On the go organic matter sensor and associated systems and methods

Abstract

An on-the-go planting system utilizing soil organic matter data to prescribed seeding rate, hybrid/variety selection or soil treatment. The system weights at least two field attribute data types, such as soil organic matter data and normalized yield data to prescribe one or more outcomes. The system can also generate a yield potential value so as to prescribe the outcomes.

Claims

1. An on-the-go planting system comprising: (a) a soil sensor configured to assess a soil quality; and (b) a processor configured to weight two or more field attributes to determine a yield potential, the field attributes comprising: (i) the soil quality; (ii) normalized yield data; (iii) terrain data; and (iv) geo-spatial data, wherein in the system is configured to prescribe one or more of a seed variety, a seeding rate, and a soil treatment based on the yield potential.

2. The on-the-go planting system of claim 1, wherein the soil quality is soil organic matter data.

3. The on-the-go planting system of claim 1, wherein terrain data comprises one or more of elevation data, soil type, drainage intensity, landscape positions, slope, grade, and wetness potential.

4. The on-the-go planting system of claim 1, wherein the soil sensor is an optical sensor.

5. The on-the-go planting system of claim 1, wherein the soil sensor is disposed on a planter row unit.

6. The on-the-go planting system of claim 1, further comprising a display housing the processor configured to display field attributes to a user.

7. The on-the-go planting system of claim 1, wherein the processor comprises an artificial intelligence for weighting.

8. The on-the-go planting system of claim 1, further comprising an artificial intelligence executed on the processor.

9. The on-the-go planting system of claim 1, wherein weights for the two or more field attributes are automatically adjusted by the system.

10. A method for prescribing a planting prescription comprising: sensing organic matter data in an open trench via a soil sensor; weighting sensed organic matter data with at least one field attribute, field attributes comprising: normalized yield data, terrain data, geo-spatial data, and user input data; determining a yield potential from the weighted sensed organic matter data and at least one field attribute; and outputting a prescription including one or more of seed variety, variable seeding rate, and soil treatment.

11. The method of claim 10, further comprising displaying on a display a map depicting organic matter data and the prescription.

12. The method of claim 10, further comprising an artificial intelligence executed on a processor for weighting sensed organic matter data with at least one field attribute.

13. The method of claim 10, further comprising an artificial intelligence executed on a processor for determining a yield potential from the weighted sensed organic matter data and at least one field attribute.

14. The method of claim 10, wherein the at least one field attribute is selected from normalized yield data, terrain data, geo-spatial data, and user input data.

15. The method of claim 10, wherein the at least one field attribute is terrain data.

16. The method of claim 15, wherein terrain data comprises one or more of elevation data, soil type, drainage intensity, landscape positions, slope, grade, and wetness potential.

17. The method of claim 10, further comprising assigning weights to the sensed organic matter data and the at least one field attribute.

18. The method of claim 17, wherein weights assigned to the sensed organic matter data and the at least one field attribute are adjusted automatically.

19. An on-the-go prescription planting system comprising: (a) a soil sensor configured to detect organic matter in a seed trench; and (b) a processor in communication with the soil sensor, the processor configured to: (i) assign a weight to and weight organic matter data from the soil sensor and terrain data; ii determine a yield potential based on the weighted weight organic matter data and terrain data; (iii) prescribe a seeding rate; and (iv) adjust a seed meter to apply the prescribed seeding rate.

20. The on-the-go prescription planting system of claim 19, wherein the weights assigned to the organic matter data and terrain data are adjusted via machine learning.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 is a top view of a tractor and planter having row units according to an exemplary implementation of the system.

(2) FIG. 2 is a three-quarters perspective view of a planter with row units, according to an exemplary implementation of the system.

(3) FIG. 3A is a side view of a tractor and planter with a row unit according to an exemplary implementation of the system.

(4) FIG. 3B is a three-quarters perspective view of the row unit of FIG. 3A showing a detailed view of various components of the system.

(5) FIG. 4A is a flow chart showing various optional steps of the system in use, according to an exemplary implementation of the system.

(6) FIG. 4B is a table showing system weighting, according to an exemplary implementation of the system.

(7) FIG. 4C is a table showing weighting calculations according to the implementation of FIG. 4B.

(8) FIG. 4D is a table showing a prescribed seeding rate according to the weighted calculations of FIG. 4C.

(9) FIG. 5 is a rendered depiction of an overlay map, according to an exemplary implementation of the system.

(10) FIG. 6 is a graphical representation of the value of weighting field attribute data via certain implementations of the system.

(11) FIG. 7 is a flow chart showing the system having machine learning, according to an exemplary implementation of the system.

