Home // International Journal On Advances in Intelligent Systems, volume 10, numbers 3 and 4, 2017 // View article


Automotive Software Product Line Architecture Evolution: Extracting, Designing and Managing Architectural Concepts

Authors:
Axel Grewe
Christoph Knieke
Marco Körner
Andreas Rausch
Mirco Schindler
Arthur Strasser
Martin Vogel

Keywords: Architecture Evolution; Software Product Lines; Architecture Quality Measures; Automotive Software Engineering

Abstract:
The amount of software in cars has been growing exponentially since the early 1970s, and one can expect this trend to continue. To keep the software development for vehicles cost efficient, modular components with a high reuse rate cross different types of vehicles are used. Often, a product line approach is used to handle variability. As the underlying software product line architecture and its evolution are generally not explicitly documented and controlled, architecture erosion and complexity within the software product line architecture are growing steadily. In the long-term, this leads to reduced reusability and extensibility of the software artifacts, and thus, to a deterioration of evolvability. First, we propose methods used to extract initial product line architectures by recovery/discovery methods and describe our experiences gained from a real world example. Furthermore, we integrate this approach into an evolutionary incremental development process and show how a knowledge based process for architecture evolution and maintenance for architectural concepts can be implemented. The approach includes methods and concepts to create adequate architectures with the help of abstract design principles, patterns, and description techniques. Our approach helps software engineers to manage system complexity by suitable architectural concepts, by techniques for architecture quality measurements and by processes to iteratively evolve automotive software systems. We demonstrate our approach on a real world example, the longitudinal dynamics torque coordination from automotive software engineering.

Pages: 203 to 222

Copyright: Copyright (c) to authors, 2017. Used with permission.

Publication date: December 31, 2017

Published in: journal

ISSN: 1942-2679