Research methodology for a new validation concept in vehicle dynamics

Abstract

The operational validation has an outstanding function in simulation-based research, since it is an essential part in quality assurance. The validation’s objective is a proof of the simulation’s appropriateness for its specific application by assessing the grade of conformity between simulation and practical experiment. Today’s validation approaches in the scope of vehicle dynamics simulation show different deficits that limit the usage of validation studies’ results in practise and lead to mistrust in the simulation results. This manuscript demonstrates the research methodology for a new validation concept. The concept’s basis is statistical testing, which means that not only the measurements of a singular vehicle are compared to simulation results. Instead, the simulation’s accuracy is investigated for a set of vehicles that are represented by different model parameterisations. This analysis of the simulation’s accuracy in different operating points enables statistically verified statements on the simulation’s validity for the considered use case. The methodology for the development and for the analysis of the new validation concept is based on logically linked research questions. This manuscript focuses on the development of an improved process structure for this new validation approach and its transformation to validation studies in the scope of vehicle dynamics simulation. The specifics in generating appropriate vehicle dynamics measurement data for testing the validation method are explained and an insight into the method’s application in conducted validation studies is given.