The three-year ASSERT project (standing for Automated proof-based System and Software Engineering for Real-Time applications) brings together the academic and industrial expertise of 29 partners from 11 countries in Europe. Co-ordinated by ESA, the project was selected in the area ‘embedded systems’ from the Information Society Technologies priority of the Sixth Framework Programme (FP6).

ASSERT’s primary goal is to define a new system and software development process, and experiment on real industrial cases. It will replace the traditional approach, which is very empirical, with a more scientific method. ASSERT will define a continuous proof-based process to ensure the correctness of computer systems for aerospace applications.

The definition of the process, integrating a large set of innovative technologies and tools provided by the 29 partners, will be the most challenging task of the project. Dr Eric Conquet, ASSERT Project Manager in ESA’s Directorate of Technical and Quality Management, explains, “We have set the standards high from the beginning of the project preparation phase in 2002. We all know that many challenges are waiting for us in the course of ASSERT, but the team is extremely motivated to fully deliver what has been promised.”

Fully complementing this new process, ASSERT will also define families of systems with their generic architectures. These families will be defined according to their required properties (safety, performance, availability, etc.), which can then be applied across all market segments - Earth observation, telecommunications, manned space and launchers.

After the ASSERT project is completed, it is hoped that the process of building a new real-time computerized application will be both more rigorous and more cost-effective. The ASSERT process will guide you fully through the development stages, from the selection of an appropriate architecture down to the final implementation with the continuous support of advanced modelling and code generation techniques. This will ensure that the right system is built, and that it is built correctly.

Because of the many technical aspects of this project, a multidisciplinary team has been assembled - with expertise spanning the fields of mathematics applied to computer systems, architecture description languages, software development and verification, ‘middleware’ technologies, CASE tools and industrial applications in aerospace fields.

Experiments performed in ASSERT will be used to define and assess the new process using real industrial cases. Two pilot projects, led by Alcatel Space and EADS, are defined on the basis of requirements from existing missions (ATV, Proteus) and from future missions (Aurora and GMES). These pilot projects will target different families of systems in order to validate particular concepts and to demonstrate the real benefits of the new system development process.

The pilot projects will include technologies to be used in the new process. A very innovative approach, called Proof-Based System Engineering (PBSE), brought in by a team of researchers led by INRIA (a French research centre), will make major scientific results available to ASSERT to ensure that a given system architecture is proven by design. This PBSE approach is fully complemented by a set of powerful technologies to support the analysis of dependability and real-time properties and the definition of the underlying middleware (activity under the lead of LAAS and ENST) on one side, and the modelling and verification of system and software designs (under the lead of VERIMAG and CS-SI) on the other side. The full integration of those technologies will ensure that the ASSERT process (defined by a team led by Synspace) can cover all phases of the system life cycle with a continuous proof-based approach. Dissemination activities (led by University of Padua) constitute another important dimension of ASSERT in order to promote the take-up of the obtained results.

The ASSERT challenge is also to show the benefits for the aerospace industry. Philippe David, Scientific and Technical Manager of the project, said, “We all know the drawbacks of the current development approach and all the consequences on our space projects. The new applications targeted by the future space programmes will reach a level of complexity that cannot be addressed with the existing technologies and processes. Our goal is also to ensure that the ASSERT results will be effectively used on future missions.”

The ASSERT Consortium started work in September 2004, with funding from the European Commission and contributions from the partners. The ASSERT partners are: ESA-ESTEC (The Netherlands) Coordinator - Alcatel Space (F), Alenia Spazio (I), Astrium (F), BSSE (D), CNRS-Verimag, CS (F), Dassault (F), DIT/UPM (E), Dutch Space (NL), EADS CRC (D), EADS MBDA (F), EADS ST (D), EADS ST (F), GET-ENST (F), ESI (E), Esterel Technologies (F), ETH (CH), Intecs (IT), Prover (S), SciScys (UK), SoftwCare (E), Terma Space (DK), TNI Valiosys (F), U. Padua (I), ONERA (F), Semantix (GR).  
 
Last update: 22 July 2008