Design of Embedded Controllers for Safety Critical Systems

Project IST-2001-38314


Project | Participants | Innovation | Work Plan  | Related Links | News


Description of the work

The Goal

Expected results





Home Project Prospects


Economic Development, Scientific and Technological Prospects

The project intends to develop novel algorithms and methodologies for the design and implementation of embedded controllers for safety-critical systems. A significant part of the project is concerned with fundamental research that has a long-time strategic horizon. However, since one of the deliverables is a design flow, we believe the impact on the industrial and academic community can be large. We expect that the EDA companies may adopt the flow as a standard at least in some application areas such as automotive design. Cadence Design Systems has initiated a focused effort in this market and it is likely that some of the ideas presented here may form a significant part of the methodology supported by the Cadence Design Systems tools. This is an interesting model of proliferation of academic research, as witnessed by the experience we had in the past. We will take as reference the area of EDA (electronic design automation) tools where the Berkeley team and the L’Aquila team have experience. In this area, revolutionary ideas developed in academia made it into the industrial world in a period of about ten years. In this case, it was essential for the developers of the ideas to participate directly in the development of the industrial messaging, of the technical directions, of the assembly of strong teams and the formation of start-ups that became large companies with leading market position (the two companies, Cadence and Synopsys, now hold the number 1 and 2 spots in the EDA world in terms of revenues, market capitalization and earnings). It is instructive to point out that the first application of the basic ideas came in large companies such as Intel, DEC, TI and ST where the industrial teams used the prototype university work having free access to the source code. In fact, the 486 and 586 design at Intel was done using extensively the tools developed at the University of California, Berkeley. This positive experience reported widely in the literature by these “illuminated” companies provided the push for the adoption of tools and methodologies by all other IC companies, thus giving impetus for the formation of the start-ups.

The essential lesson learnt throughout this experience is that, to really have an impact the economic development of industry we need:

bullet A “revolutionary” research with large potential advantages over existing techniques.
bullet A group of forward-looking companies with strong competitive pressures that create the need of innovation on a large-scale basis.
bullet The willingness and enthusiasm of researchers to engage in industrial applications and to deliver the missionary vision of the future.
bullet The creation of an appropriate support group either inside the major companies affected by the change or by start-ups whose mission is to proliferate the technology to the industrial ``masses'' on a worldwide basis.

We believe that the lack of one of these ingredients may create serious obstacles to the successful proliferation of innovation. We have seen an essential difference here between US and Europe and we believe that there is a large potential of innovation in the way research and industry interact. The composition of the partners makes it possible to exchange ideas on how to disseminate results in industry and to leverage the US experience in this topic.

In short, we are planning to place quite some effort to extend our theory to applications armed with experience contributed by some of the partners who had previous experience in this domain and with the previous results obtained over the years. In addition, we have explicitly allocated time and effort in a separate work package to disseminate results in industry and academia as well and to interact with other projects with significant overlapping with ours.