2nd International ICSE workshop on

Software Engineering for Automotive Systems

St. Louis, 21 May 2005

www.infsec.ethz.ch/events/seas05/
        
[News]   [Theme]   [CfP]   [Dates]   [Submission&Contact]   [Travel]   [Keynote]   [Program]   [PC]   [Organizers]   [Abstracts]
[4th workshop in Minneapolis, 2007]  [3rd workshop in Shanghai, 2006]   [1st workshop in Edinburgh, 2004]


News


Theme and goals of the workshop

Automotive software is one of the emerging areas of software engineering and embedded systems. Within the next five years, premium cars are expected to host a cumulated amount of up to one gigabyte of binary code of software deployed via a set of interconnected embedded platforms. To design, implement and manage the complexity of such a huge, heterogeneous distributed system with increasingly short innovation cycles and a vast installed base, neither the techniques and methods of classical embedded systems are suitable, nor the known ones in the desktop and business software domain. To tackle this challenge, we need new adapted software engineering methods for the automotive domain that allow to specifically design the different software types, corresponding to their requirements, and to later on integrate the system parts into one reliable and manageable system.


Call for Papers

This 2nd workshop (have a look at last year's program) will address system integration as a central characteristic of automotive software. Automotive systems consist of a number of independently specified and developed sub-systems that have to be integrated into the automotive system. The integration must be based on nothing but the specification and development artifacts, such as interface descriptions etc. Because of safety and quality requirements on automotive software, this integration phase is of particular relevance for software development.
We address all facets of integration of independently developed software parts to one system with emphasis on the following aspects: The workshop is intended to provide a discussion forum for researchers and practitioners working in or interested in the field of automotive software. The organizers and the program committee address both academia and industry to transfer techniques and methods from other domains to the applicability in embedded systems.

Important Dates


Submission and Contact Information

Authors should submit papers that do not exceed 6 pages in the ACM SIG proceedings format. These articles must not be submitted elsewhere. Submissions will be handled by the general ICSE CyberChair, please click here. Please direct any inquiries to Christian Salzmann, firstnamedotlastnameATbmw-carit.de.

Venue and Travel

Please refer to the corresponding ICSE webpages. The workshop place will take place in Director's row #42.

Keynote Presentation

Bruce D. Emaus
President of Vector CANtech, Inc. and Chair of the SAE Embedded Software Standards Committee

Hitchhiker's Guide to the Automotive Embedded Software Universe or Don't Panic--There's Plenty of Tasks to Execute

slides

Abstract
Considered as more complex, more demanding, and near out-of-control, automotive embedded software continues traveling into the universe of N-dimensional computing, N-dimensional networking, distributed product architectures, clusters of distributed and partial distributed functions, reliability-entangled "x-by-wire" technologies and out to the far ends of the non-competitive boundary.
With the growing attention at the industry level, the OEM level, the supplier level, and the standards level, this presentation examines many of the business and technical issues surrounding in-vehicle software development and suggests some of the solution space that might be worthy of visiting.

The speaker
Bruce Emaus is the president of Vector CANtech, a company that specializes in development tools and embedded software components for automotive distributed applications. Mr. Emaus is also the chairman of the SAE Embedded Software Standards Committee and co-chair of the SAE Distributed Embedded Systems Engineering Task Force.
With over 30 years of product development experience covering embedded software, electronics hardware, and systems engineering, he is a leading expert in the area of distributed embedded systems and small area network protocols.
As a 1971 graduate of the University of Michigan, Bruce's accomplishments include the development of Ford's first trip computer, Ford's first internally-created integrated circuit, the creation of Ford's first peer-to-peer UART-based protocol, and the development of an early electronic music system architecture before the advent of MIDI. With working experience at companies both large and small, Bruce also ran his own electronics consulting company for over 20 years before joining Vector.

Program

All talks will take 25 minutes plus 5 minutes for discussion. The invited presentation is scheduled for 45-50 minutes plus 10-15 minutes of discussion.

