AMALTHEA

With the steadily rising number of comfort, drive and driver-assistant systems in modern cars, the need for more computing power in vehicles is increasing. Systems to reduce exhaust emission and to increase occupant protection in crashes are always becoming more and more complex. This leads to a rising number of electronic control units (ECUs) and associated data transfer in modern vehicles. At the same time, ECUs not only have to be more cost-effective and optimized in energy efficiency, they also have to be more powerful. 
Yet, in the field of single-core systems, for example, the boundaries for computing power and also for instruction level parallelism (ILP) are reached, which means more and more multi-core systems on control units are used. On the one hand, this enables a small number of ECUs in a vehicle, on the other hands this means an adaption or even redevelopment of the affected, already existing, applications is needed, which were run on single-core systems up to now. For this, appropriate tool support is needed, which not only offers multi-core support but also allows for the AUTOSAR standard. Another aspect is the multitude of varieties of a type of vehicle, which are offered by the vehicle manufacturers. This diversity should preferably already be regarded in the product line development and is already regarded in the development process of a vehicle.

Therefore, the goal of the AMALTHEA project is a consistent, extendable tool platform on eclipse basis with the following features:

  • Model-based software development
  • Multi core support
  • AUTOSAR conformity
  • Integration of product line development into the development process

In doing so, OpenSource as well as commercial and corporate-owned tools should be allowed to be included into the continuous tool chain. 

In this project, C-LAB is involved in the development of concepts and methods for scheduling and timing analysis for multi core control units. Most importantly, time modeling is supposed to be integrated in the development process, so that early statements on the interaction of the developed software and the chosen target platform can be made. Also, if required, this allows making necessary corrections on the system draft already in the early development phases. 

Project funded by: BMBF (ITEA2 project)
Project duration: 07/2011 - 12/2013


Associated Partners:

 Behr Hella Thermocontrol (D), C-LAB (D), ETAS GmbH (D), FH Dortmund (D), FH Regensburg (D), ifak (D), Inno-W (FIN), itemis AG (D),  Metso Automation (FIN), Mobilera (TR), Nokia Siemens Networks (FIN), Robert Bosch GmbH (D), Timing Architects (D), TOFAS (TR), Universität Oulu (FIN), Universität Paderborn (D)


Contact: 

Dr. Bernd Kleinjohann, C-LAB