Fraunhofer Institute for Integrated Circuits IIS, Division Engineering of Adaptive Systems EASFlex4Apps
The internet of things will increasingly become a driver of digitalization in many domains. Approaches such as Industry 4.0, Smart Mobility, Smart Health or Smart Home are based on the expectation that an increasing number of systems and components will be networked together. Cyber-physical systems (CPS) permit the close integration of digital systems and components with objects and processes in the real world by means of a data infrastructure. For example, CPS data of a measurement network of distributed sensor nodes can often be accessed globally today by means of internet services.
An increasing number of applications rely on CPS and thereby on heterogeneous and distributed infrastructures such as cloud services or telecommunications networks. However, errors arising in such structures are difficult to locate and take time to correct due to the complexity and the enormous volumes of data. To correct this problem and to prepare CPS for the flexible use cases of the internet of things, we need a new generation of smart, distributed and networked systems. Flex4Apps therefore takes up the challenges arising in the development and use of complex, scalable CPS. The project partners seek to develop solutions that utilize data analysis to enable superior monitoring of operations, support direct (online) and indirect (offline) optimization and improve both the reliability and efficiency overall.
Solution Approach
To significantly increase the potential uses of CPS, the project should produce a framework for the monitoring of applications using cloud resources as well as new data analysis methods. The overall condition of a system is determined from the relationships within the collected data. Critical situations or errors can be detected early in this way, and recommended actions can be initiated. In addition, the partners also developed user-friendly control elements that should improve the handling and system monitoring. The interaction of these “analysis and control tools” was tested in example use cases drawn from practical efforts of the project partners.
The role of the EAS Division in the project is focused primarily on issues surrounding the design of reliable and robust systems, in particular with the tools of model-based analysis of critical, resource-limited communication paths in the application context. The researchers are contributing their expertise in the modeling of CPS infrastructures and usage scenarios, such as for forecasting of performance figures or for data aggregation.