Session I: Processes&Methods
09.00 09.05 Welcome A. Pretschner and C. Salzmann
09.05 09.35 Modeling Nonfunctional Requirements: A Basis for dynamic Systems M. Dinkel and U. Baumgarten
09.35 10.05 Automotive use case standard for embedded systems F. Pettersson, M. Ivarsson, and P. Öhmann
10.05 10.35 A Flexible Integration Strategy for Automotive Telematics Systems T. Bauer, J. Herrmann, P. Liggesmeyer, and C. Robinson-Mallett
10.35 11.00 Coffee all

Session II: Real Time and Safety
11.00 11.30 Real-Time Component Integration Based on Transparent Distribution E. Coste, C. Farcas, W. Pree, and J. Templ
11.30 12.00 Towards Verified Automotive Software J. Botaschanjan, L. Kof, C. Kühnel, and M. Spichkova
12.00 12.30 Overview of Existing Safeguarding Techniques for Automatically Generated Code I. Stürmer, D. Weinberg, and M. Conrad
12.30 14.00 lunch all

Session III
14.00 15.00 Keynote: Hitchhiker's Guide to the Automotive Embedded Software Universe or Don't Panic--There's Plenty of Tasks to Execute B. Emaus
15.00 15.30 Issues in Performance Certification for High-Level Automotive Control B. Weide, P. Bucci, W. Heym, M. Sitaraman, and G. Rizzoni
15.30 16.00 Coffee all

Session IV: Experience Reports
16.00 16.30 Experience of Introducing Reference Architectures in the Development of Automotive Electronic Systems U. Eklund, Ö. Askerdahl, J. Granholm, A. Alminger, and J. Axelsson
16.30 17.00 Business Situation Reflected in Automotive Electronic Architectures: Analysis of Four Commercial Cases J. Fröberg, K. Sandström, and C. Norström
17.00 17.30 Discussion and Wrap-Up all



Program Committee


Organizers

Abstracts

 M. Dinkel and U. Baumgarten
Modeling Nonfunctional Requirements: A Basis for dynamic Systems Management
The management of dynamic systems is an upcoming challenge for software engineers in automotive and other embedded systems. The complexity of current automotive computing systems is already difficult to handle for car makers and the expected growth in the area of electronic devices in vehicles will even intensify this situation. This paper presents a model based approach for for enabling automatic configuration of distributed component oriented systems. Nonfunctional requirements and capabilities of software components and platforms are explicitly modeled and provide for wellfounded statements whether a component is able to execute on a certain platform or not. With application models and platform models the validity of a configuration is defined in this paper. The models even allow reconfigurations based on information regarding the actual system context like user behavior, backend or environmental sensor information.

I. Stürmer, D. Weinberg, and M. Conrad
Overview of Existing Safeguarding Techniques for Automatically Generated Code
Code generators are increasingly used in an industrial context to translate graphical models into executable code. Since the code is often deployed in safety-related environments, the quality of the code generators is of paramount importance. In this paper, we will present and discuss state-of-the-art techniques for safeguarding automatic code generation applied in model-based development.

F. Pettersson, M. Ivarsson, and P. Öhmann
Automotive use case standard for embedded systems
Today the lack of deliveries, from suppliers to automotive OEMs, in early phases of software development cause a number of problems such as misinterpreted requirements and difficulties keeping deadlines. We have identified use cases as a mean to communicate and visualize requirements. To enable communication between different organizations a well defined standard is needed.
In this paper we propose a two level use case standard for embedded systems. The standard consists of a template and guidelines specifying how to write consistent and unambiguous use cases while capturing the necessary requirements.

T. Bauer, J. Herrmann, P. Liggesmeyer, and C. Robinson-Mallett
A Flexible Integration Strategy for Automotive Telematics Systems
This paper presents an approach for the planning of integration tests of automotive telematics systems. To our knowledge no sufficient solutions nor experiences on integration testing of automotive telematics systems on the basis of MSCs exist. No method for the determination of an integration order exists that takes the project and the system environment into account, which in our opinion greatly influence the integration order. Furthermore, most known test generation methods and structural quality measures demand syntactically sound MSCs to be applied efficiently. In our projects MSCs are often created manually from the scratch with many different tools, and MSCs are of rather low syntactical quality. Therefore, this paper addresses the determination of an integration strategy which can easily be adopted to changes in the project or in the system environment, and which can be manually applied to the given MSC-specifications.

