FL2 – How to Augment a Tractor with a “Mind”?

Lecturer: Felix Hülsmann
Fields: Computer Science, Robotics, Systems Engineering

Content

The need to feed a growing global population raises challenges in all areas of agriculture. Aggravating, the shortage of skilled labor hits farmers not only in Europe, but all over the world. Enabling the farmers to produce more food with less work force (and with reduced negative impact on the environment) is a central challenge for the agricultural industry. It is evident, that autonomous machinery is an inevitable building block to solve this challenge.

Let us imagine the following north star: In the evening, during spring, the farmer sits in the office and requests the online farm management system to pre-plan the tillage operation for the next day. The system creates, for each field on the farm that needs to be processed, all steps which the tractor should do during the next day: An ideal driving path through the field is planned, initial values for the actuation of the tractor’s implement are calculated and of course the required time and fuel are estimated. The next day, the farm manager just drives the tractor to the field, presses “play” and collects the tractor in the evening. In the meantime, planning for the next day, documentation work and other office tasks can be done. How can we make this dream becoming reality?

Autonomous machinery is – at least if it shall be used in a productive way in large units – more than just equipping existing systems with cameras and “AI”. It is first about recognizing the needs of farmers all over the world, who work in very different conditions. And it is about understanding the actual work process (e.g., tillage) and how the typical work split is between the tractor and its implement (e.g., the cultivator). Then it is about understanding how existing tractors could benefit from available technology. Technology, which is already well-known in universities, in robotics companies and partly already in the automotive industry. Even when these pre-conditions have been fulfilled, we cannot just start with plugging components together and writing the code: we need to carefully investigate the required capabilities of our system and derive a solid architecture that is able to meet
– the farmers‘ needs
– the need of our own developers and service personnel
– required levels of safety
– state-of-the-art security
– lots of further legislation and norms
and of course, tight restrictions in terms of budget and available time. Only then, we can start with hardware development, software development and of course testing. If this has all been successfully finished, we can start with first pre-series production and tests directly at the farms.

In this talk, I would like to guide you through the development of an autonomous tractor that shall be usable in a legal way in Europe. During this trip, I will give you insights into the
Systems Engineering approach. Through the whole talk, we will use the extension of a standard tractor with high-level autonomy as an example. After the talk, you will have a first impression of one of the standard approaches in the development of large technical systems, for instance in agricultural, but also in the automotive, aviation and lots of further industries. And ideally, you directly want to start developing the machinery of tomorrow.

Literature

  • INCOSE (Ed.). (2023). INCOSE systems engineering handbook. John Wiley & Sons.
  • DIN EN ISO 25119. Tractors and machinery for agriculture and forestry – Safety-related parts of control systems (ISO/FDIS 25119-:2018)
  • DIN EN ISO 18497:2018. Agricultural machinery and tractors — Safety of highly automated agricultural machines — Principles for design
  • Regulation (EU) 2024/2847 of the European Parliament and of the Council of 23 October 2024 on horizontal cybersecurity requirements for products with digital elements and amending Regulations (EU) No 168/2013 and (EU) No 2019/1020 and Directive (EU) 2020/1828 (Cyber Resilience Act)
  • DIN EN ISO 24882 (under development). Agricultural Machinery and Tractors — Cybersecurity Engineering

Lecturer

Felix has a background in Computer Science and received his PhD from Bielefeld University. In 2019 he joined CLAAS, a global manufacturer of agricultural machinery such as tractors and harvesters. There, he started as software developer for embedded Linux. Three years later, Felix became System Architect. He mainly works on the development of autonomous and highly automated tractors. You can find some further personal information here: https://www.felix-huelsmann.de

Affiliation: CLAAS E-Systems GmbH
Homepage: https://www.claas.com/