Agricultural load cycles: Tractor mission profiles from recorded GNSS and CAN bus data
Korbinian Götz, Andrew Kusuma, Adrian Dörfler, Markus Lienkamp
Abstract
Agriculture is one of the main sources of greenhouse gas emissions, with agricultural machinery playing an important role through the combustion of fossil fuels. On the upside, many farmers invest in their own renewable energy generation. To leverage the renewable produced energy and cut emissions, a promising solution is the electrification of tractors. With the advent of tractor electrification, the necessity to define the requirements of an electrified tractor emerges. In contrast to automotive and truck engineering, where the usage patterns are narrow, and testing is standardized through load cycles as the WLTP and VECTO LongHaul, the agricultural application of tractors includes a wide variability of operating conditions. Since there are no standardized, publicly available agricultural load cycles, we equipped, during a one-year period, five tractors (77-240 kW) on a farm in the South of Germany with high-resolution GNSS loggers and collected the CAN bus data from the J1939 engine bus and the ISOBUS. The data collected is enriched with information about the tractor models and implements used. Hence, the load and usage information in real-field operation is suitable as input for the development of electrified tractors. A total of 862 h of operation data has been recorded and constitutes one of the most comprehensive open datasets to date for high-resolution data of agricultural tractors. This dataset provides engine and load parameters, auxiliary power supply, and GNSS information for multiple different agricultural fields. Its applications include the simulation of electrification for agricultural tractors, load cycle construction, modeling of farming processes, the analysis of agricultural production indicators, or path planning for autonomy.