AUTILE Framework: An AUTOSAR Driven Agile Development Methodology to Reduce Automotive Software Defects Open Access
Downloadable ContentDownload PDF
This research introduces AUTOSAR Driven Agile Development (AUTILE) Framework as a new methodology to develop automotive software with an aim to reduce the number of defects and their severity. Automotive software, developed in a traditional/legacy way, currently must be rewritten from scratch to support new features or modifications in associated hardware (Briciu, Filip, and Heininger 2013, 251-256; Boss 2012, 9-15; Kant, Buhlmann, and Kalhammer 2006). Disruptive automotive mega-trends: connectivity, electrification and autonomous driving; continue to demand new features in a complex software-base (Charette 2009; Haghighatkhah et al. 2017, 82-86). As a result, greater quality issues are faced when more functionality is added to traditional, non-standardized automotive software (Briciu, Filip, and Heininger 2013; Haghighatkhah et al. 2017, 82-86; Ahrens et al. April 2010; Kant, Buhlmann, and Kalhammer 2006; Lichtenstein, Dujmovic, and Baden-Fuller 2018, 1; Schumann and Thilo 2012; Serban, Poll, and Visser Apr 2018, 120-127; Tran et al. Feb 1990). Several researchers have proposed standardization of automotive software to improve software quality (Boss 2012; Kant, Buhlmann, and Kalhammer 2006; Briciu, Filip, and Heininger 2013, 251-256; Ahrens et al. April 2010; Lichtenstein, Dujmovic, and Baden-Fuller 2018, 1). However, as yet, the industry lacks a de facto standard.At the heart of the AUTILE Framework, we recommend following the Automotive Open System Architecture (AUTOSAR) standard as the basis to design modular and open software architecture. To further realize the benefits of modular software architecture designed in the first phase, AUTILE chooses and integrates Agile methodologies as its software development approach. Widespread adoption of Agile methodologies for automotive software development has not occurred (Katumba and Knauss 2014; Hohl et al. 2016; Knauss 2014, 33-47) and this research highlights the enablers and barriers in this domain. The proposed framework was applied practically to develop automotive software projects, demonstrating that defect reduction can be accomplished using the steps outlined in this new methodology.