Initial operating settings optimization for agricultural harvester
A harvester operating settings initialization method and system receives, by electronic control architecture having a processor and memory, a plurality of aggregated data sets pertaining to multiple parameters including harvester machine parameters, harvester environment parameters, and crop parameters. The electronic processing architecture applies modeling logic to the plurality of aggregated data sets to determine a selected evaluation group from a plurality of evaluation groups each having multiple of the plurality of aggregated data sets. The electronic control architecture generates an optimization settings data set for machine and crop combinations utilizing the selected evaluation group for one or more geospatial locations. The optimization settings data set is transferred from the electronic control architecture to an operating setting controller of a harvester for initializing settings of operational systems of the harvester.
1 . A harvester operating settings initialization method comprising:
receiving, by electronic control architecture having a processor and memory, a plurality of aggregated data sets pertaining to multiple parameters including harvester machine parameters, harvester environment parameters, and crop parameters;
applying, by the electronic control architecture, modeling logic to the plurality of aggregated data sets to determine a selected evaluation group from a plurality of evaluation groups each having multiple of the plurality of aggregated data sets, wherein the modeling logic includes:
generating a plurality of data constructs associated with the plurality of evaluation groups by processing the plurality of aggregated data sets associated with the plurality of evaluation groups using one or more statistical modeling techniques; and
generating a plurality of scores for the plurality of evaluation groups by assessment of the plurality of data constructs according to one or more scoring pillars, wherein the selected evaluation group of the plurality of evaluation groups is determined based on the plurality of scores, and wherein the scoring pillars include a spread metric, a peak metric, and a generalization metric whereby:
the spread metric resolves a spread score that elevates the data constructs of the plurality of evaluation groups with aggregated data sets that vary by more than a minimum variance value and less than a maximum variance value,
the peak metric resolves a peak score that elevates the data constructs of the plurality of evaluation groups with aggregated data sets that have a quantity of identifiable prominent peaks less than a maximum peak quantity value, and
the generalization metric resolves a generalization score that elevates the data constructs of the plurality of evaluation groups with aggregated data sets that have a membership above a minimum membership value and less than a maximum membership value;
generating, by the electronic control architecture, an optimization settings data set for machine and crop combinations utilizing the selected evaluation group for one or more geospatial locations; and
transferring the optimization settings data set from the electronic control architecture to an operating setting controller of a harvester for initializing settings of operational systems of the harvester.
2 . The method of claim 1 , wherein each of the plurality of evaluation groups includes a crop-specific set of the plurality of aggregated data sets and a machine-specific set of the plurality of aggregated data sets.
3 . The method of claim 1 , wherein the statistical modeling techniques include kernel density estimation and peak detection.
4 . The method of claim 1 , wherein the parameters comprising the plurality of aggregated data sets include harvester machine data, operator behavior data, crop data, soil data, weather data, and field image data.
5 . The method of claim 4 , wherein applying the modeling logic includes receiving machine and crop inputs identifying a unique one of the machine and crop combinations and applying to the harvester the optimization settings data set corresponding to the identified machine and crop combination.
6 . The method of claim 5 , wherein the machine and crop inputs are acquired from onboard the harvester or from offboard the harvester, and wherein, when acquired from offboard the harvester, the modeling logic includes receiving a field input in terms of a target field input and a harvest date input, in which the method further includes:
building a field geometry associated with the target field input;
acquiring environment and soil inputs associated with the target field input and the harvest date input;
determining anticipated field data associated with the environment and soil inputs for the field geometry; and
incorporating the anticipated field data with the machine and crop inputs to identify the identified machine and crop combination.
7 . The method of claim 1 , wherein generating the optimization settings data set includes establishing a database of optimized settings data for multiple unique ones of the machine and crop combinations.
8 . The method of claim 7 , wherein the operational systems of the harvester include a cleaning fan, a thresher fan speed, a thresher, a chaffer, and a sieve.
9 . The method of claim 1 , wherein transferring the optimization settings data set from the electronic control architecture to the operating setting controller of a harvester includes transmitting the optimization settings data set to the harvester wirelessly from a remote facility during operation of the harvester.
10 . The method of claim 1 , further including displaying the optimization settings data set as a settings map on a display onboard the harvester.
11 . A harvester operating settings initialization system comprising a non-transitory machine-readable medium containing instructions to direct a processor to:
receive a plurality of aggregated data sets pertaining to multiple parameters including harvester machine parameters, harvester environment parameters, and crop parameters;
apply modeling logic to the plurality of aggregated data sets to determine a selected evaluation group from a plurality of evaluation groups each having multiple of the plurality of aggregated data sets, wherein the modeling logic includes:
generating a plurality of data constructs associated with the plurality of evaluation groups by processing the plurality of aggregated data sets associated with the plurality of evaluation groups using one or more statistical modeling techniques; and
generating a plurality of scores for the plurality of evaluation groups by assessment of the plurality of data constructs according to one or more scoring pillars, wherein the selected evaluation group of the plurality of evaluation groups is determined based on the plurality of scores, and wherein the scoring pillars include a spread metric, a peak metric, and a generalization metric whereby:
the spread metric resolves a spread score that elevates the data constructs of the evaluation groups with aggregated data sets that vary by more than a minimum variance value and less than a maximum variance value,
the peak metric resolves a peak score that elevates the data constructs of the evaluation groups with aggregated data sets that have a quantity of identifiable prominent peaks less than a maximum peak quantity value, and
the generalization metric resolves a generalization score that elevates the data constructs of the evaluation groups with aggregated data sets that have a membership above a minimum membership value and less than a maximum membership value;
generate an optimization settings data set for machine and crop combinations utilizing the selected evaluation group for one or more geospatial locations; and
initialize operational systems of the harvester according to the optimization settings data set.
12 . The system of claim 11 , wherein each of the plurality of evaluation groups includes a crop-specific set of the plurality of aggregated data sets and a machine-specific set of the plurality of aggregated data sets.
13 . The system of claim 11 , wherein the statistical modeling techniques include kernel density estimation and peak detection.
14 . The system of claim 11 , wherein the parameters comprising the plurality of aggregated data sets include harvester machine data, operator behavior data, crop data, soil data, weather data, and field image data; and
wherein the modeling logic receives machine and crop inputs that identify a unique one of the machine and crop combinations and applies to the harvester the optimization settings data set corresponding to the identified machine and crop combination.
15 . The system of claim 14 , wherein the machine and crop inputs are acquired from onboard the harvester or from offboard the harvester, and wherein, when acquired from offboard the harvester, the modeling logic includes receives a field input in terms of a target field input and a harvest date input, and the processor is directed to:
build a field geometry associated with the target field input;
acquire environment and soil inputs associated with the target field input and the harvest date input;
determine anticipated field data associated with the environment and soil inputs for the field geometry; and
incorporate the anticipated field data with the machine and crop inputs to identify the identified machine and crop combination.
16 . The system of claim 11 , wherein the optimization settings data set are generated as a database of multiple unique ones of the machine and crop combinations; and
wherein the operational systems of the harvester include a cleaning fan, a thresher fan, a thresher, a chaffer, and a sieve.
17 . The system of claim 11 , further including a display onboard the harvester displaying the optimization settings data set as a settings map.