Lead Project COGNAC - Cognitive Agriculture

Digitalization as a Chance for Agriculture

In the Fraunhofer COGNAC lighthouse project, eight Fraunhofer Institutes are conducting joint research on basic principles for producing agricultural products that are as environmentally friendly and resource-saving as they are highly efficient.

The industrialization has had a major impact on agricultural technology in recent decades, not only increasing productivity, but also causing negative effects: soil compaction, intensive fertilization, excessive use of herbicides, pesticides, and fungicides, or various types of genetically manipulated seedlings lead to permanent damage to the biosphere. Organic agriculture has evolved as an alternative; however, it deliberately accepts productivity losses. In the Fraunhofer Cognitive Agriculture lighthouse project, we will study what a networked ecosystem for agricultural processes must look like that can help to optimize both the economic and the ecological aspects in equal measure.

The project »Cognitive Agriculture« aims to automatically collect data about complex interrelationships in farming and, based on that, to support decision-making processes in the value network.

The focus will be on the following innovation areas:

• Open data exchange in an agriculture-specific, digitally networked ecosystem that enables multivalent use and linking of complex agricultural data volumes in secure data spaces;
• automated interpretation and decision support based on high-resolution measurement data from airborne or ground-based systems using multi-channel measurement information;
• autonomous field robotics for plant-specific field work as well as robot-guided sensor platforms with specific sensor systems.
   

Participating of Fraunhofer IOSB – Department Scene Analysis SZA

For farming, one of the most important information is the data basis about the condition of the arable soil and the plants. Without a temporal and spatial high-resolution digital monitoring of soils and plants, a specific and as automated as possible field cultivation such as weed control, fertilization, irrigation, and harvesting is not possible. On this account, Fraunhofer IOSB is concerned with biomonitoring through remote sensing. For this purpose, various autonomously operating and navigating sensor platforms (satellite, aircraft, drone/UAV) are equipped with appropriate sensors (multi-/hyperspectral, optical, UV, IR, LiDAR, etc.). Using multivariate statistics or AI, correlations between pedological or plant physiological parameters are analyzed from this data pool and machine-readable products such as digital field maps generated.

Therefore, Fraunhofer IOSB employs the latest hyperspectral sensor technology, whose ever-decreasing weight now even allows the installation on drones (DJI Matrice 600 Pro). Due to the low weight of 2.83 kg, the Hyperspec Co-Aligned VNIR-SWIR from Headwall enables a 12- to 18-minute flight per battery set. At an altitude of 80 m, this corresponds to an area coverage of approximately 4.2 ha and results in a pixel resolution of approximate 4 cm.

 

Cooperation with Startup „SmartCloudFarming“

In order to successfully establish the agricultural, digital field maps generated by Fraunhofer IOSB in practical use, Fraunhofer IOSB is collaborating with the Berlin startup "SmartCloudFarming GmbH". The young company plans to install rod-like soil sensors in the field, which locally provide 3D information about agricultural parameters. SmartCloudFarming aims to link the area-wide 2D parameter maps from Fraunhofer IOSB with their local 3D information through artificial intelligence (AI) to create area-wide 3D parameter maps. This information is crucial for understanding the variability of e.g. plant nutrient distributions in the arable soil so that excessive use of fertilizer and manure can be avoided.

With this pioneering technology approach, Fraunhofer IOSB took first place with SmartCloudFarming as a Fraunhofer startup cooperation (tandem) at the Fraunhofer Venture competition "TandemCamp 2019".