In this paper a model-based design approach currently developed is introduced to optimize the development process of automotive software. The approach plays special emphasis on a quality-oriented construction of embedded software to shorten the development life cycle and the development costs at the same time. "Quality-oriented" in this context means, that design and implementation decisions may be better traced back to the actual user requirements which are essential for the validation of the system. In contrast to low-level modeling approaches (such as Matlab/Simulink [1] or ASCET-SD [2], which mainly focus on technical aspects of the system), high-level modeling concepts are introduced to represent HW-/SW-architectures within a set of consecutive abstraction levels. A newly reworked system of automotive-specific abstraction levels is presented, where architectures are specified introducing more detail on each level. The system of abstraction levels supports the inheritance of model information from abstract levels down to concrete levels and the refinement of this information at each level. Thus the gap between (informal) requirements and the implementation is reduced. Since the higher levels abstract from technical details, reuse of models will be possible in a very easy way. The abstraction levels will form the basis for the strongly formal definition of an automotive specific architecture description language which we call "CAR-DL" (Combined Architecture Description Language). The presented approach is currently developed within the project "mobilSoft" 1. no details of HWenvironment distribution not considered lack of totality characteristic system serdistributable units technical environment concretization/completeness total functional specification abstraction
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.