Artificial intelligence in agricultural communication networks
Co-funded by the Federal Ministry of Education and Research (BMBF)
Duration: May 2020 - April 2023
Digitization is also present in large farms, where networks are created that collect and process data in order to plan and implement work processes more effectively. With independently working machines and artificial intelligence (AI), agriculture is moving towards so-called precision or smart farming, where efforts are optimized through automated, intelligent decision-making processes. However, there is a challenge in the limited mobile network coverage, which is necessary for an intelligent network of different machines, robots and drones to function. Therefore, we need an innovative solution with which automation technologies that are already available can be used.
The aim of the project is to create locally functional networks by turning drones into flying base stations and agricultural machines into mobile network nodes. This ensures that data from sensors can be transmitted wirelessly and that machines can communicate with one another. One challenge for this autonomous mobile network is the large number of available or already exisiting radio technologies, such as LTE or 5G-based technologies, which must be seamlessly integrated. At the same time, there are considerable requirements for a network that requires management in three dimensions due to the use of drones and which should reliably control time-critical processes. A regulation of complex processes such as the spraying of fields, in which several vehicles have to be localized and coordinated during operation, is not feasible for the individual user. In the project, an efficient AI-based control is therefore to be developed in order to equip this network management with the ability to self-organize.
The basis of this AI-supported control should be lightweight connection units (hardware gateways). These are used to load free-flying drones or cable copters in order to implement the smooth integration of various radio technologies. In this way, reliable and real-time data communication between machines and vehicles or with a cloud is achieved. The concept developed in the project contributes to the profit-increasing digitization of agriculture by integrating different radio technologies via self-organized network management. This is intended to demonstrate the usability of this approach under current and future market and technology conditions.