Automated Assessment and Evaluation of Digital Test Drives

Stefan Otten, Johannes Bach, Christopher Wohlfahrt, Christian King, Jan Lier, Hermann Schmid, Stefan Schmerler, Eric Sax
Springer International Publishing
Lecture Notes in Mobility
Advanced Microsystems for Automotive Applications 2017
Within the last decade, several innovations in the automotive domain were introduced in the field of driver-assistance systems (DAS). As technology rapidly advances toward automated driving this trend further continues, integrating more intelligent, interconnected, and complex functionality. This results in a constantly expanding space of system states that need to be validated and verified. Approaches for virtualization of test drives such as X-in-the-loop (XiL) are in focus of current research and development. In this contribution, we introduce a concept for automated quality assessment within a randomized digital test drive. Our aim is to analyze and assess the continuous behavior of automotive systems during multiple realistic traffic scenarios within a simulated environment. Therefore an analysis and comparison of current test approaches and their verification and validation goals are conducted. These results are utilized to derive requirements and constraints for an automated assessment. In comparison to established systematic test approaches, our concept based on a continuous assessment of the entire test drive constituting multiple driving scenarios. To consider the continuous behavior and parallel assessment of different functionality, a distinction between activation conditions and test conditions is conducted. Additionally, the hierarchization of conditions allows identification and evaluation on different abstraction levels. We include general assessments in addition to system and function-specific behavioral assessments. The approach is elaborated on an example use case of an Adaptive Cruise Control (ACC) system.
Download .bib
Download .bib
Eingetragen von