Refine
Document Type
- conference proceeding (article) (1)
- Other (1)
Language
- English (2)
Publication reviewed
- begutachtet (2)
Keywords
- Deep Learning (1)
- Digital Twin (1)
- Geoinformation Data (1)
- Remote Sensing (1)
- Simulation (1)
Institute
Simulation of autonomous driving needs a realistic representation of the environment. There are several options. We propose a method,
which uses highly accurate global geoinformation (GIS) data from the City of Kempten for testing sensors, sensor fusion and driving functions (algorithms). The geo-information data is converted into an ASAM Open Simulation Interface (ASAM OSI) representation and benchmarked against ASAM OSI ground-truth data. This allows virtual validation and testing against highly accurate geo-referenced data
and a realistic environment representation.
Autonomous driving and traffic flow simulation requires a realistic and accurate representation of the environment. Therefore, this research focuses on the semantic segmentation of aerial images for simulation purposes. Initially, a dataset was created based on true orthophotos from 2019 and Kempten’s street cadaster, with true orthophotos being fully rectified aerial images. The chosen classes were oriented towards the subsequent conversion and usage in simulation. The proposed labeling workflow used cadaster data and demonstrated significant time efficiency compared to state-of-the-art datasets. Subsequently, a neural network was implemented that was trained and tested on the dataset. In addition, the network was also trained only on the lane markings to compare the network’s performance. Both cases demonstrated excellent segmentation results. The generalizability was then tested on true orthophotos from 2021. The results indicated a solid generalizability, but still needs to be improved. Finally, the aerial information was converted into a 3D environment, that can be used in simulations. Our results confirm the usage of aerial imagery and street cadaster data as a basis for the simulations.