Dynamic "Systems of CPS" are the backbone of many technology-driven visions of the future, such as for the automated transport of people and goods, smart cities, forward-looking medical technology and intelligent factories. However, with the ability to collaborate, the complexity in the development and operation of the systems increases by leaps and bounds. This is not only due to the growing number of networked, embedded systems. It must also be considered that the networks change dynamically at runtime. In an intelligent factory, for example, robots may join for certain tasks and form new alliances with other robots for subsequent tasks.
One key to mastering complexity and bringing reliable systems to market is model-based development. This means explicitly modeling structures, behavior, interaction patterns, and functional and non-functional constraints. In order to provide companies with the appropriate methods, CrESt has developed and tested appropriate modeling approaches. They are based on the established SPES methodology, which is already used in some companies.
The approaches described in the book can help organizations to effectively and efficiently develop collaborative systems that operate in dynamic and uncertain contexts. Emphasis is placed on the reliability and variability of the individual systems, which form networks at runtime. The application will be demonstrated using use cases from the automotive, industrial production (Industry 4. 0), and robotics sectors.
Open Access and available as print copy
Besides Klaus Pohl, head of the paluno working group Software Systems Engineering, the editors of the book are: Wolfgang Böhm (TU Munich), Manfred Broy (TU Munich), Cornel Klein (Siemens), Bernhard Rumpe (RWTH Aachen) and Sebastian Schröck (Bosch). The book is freely available online at:
You can also order a bound version: ISBN 978-3-030-62135-3
Engineering of Collaborative Embedded Systems
In the research project CrESt, diverse contributions were made to the model-based collaborative development of cyber-physical systems. The different chapters of this edited volume reflect the multitude and diversity of the contributions. It is an important achievement that we succeeded together with our partners in creating a uniform methodological basic understanding and a generally valid modeling approach, which unites all contributions and forms a good starting point for future work. (Dr. Marian Daun)
Marian Daun was involved in the project in the cross-cutting work to create a consistent methodology. The results of this work can be found in the chapter "Engineering of Collaborative Embedded Systems" at: https://www.springerprofessional.de/engineering-of-collaborative-embedded-systems/18682402.
Function Modelling for Collaborative Embedded Systems
An essential aspect of collaborative systems is the interaction of functions.
Individual systems join to form system networks in order to realize functionality that the individual system cannot offer. For achieving this, it is important that the developers of these systems have a common understanding of the functional characteristics of the systems involved. In close collaboration with our partners from the Technical University of Munich and the Helmut Schmidt University of Hamburg, we have succeeded in creating an ontological meta-model that precisely meets these needs. (Lisa Krajinski)
The results of this work are summarized in the article “Function Modeling for Collaborative Embedded Systems” and available at: https://www.springerprofessional.de/function-modeling-for-collaborative-embedded-systems/18682406.
Architectures for Dynamically Coupled Systems
The chapter "Architectures for Dynamically Coupled Systems" presents methods that support the development of collaborative embedded systems that do not have a rigid composition but whose composition can change at runtime.
This allows, for example, smart factories to dynamically adapt to customer requirements in order to efficiently create a variety of different products. We describe how such systems can be reliably developed in a model-based manner. (Jennifer Brings)
The chapter is available at: https://www.springerprofessional.de/architectures-for-dynamically-coupled-systems/18682410
Goal-Based Strategy Exploration
The chapter "Goal-Based Strategy Exploration" deals with goal-based strategy development.
The chapter presents an approach to determine, based on defined goals, the best adaptation strategy for collaborative system networks competing with the goals of individual embedded cyber-physical systems. This approach can be used to develop adaptation strategies for system networks. These strategies take into account that the goals of the individual systems and the system network closely interact and evaluate these interactions systematically. (Patricia Aluko Obe)
The chapter illustrates the approach using the example of a highly dynamic autonomous robot fleet and is available at: https://www.springerprofessional.de/goal-based-strategy-exploration/18682416
Handling Uncertainty in Collaborative Embedded Systems Engineering
The chapter "Handling Uncertainty in Collaborative Embedded Systems Engineering" addresses the significance of uncertainties that may arise in dynamic system networks, for example, when collaborating CPS exchange information or context is not detected accurately.
Such uncertainties need to be identified, modeled and analyzed at an early stage of system development in order to systematically equip CPS with the appropriate capabilities to ensure smooth operation even in dynamic and unsafe environments. (Torsten Bandyszak)
The chapter describes modeling and analysis techniques for the systematic consideration of uncertainty in CPS development processes. Direct link to the chapter: https://link.springer.com/chapter/10.1007/978-3-030-62136-0_7
About the project CrEst (Collaborative Embedded Systems)
CrESt was funded by the Federal Ministry of Education and Research over 3 years with approx. 15 million Euros in funding. 800,000 Euros went to the UDE. Prof. Pohl's working group was jointly responsible for the overall management and involved in numerous sub-projects. Project partners included Bosch, Siemens, FEV, Bertrandt, Fraunhofer and, in addition to the UDE, other German universities leading in this field. https://sse.uni-due.de/forschung/projekte/crest