Air Traffic Safety

Drones are operated more and more frequently. This has advantages, but it also causes risks and dangers for other aircraft, infrastructures and third parties. To minimize these risks, government-provided legal guidelines are needed, and their implementation requires effective detection and classification technology. A very well-suited and economical technology for this purpose is analyzing drone sounds with the help of machine learning algorithms. The aim of the AuDroK Project is to compile a dataset containing the sounds of different drone models and to develop classifiers that allow the identification of drones based on their sounds. The project will end with a model experiment for assessing the performance of the approaches developed. The project's successful completion will mean a contribution to improving the safety of drone flights and facilitating the integration of drones into our society.

In Germany, several different institutions are tasked with documenting air traffic incident reports. The great level of detail contained in these reports may allow the identification of individual persons, which is a violation of data privacy regulations. To prevent conflicts, the data must be anonymized. This anonymization process is currently very time-consuming and expensive because it has not been automated so far, and because the report's context must not be distorted in any way. The aim of the project is therefore to develop a program for the automated anonymization of incident reports in order to reduce expenses in terms of staff and time needed for data processing and promote the standardization of the reporting structure. Using a sample data set provided by the associated partners, the program will be trained to automatically identify sensitive data content and anonymize it without distorting its context. The project's long-term goal is to raise the acceptance of incident reports and to encourage more people to report problematic incidents in air traffic. 

The AuRa project deals with the pre-flight risk assessment for drone flights that fall into the »special category« with the aim of simplifying the time-consuming and expensive process of obtaining a flight permit, which today is carried out largely manually. The overall goal is to create an automated, digital verion of the process in order to accelerate the granting of permits and to facilitate the integration of drones into everyday traffic. Within AuRa, the partners plan to incorporate additional influence factors into the risk assessment process and to evaluate the parameters' data volume and quality in order to make the process more versatile. The automated solution will be tested within the project using an example application case.