network for data exchange in agriculture

Improved farm management, optimised decision support models and the trace-ability of farm products require the automated acquisition of geo-referenced process data. Such a system for Automatic Process Data Acquisition in outdoor operations with tractor-implement combinations has been developed at the Technical University of Munich during the past few years. This system is based on the components LBS, GPS and IMI®. To analyse and to aggregate the huge amount of data collected in such a system an automatic evaluation program has been developed. This program, based on a SQL-based database-system and named IMIlyzer, shows a possible way for data analysing. Another developed and evaluated way is to implement measurement programs in the IMI®. The results of data analysing show a good comparability and possible assessment of fields by various accumulated values, relating to time and machinery efforts.

The Ground Truth Center Oberbayern (GTCO) is a center for the application of remote sensing data in agriculture funded by the Bavarian State. The center is coordinated by the Institute for Geography of the University of Munich. Partners are the Institute for Meteorology and the company VISTA. The goal of the GTCO is to support agricultural applications by using spaceborne or airborne remote sensing data. To achieve these operational methods for information extraction from remote sensing are developed using ground truth measurements as reference. User oriented products are derived from images depending on the customers needs. Besides the land use, also information on the biophysical stage of vegetation development can be gathered and from this information on actual growth can be derived. This is demonstrated by the spatial determination of agricultural yield based on satellite data which are used as input to an agro-meteorological growth model. Results are presented from a multi-annual analysis, which proves the validity and applicability of the method.

In the preagro project in the last two years a management information system for precision farming data was developed. To support interoperability in precision farming workflow the existing data model will be described and a way to setting up standards for data exchange by usage of XML will be demonstrated.

Es wird ein integriertes System vorgestellt, das räumliche Daten verwendet, um Informationen über standortangepasste N-Düngung zu erzeugen. Dabei liegt der Fokus des Beitrags auf einem neuronalen Netz, das vor allem zur Ertragsprognose herangezogen wird. Zudem können aus der Analyse des verwendeten neuronalen Netzes Rückschlüsse auf die Qualität der verwendeten Datensätze und deren Attribute gezogen werden.
The paper presents an integrated system that uses spatial data for the creation of information about site specific nitrogen fertilization. The focus of the article is on a neural network that is mainly used for yield prognosis. Conclusions concerning the quality of the used databases and their attributes can be drawn from the analysis of the used neural network.