Dynamic information-centric systems-of-systems
Today’s modern high tech systems offer a previously unparalleled variety of feasible applications, made possible by a trend of increasing interactions and collaborations between systems, and their sharing of information, processes, and resources. Altogether, this led to systems whose components are highly complex systems in themselves, called Systems of Systems, abbreviated by SoS.
For well defined situations and applications, the distribution of tasks and resources might be defined in advance, thus regulating the resulting collaboration as well as the integration of the individual systems during implementation. This approach, however, restricts a SoS to its predefined task and configuration, and consumes a considerable amount of resources and time, especially if it becomes necessary to modify the engineered system. The resulting limits on a system of systems flexibility and its abilities to adapt to new scenarios and configurations, as well as to evolve to cover new applications or technologies are no longer acceptable: Future SoS are expected to react adequately to ad-hoc changes, to enable collaboration across former boundaries, to deal with the unexpected, to offer possibilities.
The Poseidon project rises to the challenge to discover new ways on how to build such advanced systems of systems, and therefore on how to allow for flexibility, adaptability and evolvability in systems of systems while ensuring reliability – a crucial requirement, not only in the domain of maritime safety systems that provides Poseidon’s exemplary application and the industrial laboratory needed for its success.
The challenge of developing advanced system of systems is to gain flexibility, adaptability and evolvability while retaining reliability at the same time. Any change in the SoS configuration should be achievable with adequate efforts and the system should always remain operational and reliable, even given unforeseen events, scenarios, and applications.
Given the information-centric operations of such systems of systems, the Poseidon project will provide evidence for meeting this challenge within the context of the application domain by fulfilling the following primary goals:
- Development of a system of systems architecture beneficial to adaptability and evolvability.
- Development of an integration and acceptance methodology for reliable dynamic (re-)configurations of the system of system.
- Trustworthy fusion and processing of information originating from a variety of sources that differ in type, role, syntax, and semantics.
- Intelligent analysis of the available information to solve the application tasks.
The research will thus have strong multi-disciplinary aspects, requiring expertise in different technical and scientific domains, including object behavior analysis, machine learning, information visualization, system reasoning, information fusion, information relationship modeling, trust management, run-time integration, and acceptance test techniques. To be successful, it will be needed to establish meaningful combinations of the analytic, modeling, and implementation techniques of the various disciplines.
The research activities will be guided by three lines of attention:
Information analysis and visualization, covering
- data pre-processing and feature extraction;
- anomaly detection and situation awareness;
- multi-objective visualization of data.
Trustworthy information interoperability, covering
- ontology engineering and semantic alignment;
- managing and enacting security policies in presence of dynamic coalitions.
Integration and acceptance, covering
- model-based methods, techniques, and tools for testing, monitoring and diagnosis;
- workflow modeling and realization for integration and acceptance processes;
- automatic adapter generation for integration of subsystems;
- process-mining to discover interface processes and check conformance;
- analysis of testability and diagnosability and their influence on the SoS.
Poseidon is a joint project of a consortium of industrial and academic partners. The Embedded Systems Institute (ESI) has the responsibility for project management and knowledge dissemination. Also, ESI Research Fellows coach and supplement the research activities of the academic partners.
Thales Above Water Systems Division, the carrying industrial partner, provides the industrial challenge, expert knowledge in the domain of Maritime Safety and Security Systems, and specialized facilities, e.g. for simulations, experiments, and prototyping. The Above Water Systems Division is part of the Thales Group, a global electronics company delivering mission-critical information systems and services for the Aerospace, Defense, and Security markets. Second participating industrial partner is Noldus Technology, a research driven company that provides professional software and instrumentation for the collection and analysis of behavioral data.
The academic partners are Delft University of Technology, Eindhoven University of Technology, Tilburg University, University of Amsterdam and VU University Amsterdam.
For part of the project time, researchers are co-located at the ESI facilities or at Thales in Hengelo.
Poseidon started in June 2007, and is partly funded by the Dutch Government.
Poseidon flyer (130kB)