Various parties put new requirements onto the development of vehicles, e.g. from pollution control perspective much cleaner cars are needed which have more fuel-efficient engines. Additionally vehicles must become safer in order to reduce accidents and injuries of drivers and passengers. With the increase of comfort functions in our daily environments, we all wish to have also more comfortable cars and more driver assistance functionality.
Today these functionalities are implemented separately with dedicated control units. Such architecture has the big advantage that error propagation is limited. But on the other hand it leads to dramatically increasing complexity of the systems because of their necessary interaction. More or less every control unit has to interchange data with most of the others. This implies that the necessary performance and communication increase is dramatic.
Future cars will have even more functions with a higher complexity and a bigger software part. Many dependencies between the ECUs increase the overall complexity. Too many processors are needed and the communication overhead is disturbing the functionality of the systems. Current methods and tools already reached the limit to keep the complexity controllable. New strategies are needed, because new cars shall be developed even in a shorter time.
As a reaction to these shortcomings, the systems are changed to multi-core technology to reduce the amount of units and the AUTOSAR architecture is introduced for more flexibility in the allocation of functions.
In order to develop these new system topologies new tools are needed as well. In this project, we want to build a platform for this new generation of development environments, which supports the development of multi-core systems, considers product line engineering and produces AUTOSAR compatible software. In order to allow automatic software generation and to enable software reuse, these tools shall be based on model driven engineering, and support the related development steps.
The project was labeled in the fourth itea2 call (ITEA 09013) and is funded by the federal ministry of Education and Research.