16. Workshop Automotive Software Engineering
Software Engineering 2019 - 18.2.2019 - Stuttgart
Motivation
Wie seine Vorgänger setzt sich der 16. Workshop Automotive Software Engineering mit der Problematik der Softwareentwicklung im Automobilbereich und folglich mit dafür geeigneten Methoden, Techniken und Werkzeugen auseinander. Die Automobilsoftware spielt heutezutage mehr denn je mit zunehmend vernetzten Fahrzeugen und modernen Fahrerassistenzfunktionen einschließlich des vollautomatisierten Fahrens eine wichtige Rolle.
Dabei steigt die Systemkomplexität nicht nur stetig an, sondern auch strengere Anforderungen an Zuverlässigkeit, Sicherheit (Security und Safety) und Datenschutz (Privacy) müssen insbesondere erfüllt werden. Der Trend zur Vernetzung hat das Fahrzeug längst erreicht. Zudem bauen immer mehr Funktionen auf Sprachsteuerung auf, um eine Handybedienung während des Fahrens zu ermöglichen. Das Autofahren wird somit durch voranschreitende „digitale Kulturen“ verändert: Menschen werden bald auf Dienste wie WhatsApp, Skype oder sogar Facebook vom Fahrzeug aus zugreifen können.
Wichtige Termine
Einreichung von Beiträgen:
11.01.2019
Entscheidung über Annahme:
29.01.2019
Einreichung finaler Version:
07.02.2019
Durchführung des Workshop:
18.02.2019
Ziele
In diesem Workshop werden Herausforderungen und Lösungsansätze des Automotive Software Engineering diskutiert. Beiträge aus allen Bereichen der Entwicklung von Software für moderne Fahrzeuge sind erwünscht. Der Workshop richtet sich gleichermaßen an Forscher, Entwickler und Anwender aus der Automobilindustrie sowie an Wissenschaftler aus Forschungsinstituten und Hochschulen, die im Gebiet Automotive Software Engineering arbeiten. Im Fokus stehen traditionell weniger theoretische, als vielmehr praxisnahe Arbeiten.
Programm
Regeln für den Ablauf des Workshops
- 1. Für einen Vortrag ist ein Slot von 30 min eingeplant (20 min Vortrag und 10 min Diskussion).
- 2. Für die Keynote von Matthias Tichy ist ein Slot von 50 min eingeplant (40 min Vortrag und 10 min Diskussion).
- 3. Die Vorträge können auf Deutsch oder Englisch gehalten werden.
| 08:30 - 09:00 | Registrierung |
| 09:00 - 09:05 | Begrüßung |
| 9:05 - 10:30 | Session 1 |
|
Keynote: Matthias Tichy
Adventures in Embedded Model-driven Engineering: Requirements and Architecture Abstract der Keynote ⇒ Abstract [PDF] |
|
|
Konstantin Zichler und Steffen Helke
R2BC: Tool-Based Requirements Preparation for Delta Analyses by Conversion into Boilerplates Automotive OEMs and suppliers negotiate different documents before they sign contracts for a product development. This includes the Component Requirements Specification (CRS), which is submitted by the OEM. The CRS describes the characteristics of the product to be developed in detail and is therefore the basis for the development effort estimation of a supplier. If the specified component is a successor of an already available product, the requirements specifications of both the successor and the predecessor products can be compared to estimate the development effort for the new component. This activity is called delta analysis. Due to a lack of sufficient tool support, the delta analysis is still a predominantly manual task. The main reason for this is, that the documents to be compared are structurally too different. In this work, we introduce a new method for an automated conversion of an OEMs unstructured or otherwise structured CRS into a structured language used by the supplier. The process uses established NLP tools to analyze CRS and then translates the OEMs requirements into supplier-specific boilerplates using a newly developed technique. The concept is implemented with the R2BC prototype, which demonstrates the feasibility of the approach and enables the processing of first real CRS. ⇒ Abstract [PDF] |
|
| 10:30 - 11:00 | Kaffeepause |
| 11:00 - 12:30 | Session 2 |
|
Benedikt Walter
State Machine Generation from Structured Natural Language Requirements State of the art in (automotive) system development are system specifications in form of natural language descriptions. Specification patterns are often used to avoid ambiguous and inconsistent expressions. Yet, neither natural language nor specification patterns provide sufficient overview about inner system dependencies and architectural structure. Requirements in form of finite state machines are suited for that but are often not accepted as sufficient exclusive form of documentation. To assure consistency when combining both approaches, it seems practical to derive one representation form from the other. We did not find a general formal method that solves this problem. We therefore propose a method that formally derives a finite state machine system representation from structured textual system specification by converting structured textual requirements into linear temporal logic expressions and map these into a state based representation form. This approach enables engineers to specify systems in (structured) natural language while having the ability to derive a structured, state based system overview any time during the specification process. We use an industrial system to evaluate the derived system state machine against the provided system test cases. This allows us to show correctness of the generated state machine and indicate usefulness of the proposed method. ⇒ Abstract [PDF] |
