Home | Impressum | Datenschutz | Sitemap | Fakultät Maschinenbau | KIT

Adaptive functional testing for autonomous trucks

Adaptive functional testing for autonomous trucks
Autor:

Elgharbawy, M.

Scherhaufer, I.

Oberhollenzer, K.

Frey, M.

Gauterin, F.

Links:
Quelle:

International Journal of Transportation Science and Technology
Volume 8, Issue 2, June 2019, Pages 202-218

Datum: Juni 2019 (Print-Ausgabe) / 02.12.2018 (Online)

Long-distance trucks are predestined for automated driving due to their high driving performance and long monotonous routes. Automation has the potential to increase road safety, improve fuel efficiency, optimise vehicle utilisation, increase driver productivity and reduce freight costs. Although the widespread use of full automation is not imminent, the vision of accident-free driving accelerates the evolution of driver assistance systems to higher stages of automation on the global market. The status quo assessment refers to functional testing as one of the key challenges for an economical, reliable and safe deployment of autonomous driving in the series development of trucks. Therefore, systems engineering has established data- and knowledge-driven test methods to ensure the required reliability of its products. In this scheme, the evaluation of software releases must be carried out in various phases up to the start of production. Initially through XiL technologies, then through driving simulators, test drives with trained test supervisors on test tracks and public roads, test drives by intended users and finally the homologation of vehicle types. This paper quantifies the conflict of objectives between the requirements of the test concept. Thus, a trade-off between efficiency and effectiveness criteria is achieved through adaptive test coverage of these driving functions in truck product engineering. The basics of the adaptive functional testing are presented, including commonly used verification and validation procedures. The industry-proven framework facilitates the criteria for evaluating the performance of automated driving functions and the measures for achieving a sufficient degree of maturity within the software quality management process.