RAPR aims to develop a rule-based prototype to process space-oriented agricultural knowledge in order to economically grow organic commodities. The prototype is planned to be tested while supporting the existing advisory service for production and resource management. After its completion the system will be available as open source software.
The combination of digitized geological information and electronic data ascertainment via sensor on location provides a perfect base for numerous services, e.g. control and prognosis of growth and results. Such services enable a conservative, efficient, and economic cultivation, and eventually help to identify cultivable land to grow organic commodities.
Nevertheless, to utilize this knowledge for detailed advice in defined cases requires enormous efforts. Essential are expertise in agriculture as well as familiarity with available geological information systems, and the knowledge to control them. The manual retrieval, integration and processing of relevant information puts limits to the advisory service� scope and commensurability.
RAPR solves this problem by implementing an advisory support system: space-oriented agrarian expertise is converted into a suitable control language, building a permanent source of knowledge, which allows the system to autonomously devise inquiries to connected geological information systems, to evaluate the results and to combine the local data with the stored expertise. In doing so, originally separated sources of knowledge are integrated to build a potent device.
With RAPR, we expect to increase the advisory service� accuracy and at the same time reduce efforts significantly. In addition, the integration of geological data gathered by electronic sensors on location provides a foundation to collect sound knowledge on regional conditions in the long run.
In this joint project our task entails the development of the control system, the integration of existing geological information systems, and the acquisition and formalization of agrarian and geological expertise. We will create models and ontologies for semantic integration and realize rule editors as well as components for export, interaction, and visualization to process the results.
The system will be used and tested in existing advisory networks. It will be available for further development as open source software.
John Deere Werke Zweibrücken, Agricultural Management Solutions (AMS), Zweibrücken