Landwirtschaftlich genutztes Feld aus der Vogelperspektive
© iStock.com/sam74100

Online Seminar: »From the road to the field«

Fraunhofer IKS and the agricultural technology manufacturer John Deere discuss the extent to which technologies from the automotive sector can be applied in smart farming. Register directly:

more info

Smart Farming: Agriculture in Transition

The challenges facing agriculture

Agriculture is currently facing a multitude of challenges. On the one hand, the growing world population increases the demand for foodstuffs. Whilst, on the other hand, stricter sustainability regulations lead to a reduction in the area used for agriculture and the use of plant protection products and fertilizers. Lower margins for agricultural products, the shortage of skilled workers and the resulting increased cost pressures additionally necessitate increased levels of automation. Furthermore, climate-related extreme weather events, such as droughts and heavy rainfall, make cultivation more difficult.

Smart Farming: Digital technologies as a solution

Using conventional technology for fertilizing or harvesting fields cannot solve the current problems faced by agriculture. Approaches are required that are based on state-of-the-art information and communication technologies and that digitize and automate agriculture. In other words, Smart Farming. Smart Farming makes it possible to simultaneously render agriculture more sustainable, efficient and resilient. Lower costs for the sensor technology make investments in smart farming technologies attractive. That's why the market is currently being flooded with a variety of services from different manufacturers. In order to ensure the interoperability of these services, they are integrated into the Internet of Things for the Farm (Farm IoT), The Fraunhofer Institute for Cognitive Systems IKS, which specializes in intelligent, networked systems and the associated software applications, develops solutions for this field. It supports manufacturers of agricultural machinery, agricultural technology or agricultural software in the development of product innovations and solutions for smart farming. The focus here is on four key application areas of smart farming:

  • Sensor, satellite and drone data
  • Smart crop / livestock monitoring
  • Autonomous agricultural machinery and robotics
  • Smart building / farm management & IoT
Graphical representation of four application areas of Smart Farming:  Sensor, satellite and drone data, smart crop & livestock monitoring, autonomous agricultural machinery & robotics, and smart building, farm management and IoT.
© Fraunhofer IKS
Fraunhofer IKS focuses on four application areas of smart farming: Sensor, satellite and drone data, smart crop & livestock monitoring, autonomous agricultural machinery & robotics, and smart building, farm management and IoT.

Efficiency and sustainability through precision farming

Maisernte aus der Vogelperspektive
© iStock.com/JamesBrey

Precision farming is a subcategory of smart farming and describes the monitoring and optimization of agricultural production processes through the use of digital technologies. In precision farming, for instance, fertilizer is applied based on sensor, satellite or drone data, that measures varying conditions in the field. Based on the data, an algorithm - usually based on Machine learning and neural networks  - calculates the optimal fertilizer application (Smart Crop). More efficient use of crop protection products and fertilizers saves on costs and helps farmers comply with fertilizer regulations. Agricultural machinery manufacturers can use the data generated to improve their services. It also has a positive impact on the environment, for example by reducing nitrate levels in the soil and protecting surrounding wild plants and insects. Conversely, miscalculations lead to overfertilization of fields, resulting in crop failure and significant consequences for the environment. Due to the fact the system is capable of making its own decisions, it is called a cognitive system. Improving cognitive systems and making them safe is a core competence of the Fraunhofer IKS, which is also highly relevant for autonomous driving and Industry 4.0.

Autonomous agricultural machinery and robotics

As distinct from passenger cars, autonomous driving is already a reality where agricultural machinery is concerned. This is also due to the fact that fields are private plots of land and there are hardly any other road users to consider. However, due to the fact the fields and arable land are freely accessible, a high degree of safety must be guaranteed here as well. However, the special characteristics of the environment pose a particular challenge.

Harsh environmental conditions

In order to guarantee the safety of the driver and his environment, sensor technology for agricultural machinery must be able to cope with a complex environment. While passenger cars can orient themselves by the road including the road markings, where agricultural machinery is concerned there is at best a cutting edge for orientation. The unstructured, dynamic, and open landscapes complicate machine cognition. Added to this are uneven surfaces, dust and mud, which are particularly challenging for the optical sensors.

Hidden obstacles

Optical sensors reach their limits when harvesting tall crops such as corn, wheat or canola. This becomes a danger for people and animals that are under the tall plants, as that cannot be detected by the sensors. Therefore, in addition to optical sensors, agricultural machinery must also have infrared, microwave or other sensor technologies on board.

Precise work

A loss of precision in autonomous agricultural machinery can prove costly. This is because if, for instance, a combine harvester drives too far over the cutting edge that orients it, a strip of the field may have to be driven over twice. These efficiency losses result in huge revenue losses. In order to solve this problem, local antennas transmit correction signals with an accuracy of two centimeters, which enable precise driving.

Networked on the field

In the area of digital farm management, agricultural machinery, especially tractors, become a kind of interface for many applications from different manufacturers in Farm IoT. This is made possible by constantly transmitting data from sensors measuring the state of the machine, the crop volume or GPS position. It requires a stable connection to the cloud to this end. However, there are frequently inadequate telecommunications connections in rural areas. The absence of a connection could lead to a failure of the data transmission and thus to an early failure of the harvester. This represents a high financial risk for the farmer. To counteract these risks, the Fraunhofer Institute for Cognitive Systems IKS uses state-of-the-art web technologies for communication and development of distributed IoT applications. In a joint research project, Fraunhofer IKS and its partners Huawei and Holmer Maschinenbau GmbH were thus able to transfer the concept of predictive maintenance to harvesting machines.

Landmaschine der Holmer Maschinenbau GmbH
© HOLMER Maschinenbau GmbH