|
|
Philipp Hohl
An Assessment Model to Foster the Adoption of Agile Software Product Lines in the Automotive Domain The automotive industry is changing rapidly. E-Mobility, self-driving and other trends revolutionize the automotive industry. These trends have to be addressed when developing new car generations and future mobility solutions. Traditionally, software product lines have been used to cope with the large amount of software in cars by means of software reuse. Anyhow, there is a lack of knowledge about how to react flexibly on rapid changes with software product line development. The use of agile software development methods promise a flexible development with the possibility to react fast to changes. This talk presents the development of an approach for supporting organizations with the establishment of agile software product lines. In order to fulfill this goal, an assessment and improvement model, the so-called Agile Software Product Lines in Automotive Model (ASPLA Model), was developed to foster the adoption of agile software product lines in the automotive domain. The ASPLA Model identifies improvement potential for a change towards agile software product lines and provides a prioritized list of recommendations on how to combine agile software development with software product lines within the assessed development process. Main characteristics of the ASPLA Model are the compatibility with the widely-used ASPICE model, the integration of a new product-focused category and empirically-based recommendations. ⇒ Abstract [PDF] |
|
|
Falk Howar und Malte Mauritz Towards Scenario-based Testing for Automated Driving Functions Autonomous vehicles are envisioned to share the road with humans within the next decade. This change will revolutionize road traffic and provide a positive benefit for road safety, traffic density, and emissions. The general public and product liability regulations impose high standards on manufacturers regarding the safe operation of their autonomous vehicles. However, the documentation of the safety of such vehicles is an open challenge. Vehicles' behavior will be predominantly defined in software. It will not be possible to analyze or test software of such autonomous systems exhaustively for all possible real-world driving situations. One alternative envisioned strategy for documenting safety is to show statistically (and partly in simulations) that the software is safe enough in relevant traffic situations. To this end, it is necessary to develop techniques for modeling and simulating driving situations, specifying and monitoring functional requirements, and metrics for quantifying safety. We present our ongoing work on a framework for scenario-based testing of autonomous driving functions in simulations: Domain-specific languages are employed in the reflection of requirements as runtime monitors as well as in the modeling of test scenarios. The runtime monitors evaluate and identify relevant situations in simulations and real driving data for optimizing test suites. ⇒ Abstract [PDF] |
|
| 12:30 - 14:00 | Mittagspause |
| 14:00 - 15:00 | Session 3 |
|
Karlheinz Blankenbach, Jan Bauer, Mirko Conrad, Andreas Hudak, Frank Langner, Matthäus Vogelmann und Chihao Xu
Neue Ansätze und Methoden für die Fehlermodellierung und -behandlung bei automobilen Videodatenübertragungenstrecken Kamera-Monitor-Systeme und die damit einhergehende digitale Übertragung von Videodaten kommen zunehmend in sicherheitsrelevanten Systemen im Kraftfahrzeug zum Einsatz. Um sicherheitsrelevantes Fehlverhalten zu vermeiden, müssen diese Systeme gegen Fehler und Ausfälle abgesichert werden. Der Beitrag schlägt ein Fehlermodell für die Videodatenübertragung vor und skizziert mögliche Sicherheitsmechanismen zur Fehlererkennung und -behandlung. ⇒ Abstract [PDF] |
|
|
Patrick Kummler und Hansjoerg Fromm
A Closer Look on the Difficulties to Determine the Quality of Software Requirements Increasing demands on quality and complexity are a major challenge for the development of industrial software products. Automotive software in particular is subject to additional safety, security and legal demands. In such software projects, the specification of requirements is the first concrete output that is mostly written in natural language. However, in practice, two problem areas exist: First, due to reasons like lack of knowledge and missing experience of engineers, requirements quality often is not at a satisfactory level. Second, a massive increase of the number of requirements for software poses a scalability issue. In our research we want to take a closer look on the quality determination of software requirements. We present an overview of existing research approaches based on the standard ISO/IEC/IEEE 29148:2011 that offers nine essential characteristics for requirements quality. In addition, we analyze results from several sessions in which experts rate automotive software requirements. ⇒ Abstract [PDF] |
