Page number 4
INTRODUCTION General Agriculture has always been a driver of innovation of new technologies and developments, especially in Germany. Johnson describes a direct link between the mechanisation of agriculture and the prosperity of a country1. This is particularly evident in the number of people working in the agricultural sector. In Germany, the number of people employed in agriculture fell from 22.5% in 1950 to 1.5% today. Despite all this, more food is currently being produced in Germany than ever before2. The average yield per hectare of a wheat field has quadrupled since 1900 from 1.85 tonnes to 7.64 tonnes in 20173. The mechanisation and automation of production processes had a particular influence on the advancement in agriculture4 [4]. According to the German Federal Statistical Office, 6.4 billion euros were spent on new and highly-automated agricultural machinery in 2017 [2]. This high sum underlines the willingness of the agricultural industry in Germany to invest in new innovative technology. The requirements with regard to agricultural machinery have changed over time. Some features of earlier agricultural machines are no longer in demand today, while new skills are required. Typical changes over the years included absolute machine size, engine power, comfort, safety and modern automatic steering and driver assistance systems5. Digitisation does not stop at agricultural engineering and will continue further in the future. “Deep Learning” and “Internet of Things” (IoT) have been around for a long time in agricultural research and are now being used in the first commercial products. Some examples of this in practice are smart phone apps for determining plant diseases and weeds6. Automated robotics solutions can now also be purchased and offer added value for the user (e.g. hoeing weeds in row crops). Robots in the field offer the opportunity to extend digitisation down to individual plants. Therefore, on the level of a “digital twin”, the development of the plants can be examined even more closely7. 1 Δ G. Johnson, “Agriculture and the Wealth of Nations” in Papers and Proceedings of the Hundred and Fourth Annal Meeting of the American Economic Association, 1997, vol. 87, no. 2, pp. 1–12. 2 Statista, “Landwirtschaft in Deutschland”, 2019. [Online]. Available: https://de.statista.com/statistik/studie/id/6455/ dokument/landwirtschaft-statista-dossier/. [Accessed: 04-Sep-2019] 3 P Pascher, U. Hemmerling and S. Naß, Situationsbericht 2018/19 „Trends und Fakten zur Landwirtschaft“. Deutscher Bauernverband e.V., 2018 4 M Kassler, “Agricultural Automation in the new Millennium” Comput. Electron. Agric., vol. 30, no. 1–3, pp. 237–240, Feb. 2001 5 J. N. Wilson, “Guidance of agricultural vehicles - a historical perspective” Comput. Electron. Agric., vol. 25, no. 1-2, pp. 3–9, Jan. 2000. 6 A Kamilaris and F. X. Prenafeta-boldú, “Deep learning in agriculture: A survey” Comput. Electron. Agric., vol. 147, no. February, pp. 70–90, 2018. 7 A Linz, J. Hertzberg and J. Roters, “„ Digitale Zwillinge “ als Werkzeug für die Entwicklung von Feldrobotern in landwirtschaftlichen Prozessen Material und Methoden” in Lecture Notes in Informatics (LNI), Gesellschaft für Informatik, Bonn, 2019, pp. 125-130. 4
Page number 5
With the expansion of digital infrastructure, these technologies will create new opportunities for food production of the future. After many innovation cycles in agriculture such as through mechanisation, pesticide development, mineral fertilisers, plant cultivation and precision farming, digitisation is credited with the potential for the next innovation cycle. It is expected that by adapting and further developing the technology with the help of digitisation, new optimisation potential in agriculture can be identified and used8. This includes automation using drones and robots. If you believe forecasts by experts, many swarm-based robots will do the field work of tomorrow in the future, which will further relieve farmers. An overall and sustainable change in agriculture will therefore be promoted. Automation is expected to reduce the size of future machines and tasks will be distributed9/10/11. However, for the practical use of these machines and robots, the necessary technology and infrastructure must first be available. The success of digitisation depends heavily on the actuators used. Because only if information gained through digitisation can be suitably converted into activities and actions can a benefit be generated for the farmer. Currently, the increase in the resulting data is exponential. Artificial Intelligence (AI) applications require a doubled computing requirement every 3.5 months on average12. This also increases the costs for data analysis, data management and the necessary digital infrastructure. This generates a lot of data that is never used to increase productivity. Even if intelligent algorithms generate added value from the data, this knowledge must be put into practice. This is only possible with suitable automation of the processes. Therefore, the success of digitisation and AI is directly related to robotics and automation. New information obtained through digitisation should be automated, accurate, energy-efficient and user-friendly in actions and recommendations for action can be transferred. Therefore, the requirement arises for independently operating robots and machines to take as many decision steps as possible for the user, or at least to support the user in doing so. The mechanics must be able to convert the finely resolved data and information practically in order to be able to control sections or even perform an individual plant processing. The future lies in the direct, local application of sensor data, which is supported by cloud-based information. 8 BMEL, “Digitalisierung in der Landwirtschaft Chancen nutzen - Risiken minimieren”, 2019. 9 S. M. Pedersen, S. Fountas, H. Have and B. S. Blackmore, “Agricultural robots - system analysis and economic feasibili- ty” Precis. Agric., vol. 7, no. 4, pp. 295–308, Jul. 2006. 10 3 B S. Blackmore, HW Griepentrog and S. Fountas, “Autonomous Systems for European Agriculture” in Automation Technology for Off-Road Equipment, 2006. 11 T Duckett et al., “Agricultural Robotics: The Future of Robotic Agriculture” UK-RAS White Pap., pp. 36, 2018 12 F Müssig, “AMD- CEO Lisa Su schaut in die Halbleiter-Zukunft” heise online, 2019. [Online]. Available: https://www.heise.de/newsticker/meldung/AMD-CEO-Lisa-Su-schaut-in-die-Halbleiter-Zukunft-4500699.html. [Accessed: 04-Sep2019] 5
