Automotive Software Engineering Research Articles

Automotive Software Engineering: Real-World Necessity and Significance

The automobile industry is undergoing a fundamental shift as it transitions from a mechanical to a software-intensive business, in which most innovation and competition depend on software engineering expertise. This shift is occurring due to the industry's shift from a mechanical to an electronic focus. Over the past few decades, the significance of software engineering in the automobile industry has grown substantially. As a result, it has garnered a great deal of interest from academics and industry professionals. Even though a considerable amount of information concerning automotive software engineering has been published in various scholarly journals, there needs to be a comprehensive study of this information. This systematic mapping project aims to classify and analyze the literature linked to automotive software engineering to offer a structured body of knowledge, identify well-established themes, and uncover research gaps. This study considers 679 publications from various academic fields and subfields published between 1990 and 2015. The primary studies were dissected and categorized based on five distinct dimensions of interest. In addition, potential holes in the research, as well as suggestions for directions for further investigation, are offered. The literature mainly focused on three different areas: system and software architecture and design, qualification testing, and reuse. These were the issues that were discussed the most frequently. There were fewer comparative and validation studies, and the research body needs to contain practitioner-oriented suggestions. Overall, the research activity on automotive software engineering has a high industrial relevance, but its scientific quality is relatively lower.

18th Workshop on Automotive Software Engineering (ASE’21)

18th Workshop on Automotive Software Engineering (ASE’21)

Case Study Experiences on Active Learning Methods in Germany

Abstract This article describes the implementation of active learning methods in engineering courses. The implementation was part of the activities of the international Erasmus project DriveMATH, which ran between 2017 and 2020. The active learning methods eduScrum, JigSaw, problem based learning, practical and team-based learning were used for courses for the master’s degree at Chemnitz University of Technology. Research seminar computer science, research internship and internship automotive software engineering were selected as courses for implementation. The students were invited to participate in the implementation on a voluntary basis. At the end of the implementation, the project team asked the students for feedback in the form of an evaluation survey. Criteria for the evaluation were developed in the team of the partner universities of the project. The data obtained were evaluated with the help of the structure-oriented evaluation model (SURE model).

Automated Detection of Autism Spectrum Disorder Using a Convolutional Neural Network.

Background: Convolutional neural networks (CNN) have enabled significant progress in speech recognition, image classification, automotive software engineering, and neuroscience. This impressive progress is largely due to a combination of algorithmic breakthroughs, computation resource improvements, and access to a large amount of data.Method: In this paper, we focus on the automated detection of autism spectrum disorder (ASD) using CNN with a brain imaging dataset. We detected ASD patients using most common resting-state functional magnetic resonance imaging (fMRI) data from a multi-site dataset named the Autism Brain Imaging Exchange (ABIDE). The proposed approach was able to classify ASD and control subjects based on the patterns of functional connectivity.Results: Our experimental outcomes indicate that the proposed model is able to detect ASD correctly with an accuracy of 70.22% using the ABIDE I dataset and the CC400 functional parcellation atlas of the brain. Also, the CNN model developed used fewer parameters than the state-of-art techniques and is hence computationally less intensive. Our developed model is ready to be tested with more data and can be used to prescreen ASD patients.

An Implementation and Evaluation Report of the Active Learning Method EduScrum in Flipped Class

This paper describes an implementation and evaluation report of the active learning method eduScrum in flipped class. The eduScrum test class was organized by volunteer participants of the master program Automotive Software Engineering at Chemnitz University of Technology, Germany. The “AUTOSAR based software design” seminar was selected as course for implementation the eduScrum method. The eduScrum method applied during the winter semester 2018/19 and the summer semester 2019. In the first round 30 of 80 students registered to the eduScrum class, and during the second round 40 of 85 students selected the eduScrum test class. The selected course is usually taught in a flipped class format. That was a challenging condition for the implementation of eduScrum. Due to this constrain the traditional way of implementing the eduScrum method needs to be modified and adapted to the given situation. Therefore, e-learning materials developed and placed into the university Learning Management Systems OPAL. The OPAL is online-based platform for academic teaching and learning at universities of Saxony. To measure the success of the eduScrum implementation the students within the test class did a self-evaluation based on the structure oriented evaluation method. The structure oriented evaluation model focuses on evaluation goals.

Safely Entering the Deep: A Review of Verification and Validation for Machine Learning and a Challenge Elicitation in the Automotive Industry

Deep neural networks (DNNs) will emerge as a cornerstone in automotive software engineering. However, developing systems with DNNs introduces novel challenges for safety assessments. This paper reviews the state-of-the-art in verification and validation of safety-critical systems that rely on machine learning. Furthermore, we report from a workshop series on DNNs for perception with automotive experts in Sweden, confirming that ISO 26262 largely contravenes the nature of DNNs. We recommend aerospace-to-automotive knowledge transfer and systems-based safety approaches, for example, safety cage architectures and simulated system test cases.

SMArDT modeling for automotive software testing

SummaryEfficient testing is a crucial prerequisite to engineer reliable automotive software successfully. However, manually deriving test cases from ambiguous textual requirements is costly and error‐prone. Model‐based software engineering captures requirements in structured, comprehensible, and formal models, which enables early consistency checking and verification. Moreover, these models serve as an indispensable basis for automated test case derivation. To facilitate automated test case derivation for automotive software engineering, we conducted a survey with testing experts of the BMW Group and conceived a method to extend the BMW Group's specification method for requirements, design, and test methodology by model‐based test case derivation. Our method is realized for a variant of systems modeling language activity diagrams tailored toward testing automotive software and a model transformation to derive executable test cases. Hereby, we can address many of the surveyed practitioners' challenges and ultimately facilitate quality assurance for automotive software.

