Skip to main content

Safety critical embedded systems: from requirements to system architecture


Three major software-related trends dominate most technological industries today: The costs for software development are exploding and already dominate the development costs, the features that determine purchase decisions are implemented in software, and the demand for complex safety-critical systems is rapidly increasing. Academic advances in the specification, verification and code generation of such systems have provided new technologies which can be used to improve dramatically the conventional way of building them. While commercial tools are already available, which successfully apply such techniques to the design of system components or smaller systems, it is generally recognised that the design of complex safety-critical systems requires a multi-formalism approach which combines visual programming as an engineer-friendly means of capturing and validating user requirements and formal methods for guaranteeing high quality specifications as well as provably correct generated code.

The objective of providing an integrated set of formalisms and supporting methods and tools brings together in SACRES industrial developers of safety critical systems, tool vendors and technology providers. The industrial users (British Aerospace, Siemens and SNECMA) will provide design tasks, help develop supporting design methodologies and generally assess the new approach in the context of real applications. The vendors (i-Logix, SNI and TNI) adapt and enhance their products in accordance with the needs of the users. The technology providers (INRIA, OFFIS, Siemens, and Weizmann Institute) will close the remaining technology gaps.

Since non-functional requirements such as real-time properties and distributed architectures are strongly requested by users, the project will develop new technology, which supports real-time verification and distributed code generation. Novel techniques for modular verification and code generation will also be developed. These will allow large industrial applications to be treated with the new technology. Certification techniques for the new technology will be developed in order to ensure the realisation of savings in development time and cost.

Usage of intuitive visual techniques as well as highly automated verification and code generation ensures that formally-based technology can be utilised by systems engineers without a background in formal specification methods. This will substantially reduce the heavy costs of developing safety-critical systems, and at the same time increase flexibility and the ability to react to customer needs. In consequence, SACRES will substantially contribute to increasing the competitiveness of European industry.

An important goal of the SACRES project is to secure the vendor participants an adequate share of the strategically important development tool markets. Combining efforts to provide an integrated set of formalisms and tools will increase the attractiveness of the offer of all three vendors and offer the unique advantages of a multi-formalism approach to the user.

Marketing and acceptance of the SACRES results will be supported by information dissemination targeted to the technical community of safety-critical embedded systems and to target industrial clients. Broad acceptance and usability of the results will also be supported by providing and disseminating a methodology, which ensures that the tools and techniques developed within SACRES can be smoothly integrated into typical design flows for safety critical systems.


Siemens Aktiengesellschaft
Otto-hahn-ring 6
81739 Muenchen

Participants (3)

British Aerospace Limited, Military Aircraft Division
United Kingdom
Lancaster House, Fanborough Aerospace Centre
GU146YU Farnborough
Institut National de Recherche En Informatique et Automatique
Domaine De Voluceau, Rocquencaut
78158 Le Chesnay
Cp1,technopole Brest-iroise
29608 Brest