U. Eklund, Ö. Askerdahl, J. Granholm, A. Alminger, and J. Axelsson
Experience of Introducing Reference Architectures in the Development of Automotive Electronic Systems
The requirements on increasing functionality, quality, and, customisation, while reducing cost has lead to the introduction of an architecture centred development process for electronic systems at Volvo Cars. This process enables better control of system integration and achieving non-functional requirements, such as reusability, understandability, etc. The result of the process is a reference architecture that includes strategies for implementing the balanced requirements, architectural views that provide means for reasoning about all the concerns of all stakeholders, and a top level design of the architecturally significant parts. The reference architecture guides the design of several projects, and thus, cost is optimised accordingly. The main contribution of this paper is that we present experiences from introducing the architecture centred process. The main conclusions are that disseminating and maintaining the reference architecture actually require more resources than developing it. Furthermore, experience shows it is difficult to create an architecture that enables a lot of different variants that is also strategically useable in the long term.

B. Weide, P. Bucci, W. Heym, M. Sitaraman, and G. Rizzoni
Issues in Performance Certification for High-Level Automotive Control Software
High-level supervisory control software for automotive applications (e.g., drive-by-wire) presents many challenges to making performance guarantees, which are a necessary part of the software's certification for deployment. The features of such systems demand that a compositional, or modular, approach to reasoning about performance be devised and applied. We discuss one such analytical approach as an alternative to simulation and testing.

J. Fröberg, K. Sandström, and C. Norström
Business Situation Reflected in Automotive Electronic Architectures: Analysis of Four Commercial Cases
The electronic architecture, of an automotive product, form a complex construct; including technology and methods, which ultimately should be chosen to optimally support the organization's own business situation. In this paper, we have analyzed the relationship of four automotive electronic architectures to their respective business requirements and business context. In the light of the business situation, we explain the solutions and why design principles are pursued. The analysis shows four electronic architectures that are quite different and the reason for this becomes apparent when looking at different business context and business requirements. Differences in business situation cause different focus in development effort.
An important conclusion from this is that applying technical solutions from one industry in another may not optimally meet business requirements and business context. Understanding the requirements from the business situation is the key to choosing architectural solutions.

E. Coste, C. Farcas, W. Pree, and J. Templ
Real-Time Component Integration Based on Transparent Distribution
This paper introduces a real-time component model that offers a separation of concerns which allows a straight-forward integration of independently developed components. So-called transparent distribution forms the backbone of the integration process. Transparent distribution means that (1) the functional and temporal behavior of a system is the same no matter on which node of a distributed system a component is executed and (2) the developer does not have to care about the differences of local versus distributed execution of a component. We first present the concepts of a component model for real time systems that is well suited for transparent distribution. The component model is based on logical execution time, which abstracts from physical execution time and thereby from both the execution platform and the communication topology. Then we discuss the resulting tool chain and integration process. A case study rounds out the paper.

J. Botaschanjan, L. Kof, C. Kühnel, and M. Spichkova
Towards Verified Automotive Software
Automotive software is one of the most challenging fields of software engineering: it must meet real time requirements, is safety critical and distributed over multiple processors. With the increasing complexity of automotive software, as for example in the case of drive-by-wire, automated driving, driver assitents, etc. software correctness becomes more and more a crucial issue. In order that these innovations can become reality, it is necessary to be able to guarantee software correctness.
The presented work aims at verification of automotive software. For this purpose it introduces a verification approach, including a framework of verified modules that assists the verification of the actual application. Feasibility of this approach was validated on a case study that also showed how verification can be integrated into the development process.

[News]   [Theme]   [CfP]   [Dates]   [Submission&Contact]   [Travel]   [Keynote]   [Program]   [PC]   [Organizers]   [Abstracts]
[3rd workshop in Shanghai, 2006]   [1st workshop in Edinburgh, 2004]
last modified Jan-26-2005, Alexander Pretschner