The project brings together several research institutions: the Technology and Support Centre (TFZ) in Straubing, Weihenstephan-Triesdorf University of Applied Sciences (HSWT) at the Straubing Campus of the Technical University of Munich (in particular the Professorship of Bioinformatics led by Prof. Dr. Dominik Grimm), and the Bavarian State Research Center for Agriculture (LfL).
The increasing demands for sustainable crop production require innovative solutions, as conventional large-scale chemical plant protection measures are associated with significant environmental and cost-related risks. The use of drones in combination with modern pattern-recognition methods represents a promising approach that can increase the efficiency of weed control in the medium to long term while simultaneously reducing the use of plant protection products.
Within the project, weed application maps are being developed that depict the spatial distribution patterns of weed infestation in agricultural fields. These maps are based on extensive data from drone flights over sorghum and maize fields, which are subsequently analyzed using AI algorithms. The resulting maps are intended not only to facilitate the localization of weeds, but also to serve as a control basis for field robots that carry out mechanical weed control measures precisely where they are needed. This enables site-specific treatment, reduces soil erosion, and strengthens plant health.
A video feature on the project has now been released by the Bavarian Competence Network for Digital Agriculture (KNeDL): https://youtu.be/XyFvBe4bY3Q
