Fertilizing crops more selectively

The growing world population needs more food than ever before. But arable land is limited. In addition, heat and drought - caused by climate change - as well as the use of pesticides have a detrimental effect on agriculture. Researchers at the University of Bonn and Forschungszentrum Jülich want to use AI methods and robotics to show ways towards sustainable agriculture.

The PhenoRob Cluster of Excellence conducts research on a more resource-efficient and sustainable production of crops. The project combines robotics, digitalization and machine learning with modern phenotyping, modeling and plant production. Agricultural crops will be observed, analyzed and treated in a more targeted way using new methods and technologies. Research is also being conducted into soil properties, local biodiversity and other important environmental factors. Since March 2020, the scientists have been conducting various field experiments at 15 locations to measure, among other things, water consumption, heat resistance and crop yields.

"With the PhenoRob project, we are trying to reduce the negative impact of crop production on the environment, such as the excessive input of nitrate into the soil or the application of too much chemical crop protection," says Prof. Heiner Kuhlmann from the University of Bonn. "With these techniques, we want to bring the fertilizer directly to the plant, remove weeds mechanically or by laser using robots, and treat plant diseases directly on the leaf".

Exploring plant growth with machine learning

The scientists use drones to document the vegetation.
The scientists use drones to document the vegetation. © University of Bonn/Excellence Cluster PhenoRob

Using ground robots and specially developed drones, the scientists systematically monitor all essential aspects of plant growth. They evaluate the huge amounts of data that accumulate in a short time using machine learning methods. In this way, they hope to improve our understanding of plant growth and the efficiency of nutrient and water use. The researchers are also working on 4D models of their database, which will be used for further simulations in the future.

The aim is to better understand the growth of plants. For practical applications in the field, this may mean that fertilizers and crop protection can be used in a more targeted and thus more environmentally friendly and economical manner in the future. The scientists are also investigating methods to strengthen the adaptability of crops to changing climatic conditions and to preserve biodiversity.

Application facts


Technology field
Data Management and Analysis
Image Recognition and Understanding
Sensors and Communication
Robotics and Autonomous Systems
Application industry
Agriculture
Education
Field of application
  • Robotics
    Miscellaneous
Value-added activity
Research and Development [R&D]
Funding
Bundesministerium für Bildung und Forschung (BMBF)
Sonstiges
AI Developers

Rheinische Friedrich-Wilhelms-Universität Bonn

University/Research Institution
Website

AI Development partner

Forschungszentrum Jülich GmbH
Fraunhofer-Institut für Intelligente Analyse- und Informationssysteme IAIS
Institut für Zuckerrübenforschung (IfZ)

Video