibility with Phoenix Contact PLCnext (OPC UA, EEM integration), Beckhoff IPC (OPC UA), VDW UMATI (OPC UA), Bitmotec (MQTT), Lenze (MQTT), MIP (MQTT), Beckhoff ADS (Windows), Phoenix Contact EEM-MA 370 (Modbus/TCP), Siemens SENTRON PAC 3200 (Modbus/TCP) ...

Overview

Architecture Overview

For detailed links to the individual components, please refer to the platform overview. The platform repository also contains a growing set of examples and demonstration use cases.

For more details on how to develop application-specific services for the IIP-Ecosphere platform, please watch our series of video tutorials on YouTube or IIP-Ecosphere.

Documentation

Platform installation, use and concepts are documented in several documents:

The platform installation guide.
The platform FAQ.
The platform handbook and on Zenodo
The platform releases overview
Service development tutorial YouTube and IIP-Ecosphere.
The IIP-Ecosphere platform requirements (functional and quality view)
The IIP-Ecosphere platform usage view
The IIP-Ecosphere Industry 4.0 platform overview
For more information, see the documentation overview

Publications

Holger Eichelberger, Alexander Weber, Model-driven realization of IDTA submodel specifications: The good, the bad, the incompatible? (preprint on arxiv).
Alexander Weber, Holger Eichelberger, Per Schreiber and Svenja Wienrich, Performance comparison of TwinCat ADS for Python and Java, Symposium on Software Performance (SSP'23), 2023
Ahmad Alamoush and Holger Eichelberger, Analyzing and Improving the Performance of Continuous Container Creation and Deployment, Symposium on Software Performance (SSP'23), 2023
Holger Eichelberger and Claudia Niederée, Asset Administration Shells, Configuration, Code Generation: A power trio for Industry 4.0 Platforms, IEEE 28th International Conference on Emerging Technologies and Factory Automation (ETFA'23), pp. 1-8, doi: 10.1109/ETFA54631.2023.10275339.
Christian Sauer and Holger Eichelberger: Experiences in Collecting Requirements for an AI-enabled Industry 4.0 Platform, GI Fachgruppe Requirements Engineering (to appear in November 2023)
Kevin Feichtinger, Kristof Meixner, Felix Rinker, István Koren, Holger Eichelberger, Tonja Heinemann, Jörg Holtmann, Marco Konersmann, Judith Michael, Eva-Maria Neumann, Jérôme Pfeiffer, Rick Rabiser, Matthias Riebisch and Klaus Schmid (2023): Software in Cyberphysischen Produktionssystemen - Herausforderungen zur Umsetzung in der Industrie In: ATP-Magazin, 2023 (4): 62-68.
Holger Eichelberger, Gregory Palmer, Svenja Reimer, Tat Trong Vu, Hieu Do, Sofiane Laridi, Alexander Weber, Claudia Niederée, Thomas Hildebrandt in Thais Batista, Tomás Bures, Claudia Raibulet and Henry Muccini (Hrsg.) (2023): Developing an AI-Enabled IIoT Platform - Lessons Learned from Early Use Case Validation In: Software Architecture. ECSA 2022 Tracks and Workshops Bd. 13928. S. 265-283. Springer International Publishing.
Holger Eichelberger, Gregory Palmer and Claudia Niederee (2023): Developing an AI-enabled Industry 4.0 platform - Performance experiences on deploying AI onto an industrial edge device In: Softwaretechnik-Trends, 43 (1): 35-37.
Christian Sauer and Holger Eichelberger (2023): Performance Evaluation of BaSyx based Asset Administration Shells for Industry 4.0 Applications In: Softwaretechnik-Trends, 43 (1): 47-49.
Ahmad Alamoush and Holger Eichelberger (2023): Adapting Kubernetes to IIoT and Industry 4.0 protocols - An initial performance analysis In: Softwaretechnik-Trends, 43 (1): 41-43.
Holger Eichelberger, Svenja Reimer, Claudia Niederée and Gregory Palmer (2022): Virtuelle IIoT-Plattform für die Digitalisierung der Fertigung In: Zeitschrift für wirtschaftlichen Fabrikbetrieb, 117 (12): 884-887.
Kevin Feichtinger, Kristof Meixner, Felix Rinker, István Koren, Holger Eichelberger, Tonja Heinemann, Jörg Holtmann, Marco Konersmann, Judith Michael, Eva-Maria Neumann, Jérôme Pfeiffer, Rick Rabiser, Matthias Riebisch and Klaus Schmid (2022): Industry Voices on Software Engineering Challenges in Cyber-Physical Production Systems Engineering In: 2022 27th IEEE International Conference on Emerging Technologies and Factory Automation (ETFA) S. 8. 27. International Conference on Emerging Technologies and Factory Automation, Stuttgart (Germany), IEEE.
Holger Eichelberger, Heiko Stichweh and Christian Severin Sauer (2022): Requirements for an AI-enabled Industry 4.0 Platform – Integrating Industrial and Scientific Views In: SOFTENG 2022 The Eighth International Conference on Advances and Trends in Software Engineering S. 7-14. ThinkMind.
Claudia Niederée, Holger Eichelberger, Hans-Dieter Schmees, Alexander Broos and Per Schreiber Per, KI in der Produktion – Quo vadis?, IIP-Ecosphere Whitepaper
Miguel Gomez Casado and Holger Eichelberger (2021): Industry 4.0 Resource Monitoring - Experiences With Micrometer and Asset Administration Shells In: CEUR-WS Proceedings of Symposium on Software Performance 2021 (SSP'21) CEUR-WS.org.
Holger Eichelberger, Christian Sauer, Shayan Amir Ahmadian, Michael Schicktanz, Andreas Dewes, Gregory Palmer and Claudia Niederée, IIP-Ecosphere Plattform – Anforderungen (Funktionale und Qualitäts-Sicht), IIP-Ecosphere Whitepaper
Heiko Stichweh, Christian Sauer and Holger Eichelberger, IIP-Ecosphere Platform Requirements (Usage View), IIP-Ecosphere Whitepaper
Christian Sauer, Holger Eichelberger, Amir Shayan Ahmadian, Andreas Dewes, Jan Jürjens, Current Industrie 4.0 Platforms – An Overview, IIP-Ecosphere Whitepaper

Building the Platform

The platform consists of some core and many alternative components which can be built locally via the provided Maven multi-module POM. However, to save time, we recommend relying on the pre-built snapshot and release binaries in Maven Central from our Continuous Integration server. To speed up this process when changes to the individual modules occur, the build sequence is incremental and not based on the Maven multi-module build. So we recommend that you install Java, Maven and an IDE (supported/working versions see documentation overview) and let Maven resolve the actual binaries for you. For more detailed information on the build dependencies, please refer to the platform handbook.

We've heard about strange problems on Mac OS. So far, we do not have enough information to resolve the potential problems.

To ease the first steps, we provided Demonstration and Development Docker Containers on Docker Hub.

Please note that several python models exceed the permitted file size of github, thus, GIT Large File Support is required.