The industrialization of automated driving functions according to level 3 requires an efficient test and calibration concept to deal with an increased complexity, growing customer demands, and a larger vehicle fleet offered. Therefore, a method for a complexity reduction of the calibration parameter space is presented. In the two-step approach, a qualitative sensitivity analysis is used to identify valid regions in the search space and subsequently decrease dimensionality based on the parameterspecific global influences. The reduced parameter space and sensitivity information can then serve as a starting point for an efficient calibration process on the target hardware. To examine the method's potential, our approach is applied to the parameter space of an automated driving function. The results expose clear dependencies between parameters and driving scenarios and allow an exclusion of parameter space dimensions based on sensitivity values. The predefined search space can be narrowed down to valid regions using the parameter range identification approach. Finally, the findings are validated with a quantitative variance-based sensitivity analysis. The validation confirms that our method provides equivalent results with a comparably smaller number of system evaluations.

No abstract

The upcoming market introduction of highly automated driving functions and associated requirements on reliability and safety require new tools for the virtual test coverage to lower development expenses. In this contribution, a computationally efficient and accurate simulation environment for the vehicle’s lateral dynamics is introduced. Therefore, an analytic single track model is coupled with a long-short-term-memory neural network to compensate modelling inaccuracies of the single track model. This ‘Hybrid Vehicle Model’ is parameterized with selected training batches obtained from a complex simulation model serving as a reference to simplify the data acquisition. The single track model is parameterized using given catalogue data. Thereafter, the long-short-term-memory network is trained to cover for the single track model’s shortcomings compared to the ground truth in a closed-loop setup. The evaluation with measurements from the real vehicle shows that the hybrid model can provide accurate long-term predictions with low computational effort that outperform results achieved when using the models isolated.