The objectives of this research were to collect and parse CAN, topography, timing, and path data for forage operations leading to individual cylindrical round bales with highest possible accuracy for subsequent analysis, to analyze fuel consumption rate and engine load data to ascertain loading differences which could reflect baler consumption rate, and to determine the possible resolution extent of forage yield maps generated without load cells or specific baler instrumentation.

An ISOBlue unit (open-source telematics) was used to record CAN messages pertaining to baling of large round bales which included time, Global Positioning System (GPS) Coordinates (hence speed), fuel consumption rate, engine load, and engine speed. The change in power requirement (a combination of PTO load, rolling resistance, and elevation change) was used to infer the rate of forage accumulation across the field. Post-processing of the data also included analysis of elevation data which was obtained from the CKT Tracker App.

The data was processed using Matrix Laboratory (MATLAB), ArcMap, and MS Excel. MATLAB was used for operations such as data alignment, calculation of slope between elevation points, and calculation of change in power requirement. ArcMap was used for operations such as locating each bale, splitting the complete dataset into a dataset for each bale, and joining the baling, no-PTO load, and elevation datasets together based on spatial location. Of the ten test bales produced, an increase in power requirement due to increased accumulation of mass in the baling chamber was only visible for three of the bales. Within the datasets of these three bales, the increase in power requirement was not consistent enough to determine the yield of the crop across the field, given the data collected.