Fraunhofer-Gesellschaft

Publica

Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Simulations on Consumer Tests: A Systematic Evaluation Approach in an Industrial Case Study

 
: Berger, C.; Block, D.; Heeren, S.; Hons, C.; Kühnel, S.; Leschke, A.; Plotnikov, D.; Rumpe, B.

:

IEEE intelligent transportation systems magazine 7 (2015), Nr.4, S.24-36
ISSN: 1939-1390
Englisch
Zeitschriftenaufsatz
Fraunhofer FIT ()

Abstract
Consumer tests which assess safety features of modern vehicles have a tradition in Europe. Recently, such test protocols have been substantially extended to also cover active safety systems like Volkswagen's Front Assist. Objective: Simulations for passive safety systems are already a widely adopted approach during vehicle development and internal assessments. As active safety systems are becoming an increasingly important element in a vehicle's safety concept and a differentiating feature, a systematic validation and assessment of such systems is necessary to successfully pass consumer tests and complimentarily identified, relevant traffic scenarios. Method: With this work, we extend our previous conference publication about EuroNCAP CCRs tests by additionally investigating US NCAP scenarios for an AEB system. Therefore, we systematically modeled the allowed variations with a graph where the paths represent concrete test scenarios. These paths are used in a virtual test environment to assess the AEB system. In our previous publication, we illustrated our method of test case generation and simulating consumer test scenarios by showing results of 27 specific test cases. In this work, we focused on integrating a test automatization routine as well as evaluating a set of test cases with a factor of 100 compared to our previous paper. Results: We demonstrate the approach for both EuroNCAP's and US NCAP's CCRs scenarios with a total quantity of more than 2,700 test cases including re-runs to systematically evaluate an AEB algorithm. Our results unveiled varying action points in time for the same initial values for a given consumer test scenario while applying different allowed variations. Conclusion: We foresee the importance of complementary virtual testing for real-world tests on proving grounds especially during the design phase. Our study shows that already small variations that yet accord with the test procedure specification influence the behavior of - n active safety system and need to be investigated during development and vehicle testing.

: http://publica.fraunhofer.de/dokumente/N-402461.html