Donau Soja is participating in the EU funded H2020 project CYBELE. Starting in January 2019, 31 organizations collaborate to generate innovation in the European agriculture and food sector through the convergence of High Performance Computing infrastructure (HPC), Big Data analysis, Machine Learning methods, Cloud Computing and Internet of Things. Social, environmental and economic benefits, such as reduced scarcity and increased food security shall be achieved.
Soya yield and protein prediction demonstrator
Together with Biosense, Donau Soja contributes to the development of a yield and protein prediction model for soya fields. Based on satellite images, as well as on soil analysis, weather and further data, a software-model is trained to predict the harvest of soya fields, both in terms of quantity and quality. In order to elaborate a solid and reliable model, Biosense is going to derive methods, such as machine learning algorithms, to process as much available data as possible. Donau Sojas’ task is to collect data to verify ‘ground truth’. Therefore, georeferenced data on actual yields and protein contents will be gathered by the principles of crowdsourcing. Through the combination of satellite images and data measured in the field, a proper model will be derived, allowing reliable statements on the expected yield and protein content of the harvest. To achieve this, huge processing power and smart algorithms are needed, as the data must be analyzed as time series and cover the whole area of European Soya production (4.2 Mio ha in 2018).
In a further step, Donau Soja is going to elaborate and validate business models for the soya harvest prediction model and further applications based on it. Optimizing the sustainable European protein production through an overall increase in the efficiency by simultaneously decreasing the environmental impact motivates us to contribute to the Cybele project!
Find more information on the project website.
If you have further questions, please contact spreitzer(at)donausoja.org.
This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement no. 825355