Search Results

Car Periphery Supervision (CPS) systems comprise a family of automotive systems that are based on sensors installed around the vehicle to monitor its environment. The measurement and evaluation of sensor data enables the realization of several kinds of higher level applications such as parking assistance or blind spot detection. Although a lot of similarity can be identified among CPS applications, these systems are traditionally built separately. Usually, each single system is built with its own electronic control unit, and it is likely that the application software is bound to the controller's hardware. Current systems engineering therefore often leads to a large number of inflexible, dedicated systems in the automobile that together consume a large amount of power, weight, and installation space and produce high manufacturing and maintenance costs.

Legislative bodies are directing that automotive products comply with stringent safety levels. The liability for the safety of passengers in an automobile has traditionally been quite complex. Other transport sectors are externally regulated, and liability lies with the manufacturer or the transport service provider. The automotive industry is self-regulated and the individual driver carries a significant liability. Software and electronics increasingly provide greater control of automotive safety, possibly reducing driver liability, and increasing the need for more formal software development methods. The automotive business model, however, also presents challenges to the effective use of formal methods. An automotive design change costing €600 per vehicle could consume 100% of gross margin. In aviation, this cost represents 0.01% of gross margin [1] [2].

In the automotive industry a growing number of mono-functional control units with increasing complexity on one hand and requirements for reduced power con-sumption and mounting space on the other hand are enforcing an architectural change of the car electronics. Computer platforms with a client/server architecture are candidates to reduce the number of control units drastically followed by a reduction of costs, space and better integration possibilities for enhanced func-tionalities as well as additional services. The problem which comes up now if those architectures are coming to the cars is to cope with software complexity and reliability issues under the aspect of continuously evolving hardware infrastructure. To understand the problems better, the Corporate Engineering within Robert Bosch GmbH has build a multifunctional look-and-feel demonstrator in a first step using a component-based software architecture on a standard PC-platform. Experiences are reported.

The ever increasing complexity of embedded automotive software is not matched by the current development and test processes of automotive embedded software and the latter have become the limiting factor. A model-based software development and testing approach has the potential to reduce software development times, to produce executable specifications very early in the process as well as facilitate automatic code generation. Not surprisingly, the above are regarded as highly beneficial for the automotive industry. The automotive industry is increasingly using model-based testing techniques. Despite this, model-based testing tends to be done in a bespoke and non-systematic fashion [1] and easy to use, high quality, formalised, model-based testing methodologies that cater for the specific needs of in-vehicle software are hard to find.

Product line approaches are well-known in many manufacturing industries, such as consumer electronics, medical systems and automotive [1]. In recent years, approaches with a similar background have rapidly emerged within Software Engineering, so called Software Product Line (SPL) approaches [2], [3]. As automotive manufacturers and suppliers design and implement complex applications, such as driver assistance [4], they strive for mechanisms that allow them to implement such functionality on integrated platforms. This offers the opportunity to build a variety of similar systems with a minimum of technical diversity and thus allows for strategic reuse of components. This has resulted in a growing interest in SPL approaches both in the software engineering and the automotive systems domain. This paper discusses the increasing importance that SPL approaches could play within the context of Automotive Systems Engineering.

The amount and complexity of software in automotive systems is constantly increasing. Today's luxury cars include numerous electronic control units. A large part of the functionality of these units is driven by software. In the future even more software-intensive automotive systems are expected as automotive manufacturers and suppliers tend to integrate and combine applications on more powerful platforms. The increasing amount and complexity of software in these platforms has led to the situation where software engineering has become an essential discipline within automotive systems development. This paper identifies essential areas of software engineering that will have a significant impact on future automotive systems and systems development.