Automotive Systems Engineering: A Personal Perspective
The complexity of modern vehicles requires new methods for vehicle development. Based on two decades of experience in automotive R&D the author motivates to take more benefits from systems engineering. Advantages are expected in three areas: objective
- PDF / 1,024,324 Bytes
- 19 Pages / 439.37 x 666.142 pts Page_size
- 83 Downloads / 246 Views
Automotive Systems Engineering: A Personal Perspective Markus Maurer
2.1 Motivation: Do We Need Yet Another Theoretical Framework? The complexity in the development of automobiles grows. In the 1960ies three body variants (sedan, sports car, and spider) were sufficient to satisfy customer needs (Heißing and Ersoy 2008). After 2000, Heißing distinguished 16 different types of variants (Heißing and Ersoy 2008). The portion of software and electronics at the added value is rising significantly. It changes the requirements on the development process itself. In addition, other tools and methods in car development are becoming necessary. The concept of automotive systems engineering sketched in this article is shaped by my personal experience as a researcher and developer in the field of autonomous road vehicles and driver assistance systems.1 In the late 1980s and the early 1990s researchers packed more computers into road vehicles than a sedan could carry. It was a milestone when sedans were able to demonstrate automatic driving functions in public traffic for the first time (Fig. 2.1, Maurer et al. 1994). In 2007 autonomous vehicles demonstrated in the framework of the DARPA Urban Grand Challenge that they were able to complete missions and, for a limited time, survive unmanned in a cooperative environment on a shut-down military airport among other autonomous vehicles and carefully operating human drivers (Darpa 2008). Driver assistance systems require complex software systems in order to perceive the environment and to execute appropriate actions supporting the goals of the 1
In this article systems are called driver assistance systems if they recognize their environment and the current situation with machine perception in order to support the driver appropriately in his driving task (Maurer 2012a, b).
M. Maurer (B) Institute of Control Engineering, Technische Universität Braunschweig, Hans-Sommer-Str. 66, D-38106 Braunschweig, Germany e-mail: [email protected] M. Maurer and H. Winner (eds.), Automotive Systems Engineering, DOI: 10.1007/978-3-642-36455-6_2, © Springer-Verlag Berlin Heidelberg 2013
17
18
M. Maurer
Fig. 2.1 VaMP: Automated driving functions with computer vision in the 1990s (Maurer 2000a, 2000b)
drivers. After two decades of personal experience, joint projects, and interviews with colleagues from Audi, BMW, Bosch, Continental, Daimler, Toyota, and Volkswagen, I can conclude that it is still a major challenge to develop complex, distributed safety-critical, mechatronic systems on time and in accordance to the quality and financial goals of the projects. Lately the reports of developers of hybrid vehicles from the companies above resemble those of the developers of driver assistance systems. Here too, the focus is on complex distributed safety-critical systems; again innovations are created which deeply affect the architecture of the overall vehicle. This article tries to elucidate why it is so challenging to develop driver assistance systems or hybrid drive train systems. An analysis by
Data Loading...
