Synthetic Object Placement with Statistical Representations Regarding Real Data Sets

Collecting automotive sensor data for AI-applications is time and cost expensive. One solution is the usage of simulated data. However, the quality of the applications results highly depends on how close the simulated scenery is to reality. One main challenge is the definition of the properties of dynamic objects. A simulation of their behaviour over time is possible, but for this, extra software e.g. SUMO or CARLA is needed. This paper therefore gives a proposal for defining those properties without further simulation software. The proposal uses statistical representation from real automotive data sets. This gives the advantage of direct object placements. In addition the objects properties are as near as possible to reality. Another advantage is the creation of more data frames as a data set can provide. First, several parameters are collected from real data sets. Then the parameters will be analysed regarding their statistical appearance. This analysis is then the basis for mathematical representations. Further metadata will be collected separately. Here, standards as openDrive are important tools. For validation, the objects are placed in a simulation environment (CARLA). For the Definition of Experiment (DoE) here, sampling methods as Latin Hypercube Sampling are fed with the previously determined analysis. In the end, the synthetic data is validated.