The resKIL project aims to holistically optimize resources (both people and hardware) through detailed problem analysis and generalization, development of a variable software architecture, and development of an AI toolchain. As a result, a scalable and variable solution will be developed so that machine learning will experience a much greater spread in agriculture in the near future. There will be a focus on the sub-areas described below. The developments will be integrated and implemented in functional models using the example of crop quality determination and feature recognition in the machine environment. An agile development strategy will be pursued in order to ensure a continuous increase in quality.

Three harvesting periods are at the center of the project planning. These define the sequence and iterations of the individual work packages. The transfer of the developments across several harvests is a further component in order to be able to demonstrate the generalizability and, in addition, an increase in the robustness and quality of the approaches.

Homepage