At the end of the year 2002, ESA (European Space Agency) and a group of partners decided to build together a major research project to improve the system and software engineering process for critical and real-time system.
This first group of partners has been very quickly extended to a consortium of around 30 partners embracing a large spectrum of competences from research laboratories to highly experienced industrialists from the aerospace domain, through SMEs and SW houses addressing tool development.
Starting with a huge ambition and a strong vision to the future, the consortium started to operate in September 2004 with the financial support of the European Commission under the 6th framework programme. The project, named assert (Automated proof-based Software and System Engineering for Real-Time systems), targeted the definition and implementation of a new process securely built on strong foundations :
system families and the use of modelling and proofs.
This project is there to push innovation in software engineering. Software engineering and many activities in the IT sector are too much focused on programming, producing code and optimising it. Software is something invisible until it can be materialised through real code and that is the obsession of software engineer to show real code. Although there are many design methods and languages, the discipline is still immature when real engineering is concerned. There are too few ways of guaranteeing software properties (reliability, maintainability, safety, performance,…) before implementation, and the existing methods (the so-called formal languages) are rejected by the programmers because of their complexity or lack of integration in the process and their incompatibility with the culture of software people.
The main consequences of such an approach are well known and their impact can be seen every time you use software, which really means every time of your life. If the engineering approach we use in the software business was used in the civil engineering domain, we probably will think twice before crossing a bridge. We have to admit that such an approach will not be able to cope with the increasing complexity of autonomous spacecrafts cooperating to perform a single mission.