2nd International Workshop on Automotive Systems and Software Architectures (WASA)—Introduction to special section

Automotive software engineering was officially introduced more than a decade ago into the software community addressing research challenges and technical issues encountering software development in the automotive domain. A modern-day premium class vehicle contains up to 100 000 000 lines of executable code running on multiple microcontrollers creating a heterogeneous, deeply coupled networking system. Nowadays trends like car2x, (fully) electric vehicles or self-driving cars are all based on software. Those new features will require new engineering approaches and more advanced software architectures suitable for automotive domain.

Automotive software engineering: A systematic mapping study

A comprehensive survey of literature on Automotive Software Engineering (ASE).679 primary studies were identified, classified and analyzed with respect to five dimensions.Three most investigated areas include software architecture & design, testing and reuse.ASE seems to have high industrial relevance but is relatively lower in its scientific rigor.Validation & comparative studies are less represented and literature lacks practitioner-oriented guidelines. The automotive industry is going through a fundamental change by moving from a mechanical to a software-intensive industry in which most innovation and competition rely on software engineering competence. Over the last few decades, the importance of software engineering in the automotive industry has increased significantly and has attracted much attention from both scholars and practitioners. A large body-of-knowledge on automotive software engineering has accumulated in several scientific publications, yet there is no systematic analysis of that knowledge. This systematic mapping study aims to classify and analyze the literature related to automotive software engineering in order to provide a structured body-of-knowledge, identify well-established topics and potential research gaps. The review includes 679 articles from multiple research sub-area, published between 1990 and 2015. The primary studies were analyzed and classified with respect to five different dimensions. Furthermore, potential research gaps and recommendations for future research are presented. Three areas, namely system/software architecture and design, qualification testing, and reuse were the most frequently addressed topics in the literature. There were fewer comparative and validation studies, and the literature lacks practitioner-oriented guidelines. Overall, research activity on automotive software engineering seems to have high industrial relevance but is relatively lower in its scientific rigor.

Deep Learning in Automotive Software

Deep-learning-based systems are becoming pervasive in automotive software. So, in the automotive software engineering community, the awareness of the need to integrate deep-learning-based development with traditional development approaches is growing, at the technical, methodological, and cultural levels. In particular, data-intensive deep neural network (DNN) training, using ad hoc training data, is pivotal in the development of software for vehicle functions that rely on deep learning. Researchers have devised a development lifecycle for deep-learning-based development and are participating in an initiative, based on Automotive SPICE (Software Process Improvement and Capability Determination), that's promoting the effective adoption of DNN in automotive software. This article is part of a theme issue on Automotive Software.

Improving the State of Automotive Software Engineering

The automotive industry is fundamentally changing by becoming software intensive, rather than mechanically intensive. To stay ahead of the game, automakers must continuously improve their software engineering. For this article, the authors studied the existing literature on the subject and made practitioner-oriented recommendations.

Yet Another Challenge for the Automotive Software: Deep Learning

Deep-learning-based systems are becoming pervasive in automotive software. So, in the automotive software engineering community, the awareness of the need to integrate deep-learning-based development with traditional development approaches is growing, at the technical, methodological, and cultural levels. In particular, data-intensive deep neural network (DNN) training, using ad hoc training data, is pivotal in the development of software for vehicle functions that rely on deep learning. Researchers have devised a development lifecycle for deep-learning-based development and are participating in an initiative, based on Automotive SPICE (Software Process Improvement and Capability Determination), that's promoting the effective adoption of DNN in automotive software.

A structured hazard analysis and risk assessment method for automotive systems—A descriptive study

The 2011 release of the first version of the ISO 26262 standard for automotive systems demand the elicitation of safety goals following a rigorous method for hazard and risk analysis. Companies are struggling with the adoption of the standard due to ambiguities, documentation demands and the alignment of the standards demands to existing processes. We previously proposed a structured engineering method to deal with these problems developed in applying action research together with an OEM. In this work, we evaluate how applicable the method is for junior automotive software engineers by a descriptive study. We provided the method to 8 members of the master course Automotive Software Engineering (ASE) at the Technical University Munich. The participants have each been working in the automotive industry for 1–4 years in parallel to their studies. We investigated their application of our method to an electronic steering column lock system. The participants applied our method in a first round alone and afterwards discussed their results in groups. Our data analysis revealed that the participants could apply the method successfully and the hazard analysis and risk assessment achieved a high precision and productivity. Moreover, the precision could be improved significantly during group discussions.

English

English

Engineering Automotive Software

The amount of software in cars grows exponentially. Driving forces of this development are the availability of cheaper and more powerful hardware, as well as the demand for innovation through new functionality. The rapidly growing significance of software and software-based functionality is at the root of various challenges in the automotive industries. These concern their organization, definition of key competencies, processes, methods, tools, models, product structures, division of labor, logistics, maintenance, and long-term strategies. This paper pinpoints the idiosyncrasies of the domain, characterizes the essentials of automotive software, and discusses the challenges of automotive software engineering

Software engineering for automotive systems at ICSE 2004 workshop summary

This paper provides a summary of the ICSE 2004 workshop on software engineering for automotive systems. We outline the main characteristics of automotive software engineering and indicate how the workshop papers relate to these characteristics. Furthermore, we summarize the discussion at the workshop.