Generative programming: methods, tools, and applications
Generative programming: methods, tools, and applications
| Hi-index | 0.00 |
The development of software functions in the automotive domain is subject to multiple conditions. These conditions are for example the rising number of various functions in the car, the simultaneously increasing cost pressure and shortened development cycles. To come up with these conditions, an improvement of reuse is very promising. The use of model-based development techniques (e.g. the development with tools like Matlab ®Simulink ®/Stateflow ®) is a first step for reaching a higher level of reuse. But to have the possibility of modelling a function is not a guarantor for managing the increasing variability. The use of additional software techniques is needed to handle the variability. An example for such a technique is the software product line approach. In this approach the possible variability of a function is included in the development process, right from the start. In this paper, a concept for signal-oriented feature trees is presented, which is the precondition for using software product lines in the model-based development process. The advantages by using these feature trees, e.g. inclusion of nonfunctional features and generation of model-wrappers, are shown with an example of a battery-energy-management (BEM) function, which is developed in Matlab and has to be used in many different vehicle variants.