Generating highly accurate position data using cooperative sensor data fusion

This subproject investigates cooperative sensor data fusion to increase the accuracy of estimated positions of road users. Thereby, the accuracy improvement compared to conventional methods shall be achieved by the fusion of different data sources, especially data from the sensitive surface layer. By using classical sensors such as Global Navigation Satellite Systems (GNSS), Inertial Measurement Units (IMU) and highly accurate map data, the coverage of the sensitive surface layer can be artificially extended to a longer road segment. Thus, highly accurate position information on the entire road surface is available to other subprojects. The position data make a fundamental contribution to the digital of the road system, as the other subprojects can use them to investigate, for example, complex road surface loads.

The service-oriented architecture (SOA) enables different data sources to be dynamically integrated and managed in the sensor data fusion at runtime. In addition to the sensors installed in the vehicle, such as GNSS and IMU, the SOA enables the integration other sources via vehicle-to-infrastructure (V2I) communication without interrupting the operation of the digital twin.

This subproject represents a central research contribution within the research field of cooperative localization, primarily due to the explicit consideration of the sensitive surface layer and the heterogeneous data sources.