System and method for irrigation management using machine learning workflows
The present invention provides a system and method which includes a machine learning module which analyzes data collected from one or more sources such as UAVs, satellites, span mounted crop sensors, direct soil sensors and climate sensors. According to a further preferred embodiment, the machine learning module preferably creates sets of field objects from within a given field and uses the received data to create a predictive model for each defined field object based on detected characteristics from each field object within the field.
1. A system for use with a self-propelled irrigation system having at least one span and a drive system for moving the span across a field to be irrigated, the system comprising: span mounted sensors, wherein at least one span mounted sensor comprises at least one sensor configured to allow for detection of the rate of crop growth; climate sensors, wherein at least one climate sensor is configured to detect at least one climate condition, wherein the climate condition is selected from the group of climate conditions comprising: humidity, pressure, precipitation and temperature; aerial sensors, wherein the aerial sensors include at least one sensor located on an unmanned aerial vehicle, plane or satellite; and a machine learning module, wherein the machine learning module is configured to receive characteristic data for the field; wherein the machine learning module is configured to create a set of field objects for the field and use the characteristic data to create a predictive model for each defined field object based on the detected characteristic data for each field object within the field; wherein the machine learning module receives field measurements and dimensions; wherein the set of field objects are stored as annular sectors; wherein the annular sectors are formed as subsections of rings defined by an inner and outer circle with the shape preferably bounded by the difference in radial length, and an angle (0) derived from two radii connecting to the ends of an outer length L determined by the selected angle (0); wherein each annular sector is defined as having an area=(Ru 2 −Ri 2 )/2θ; wherein ⊖=L/r, Ru is the radius of the outer arc, Ri is the radius of the inner arc, r is the radius of the irrigable field, and L is the arc length of the outer circumference for the selected angle (⊖); wherein the characteristic data comprise data received from onboard sensor arrays; wherein the characteristic data comprise spectral bands generated off of the soil and the crop canopy; wherein the characteristic data comprises water chemistry data; wherein the characteristic data comprises data regarding the specifications of the irrigation system and its subcomponents; wherein the machine learning module further analyzes data regarding soil chemistry, water chemistry and yield data; wherein the characteristic data comprises: soil moisture by depth; soil moisture forecast in root zone; and soil moisture forecast by depth; wherein an annular sector comprises a discrete data point which is linked to characteristic data; wherein the machine learning module is configured to create a predictive module for a plurality of annular sectors over time and to create yield optimizing solution models for the plurality of annular sectors; wherein the characteristic data further comprises: the chemigation material amount ready for injection, and the base chemigation application amount per unit area.
2. A system for use with a self-propelled irrigation system having at least one span and a drive system for moving the span across a field to be irrigated, wherein the system comprises: span mounted sensors, wherein at least one span mounted sensor comprises at least one sensor configured to allow for detection of the rate of crop growth; climate sensors, wherein at least one climate sensor is configured to detect at least one climate condition, wherein the climate condition is selected from the group of climate conditions comprising: humidity, pressure, precipitation and temperature; aerial sensors, wherein the aerial sensors include at least one sensor located on an unmanned aerial vehicle, plane or satellite; and a machine learning module, wherein the machine learning module is configured to receive characteristic data for the field; wherein the machine learning module is configured to create a set of field objects for the field and use the characteristic data to create a predictive model for each defined field object based on the detected characteristic data for each field object within the field; wherein the machine learning module receives field measurements and dimensions; wherein the set of field objects are stored as annular sectors; wherein the annular sectors are formed as subsections of rings defined by an inner and outer circle with the shape preferably bounded by the difference in radial length, and an angle (⊖) derived from two radii connecting to the ends of an outer length L determined by the selected angle (⊖); wherien wherein each annular sector is defined as having an area=(Ru 2 −Ri 2 )/2θ; wherein ⊖=L/r, Ru is the radius of the outer arc, Ri is the radius of the inner arc, r is the radius of the irrigable field, and L is the arc length of the outer circumference for the selected angle (⊖); wherein the characteristic data comprise data received from onboard sensor arrays; wherein the characteristic data comprise spectral bands generated off of the soil and the crop canopy and NIRNRGB bands; wherein the characteristic data comprise data regarding the irrigation machine; wherein the characteristic data comprises water chemistry data; wherein the characteristic data comprises data regarding the specifications of the irrigation system and its subcomponents; wherein the machine learning module further analyzes data regarding soil chemistry, water chemistry and yield data; wherein the characteristic data comprises: soil moisture by depth; soil moisture forecast in root zone; and soil moisture forecast by depth; wherein an annular sector comprises a discrete data point which is linked to characteristic data; wherein the machine learning module is configured to create a predictive module for a plurality of annular sectors over time and to create yield optimizing solution models for the plurality of annular sectors; wherein the characteristic data further comprises: the chemigation material amount ready for injection, and the base chemigation application amount per unit area; wherein input data comprises positional data for a given time (P1); wherein the input data comprises torque application data from the drive system; wherein the torque application data comprises data indicating the amount of torque applied to a drive wheel over a given interval of time (T+1); wherein the system of the present invention is configured to calculate an expected position (PE) of the drive tower after the given interval of time; wherein the system is configured to directly detect the geo-location of the drive tower (P2) after the given time interval; wherein the system is configured to calculate a slip ratio based on a comparison between the expected position (PE) and the detected position (P2); wherein the system is configured to calculate an estimated ground moisture level based on the calculated slip ratio.