FELD 01/2022

Artificial Intelligence

Artificial Intelligence THE ART OF KNOWING BETTER Farmer Schulz examines the blighted spots on his maize with suspicion. He cannot yet say exactly what is damaging his plants. Is it a fungus? Or is it something else? As a result of climate change, new pests are increasingly appearing in his region. More and more often, farmer Schulz can no longer rely on his decade-long experience. New knowledge and new solutions are needed. So far only a small corner of his field is infested, but he has to act quickly! He has little choice but to generously spray the field with pesticides and warn the neighbouring farms. The news can be spread quite quickly via his smartphone. The prerequisite, however, is that all farmers are networking with each other and that the affected farm voluntarily reports the unpleasant discovery. A NEW ERA 9 13 15 Gone are the days when artificial intelligence only appeared in science fiction. Thanks to increasingly powerful computers, it is becoming more and more an important part of our lives. In the future, even agriculture will not be able to manage without artificial intelligence. Already today, robots and drones can take over certain field operations. In times of climate change and species loss, however, there are far greater challenges to be faced. Artificial intelligence is expected to help develop the sustainable agricultural systems of the future. But before that becomes reality, even artificial intelligence still has a lot to learn. The example of farmer Schulz is a fictional one. However, it helps Prof. Masahiro Ryo to point out the future challenges and the potentials of new technologies. Ryo heads the new working group »Artificial Intelligence«, or AI for short, at the Leibniz Centre for Agricultural Landscape Research (ZALF). »When it comes to new findings, for example on plant pathogens, the effectiveness of protective measures or climatic changes, even experienced farmers are dependent on personal exchange, further training and their own observations and enquiries«, says Ryo. »This is an additional workload in the already crowded daily routine of many farms. It can take a long time for new knowledge to arrive where it is actually needed«. A field robot controlled by artificial intelligence, on the other hand, is connected to a network of hundreds of other robots and drones. This network simultaneously tracks the spread of diseases or pests in a region with hundreds of cameras. Another advantage is that the network can also be linked to other sources of information such as weather and climate data, market prices or legal requirements. »Even the knowledge of how to safely detect the new pest and to treat infested plants can be accessed at the push of a button, even if the knowledge originates in a different country«, says Ryo. Another advantage: modern machine technology can specifically treat only those areas of the field that are actually infested. Other tasks, such as fertilization or weeding, are also only carried out where deemed necessary. The result of this so-called »precision farming« is a significantly more sustainable and resource-efficient agriculture. For many years, the machine 34 35