How can ever larger volumes of scientific data be processed and evaluated? And how can Earth observation data be meaningfully combined with ground measurements, thereby opening up new sources of information? In the cross-sectoral Big Data Platform project, researchers from the German Aerospace Center (Deutsches Zentrum für Luft- und Raumfahrt; DLR) are devising new methods for the future-oriented field of Big Data Science. The interdisciplinary research project involves 21 DLR institutes from the research fields of spaceflight, aeronautics, transport, energy, digitalization and security. The project is set to run for four years and has received more than 21 million euros of funding. It is led by Simulation and Software Technology at the DLR.
The project addresses four key issues relevant in Big Data:
How can standardized access to data be enabled?
How can high data quality be ensured?
How can actual knowledge be derived from the data?
What are the potential benefits for society?
In the project we give answer to these questions in four separate work packages (Hauptarbeitspaket; HAP).
In the work package on platform techniques (Grundlegende Plattformtechniken) we focus on building a data science platform at the DLR and furthermore for the Helmholtz Association of German Research Centers. By using standardized tools and interfaces for software development, the double development of software can be avoided.
The work package on data management techniques (Datenmanagementtechniken) concentrates on high data quality. Even the best analysis techniques will fail if the provided information is insufficient or heterogeneous. Possible reasons for this are measurement errors, gaps in the dataset or data without relevance for the considered problem. Data in the Big Data Platform is collected from many different sources, such as planetary research, air traffic management and citizen science.
The data analysis is performed in a further work package (Intelligente Analysemethoden). Intelligent methods of data analysis, such as machine learning, allow deriving reliable knowledge from data. One specific example of this type of application is the high-precision identification of roads and road markings, from which free parking spaces within a city can be filtered out in real-time analysis.
Finally, pilotscale demonstrations (Pilotdemonstrationen ) address the potential benefits for society. As an example, due to automatically evaluated Earth observation data, rescue workers can be provided with support in crisis situations.