This content is not included in
your SAE MOBILUS subscription, or you are not logged in.
Lean Model-Based Development: Practical Approach
Abstract:
Development pace of new embedded projects often requires usage of model-based design process (MBD). More individuals start using MBD without previous experience with tools and new processes. Matlab/Simulink/Stateflow is a common tool that is used in control applications in automotive and airspace industries. Because of its complexity, the tool has a steep learning curve. Therefore, it is vitally important to set the MBD environment that allows persons to develop real-life projects even without a deep knowledge of the tool. The quality of the product should not be compromised and the development time should not be extended due to the initial lack of knowledge of the tool by the developers.
The shifting to MBD leads to changes of roles and responsibilities of algorithm designers and software implementers. This shift is due to ability of creating of efficient production code by code generators. Code generators empower algorithm developers to produce production code of almost the same quality as the software engineers used to create by hand. However, the generation of that software is possible in only a well-defined MBD environment. That environment is set by those with deep knowledge of the embedded software as well as modeling and autocoding tool.
This paper describes the components of the Simulink-based MBD environment that are proven be effective in practice. Usage of the MBD environment streamlines the development and allows its automation. That, in turn, facilities the execution of lean MBD development process, increase the robustness of the processes, and leads to a high quality final software.