The trend toward pervasive computing and embedded networked systems leads to an increased complexity and dynamics of technical systems. Many current embedded systems suffer already from this complexity issue, which results in increased maintenance costs and the need of frequent firmware updates after deployment. Future systems are expected to be even more complex, which requires to research new ways to handle complex embedded systems. One approach to solve this problem is to increase the level of self-organization in networked systems. Self-organizing systems (SOS) usually consist of a large number of autonomous components, which interact with each other and also with their environment. These interactions are guided by internal rules presenting the agents’ behaviour, which allows them to work in a completely decentralized way and also makes them highly adaptable to external effects. Although self-organizing systems offer numerous advantages over traditional ones like robustness against a failure of a component and scalability, due to the distributed structure there is no straightforward way to design such a system. This project targets the three key issues for developing self-organizing embedded systems:
- Identification of system properties and requirements necessitating the application of SOS structure.
- Investigation of possible design methods based on automated parameter search using evolutionary computing.
- The problem of verification and trust in such evolved systems.
The work will be complemented with technical contributions to the field of modelling and design paradigms of self-organized networked embedded systems.
This project is funded by the Österreichische Forschungsförderungsgesellschaft (FFG).