|
| 15:00 - 15:30 | Kaffeepause |
| 15:30 - 17:00 | Session 4 |
|
Dennis Hild, Martin Beckmann und Andreas Vogelsang
Traceability Analysis of a High-Level Automotive System Architecture Document More and more functions in automotive systems are enabled and controlled by software that is distributed over a large number of systems and components. In addition, the systems are more and more interconnected to implement the desired vehicle functions. Creating and maintaining a high-level overview of the relations between vehicle functions, systems, and components in a complete and consistent way requires a lot of effort. On the other hand, such a high-level architecture is beneficial for planning the development, analyzing the architecture, or defining product variants. In this paper, we analyze a real-world document that was manually created to document all vehicle functions, systems, and components of one car series and relations between these. We formalized the content of this model by providing a model of the concepts and a set of consistency rules. By evaluating the consistency rules on the given document, we found 213 contradictory relations and 547 missing relations. Based on these results, we conclude that manually maintaining such high-level architectures is highly error-prone and should thus be supported by automation and appropriate tooling. ⇒ Abstract [PDF] |
|
|
Ina Schaefer (Moderation)
Treffen der GI-Fachgruppe Automotive Software Engineering
⇒ Tagesordnung [PDF] |
Call for Papers
In diesem Workshop werden Herausforderungen und Lösungsansätze des Automotive Software Engineering diskutiert. Beiträge aus allen Bereichen der Entwicklung von Software für Automobile und Nutzfahrzeuge sind erwünscht. Beispiele für Themen sind:
- Architekturen, Schnittstellen und Technologien für die Fahrzeugvernetzung
- Architekturen und Methoden zur Entwicklung von Fahrerassistenz- und autonomen Fahrfunktionen
- Apps im Fahrzeug
- Big Data für intelligente Fahrzeuge
- Entwicklungsprozesse für Software im Fahrzeug
- Standardisierte Softwarekomponenten, sowie Baukästen für die Entwicklung von innovativen Applikationen
- Varianten- und Konfigurationsmanagement
- Evolution und die Aktualisierung von Software im Betrieb
- Qualitätssicherung sowie Analyse der funktionalen und nicht-funktionalen Eigenschaften
- Sicherheitsaspekte (Security und Safety)
- Verhaltensadaption auf Basis des Umgebungskontexts, z.B. des Fahrzeugs oder des Fahrers
- e-mobility
Die Einreichung für den Workshop (4-8 Seiten) erfolgt über das Konferenz-Management-System EasyChair. Ferner sind die IEEE-Formatvorgaben (2-spaltiges A4-Format) zu beachten. Alle Einreichungen werden einem PeerReview durch das Programmkomitee unterzogen. Akzeptierte Beiträge werden im Workshopband der SE2019 veröffentlicht. Für angenommene Einreichungen wird von mindestens einem der Autoren eine Registrierung für den Workshop erwartet.
Programmkomitee
Dr. Christian Allmann
Audi AG
Prof. Dr. Marcel Baunach
Technische Universität Graz
Dr. Mirko Conrad
samoconsult GmbH
Dr. Heiko Dörr
Model Engineering Solution GmbH
Prof. Dr. Volker von Holt
Ostfalia Hochschule für angewandte Wissenschaften
Prof. Dr. Thomas Kropf
Robert Bosch GmbH
Dr. Thomas Noack
Individual Standard IVS GmbH
Prof. Dr. Dirk Nowotka
Universität Kiel
Prof. Dr. Jörn Schneider
Hochschule Trier
Dr. Thomas Sauer
Volkswagen AG
Historie
- 15. Workshop ASE, 47. Jahrestagung der GI, Informatik 2017 in Chemnitz
- 14. Workshop ASE, 46. Jahrestagung der GI, Informatik 2016 in Klagenfurt
- 13. Workshop ASE, 45. Jahrestagung der GI, Informatik 2015 in Cottbus
- 12. Workshop ASE, 44. Jahrestagung der GI, Informatik 2014 in Stuttgart
- 11. Workshop ASE, 43. Jahrestagung der GI, Informatik 2013 in Koblenz
- 10. Workshop ASE, 42. Jahrestagung der GI, Informatik 2012 in Braunschweig
- 9. Workshop ASE, 41. Jahrestagung der GI, Informatik 2011 in Berlin
- 8. Workshop ASE, 40. Jahrestagung der GI, Informatik 2010 in Leipzig
- 7. Workshop ASE, 39. Jahrestagung der GI, Informatik 2009 in Lübeck
- 6. Workshop ASE, 38. Jahrestagung der GI, Informatik 2008 in München
- 5. Workshop ASE, 37. Jahrestagung der GI, Informatik 2007 in Bremen
- 4. Workshop ASE, 36. Jahrestagung der GI, Informatik 2006 in Dresden
- 3. Workshop ASE, 35. Jahrestagung der GI, Informatik 2005 in Bonn
- 2. Workshop ASE, 34. Jahrestagung der GI, Informatik 2004 in Ulm
- 1. Workshop ASE, 33. Jahrestagung der GI, Informatik 2003 in Frankfurt
Organisation
Prof. Dr. Andreas Vogelsang
Prof. Dr. Ina Schaefer
Prof. Dr. Steffen Helke
Die Organisation erfolgt in enger Abstimmung mit der GI-Fachgruppe Automotive Software Engineering, die den ASE-Workshop seit vielen Jahren veranstaltet.