Software verification of an active automotive safety system using hardware-in-the-loop
In: ECCOMAS Thematic Conference on Advances in Computional Multybody Systems
University of Duisburg-Essen Institute of Mechatronics and System Dynamics Lotharstr. 1, 47057 Duisburg, Germany (2003)
Buchaufsatz / Kapitel / Fach: Maschinenbau
During the last few years vehicle safety has been improved significantly by introducing active and passive automotive safety systems. Whereas the complexity, interdependency and demands on the safety systems continuously increase, a trend to shorter development times with consistent quality occurs. Subject of the considered development process is the vehicle dynamics control system Electronic Stability Program (ESP, Robert Bosch GmbH). A challenge within the development process of complex mechatronic systems like ESP is to ensure the safety requirements on the system. Thus the development of the monitoring software as a part of the ESP self-diagnosis is an important element and has to be verified systematically by tests for quality assurance in all development phases. The combination of Hardware-in-the-Loop (HIL) real-time simulations and road tests is considered as the optimal solution for the monitoring software verification. The HIL application offers time and cost savings, reproducibility, performance of life-threatening driving maneuvers, the possibility for a test automation and the supply of state variables, which are not available in the real vehicle. But even the best simulation environment cannot exactly represent the reality and the results are only as good as the vehicle model, whereby road tests are still indispensable. The monitoring software is verified using a generic real- time vehicle model with one parameter set for a class of vehicles. The generic model is analyzed using a complex and validated vehicle model, which is based on a multibody systems approach including closed kinematical loops for wheel suspensions and drive train. Furthermore, alternatives to improve the real-time model by parameter adaptation are discussed. The simulation results with a tire parameter adaptation to compensate all modeling simplifications point out the potential to enhance the field of application.