DETAILED DESCRIPTION

(12) The various implementations disclosed herein relate to technologies for real time adjustments to variable seeding rates, hybrid selection, and/or soil treatment during agricultural processes, such as planting. In certain implementations, the system is an on-the-go, active, multi-factor variable rate prescription system. In various implementations, the system is used in conjunction with both a soil sensor that may be constructed and arranged to assess soil qualities such as organic matter and another augmenting data type, such as normalized yield to adjust the rate of seeding, hybrid selection and/or soil treatment. In various implementations, the organic matter measurements and augmenting data values are weighted to prescribe the seeding rate on-the-go.

(13) Accordingly, the implementations disclosed herein relate to systems, methods and devices for real-time monitoring of soil organic matter and other augmenting data for the prescription of variable rate seeding and/or variation of alternative planting parameters, or, for brevity, an on-the-go planting system or system. While the term system may be used, it is in no way intended to limit the scope of the disclosed implementations.

(14) The various implementations of the system disclosed herein can be incorporated into and/or used with various other known devices, systems, and methods. For example, various implementations of the system may be incorporated with the disclosures of U.S. patent application Ser. No. 16/523,343, filed on Jul. 26, 2019, entitled Closing Wheel Downforce Adjustment Devices, Systems, And Methods, U.S. patent application Ser. No. 16/142,522, filed on Sep. 26, 2018, entitled Planter Downforce and Uplift Monitoring and Control Feedback Devices, Systems, and Associated Methods, which are hereby incorporated by reference in their entirety.

(15) Turning to the figures in greater detail, FIGS. 1-3 depict various implementations of the augmented on-the-go system 10. FIG. 1 shows an exemplary tractor 12 with an attached planter 14; FIG. 2 shows another exemplary planter 14; and FIG. 3 shows an exemplary row unit 16, each for use with the disclosed system 10. In various implementations, the planter 14 includes a plurality of row units 16, as has been previously described. It is readily appreciated that the system 10 according to various implementations can be incorporated into any known row unit 16 having any configuration, and that any such row unit 16 can be incorporated into any known seeding machine.

(16) In certain implementations like those depicted in FIGS. 1-3B, the system 10 includes one or more liquid reservoirs 18. In various of these implementations, the liquid reservoirs 18 store various chemicals, such as fertilizer, for application to the soil 2 during planting. In further implementations, the system 10 includes one or more seed hoppers 20. In various implementations, the liquid reservoir(s) 18 and/or seed hopper(s) 20 are disposed on the tractor 12, planter 14, and/or row units 16 and constructed and arranged to plant seed and/or apply treatments such as pesticide, herbicide or the like at prescribed rates, as would be readily appreciated.

(17) As shown best in FIGS. 3A-3B, the row units 16 of various implementations may include a variety of features, including but not limited to: a supplemental downforce system 15, liquid reservoirs 18 and/or seed hopper 20, opening discs 22, gauge wheels 24, closing wheels 26, row cleaners 28 and the like. In some implementations, each row unit 16 includes both a liquid reservoir 18 and a seed reservoir 20. As would be understood, the various features and components of the row units 16 may be present in myriad configurations.

(18) Continuing with FIGS. 3A-3B, exemplary implementations of the system 10 in certain implementations can operate with a variable rate planting system 25 having a seed tube 25A and seed meter 25B, wherein the seed meter 25B can operate to quickly and easily adjust the seed cell position, geometry, number of the seed meter seed cells in real-time, including in the middle of the field without having to stop, such as SeedCommand/SureDrive or other products known in the art. It is appreciated that various of these variable rate changes are made via hydraulically or electronically altering the RPM of the seed meter 25B.

(19) Exemplary implementations of the system 10 also include a soil sensor 30, such as an organic matter soil sensor 30. The soil sensor 30 may be an optical sensor or other sensor type as would be known and appreciated by those of skill in the art. In some implementations, the soil sensor 30 is disposed on the tractor 12, as shown in FIG. 1. In alternative implementations, the soil sensor 30 is disposed on the planter 14, as shown in FIG. 2.

(20) In exemplary implementations, the soil sensor 30 is disposed on one or more of the row units 16, shown for example in FIGS. 3A-3B, such that the soil sensor 30 is able to take readings of organic matter in the opened trench 2. In various of these implementations, the soil sensor 30 is placed adjacent to the seed tube 25A, as would be readily appreciated.

(21) The soil sensor 30 according to these and other implementations is in communication with a processor 32. In some implementations, the processor 32 is in direct and/or indirect wired or wireless communication with storage and/or computing media 40. As shown in FIG. 1, the storage 40 may be accessed via a cloud platform 34 or server 34 and/or storage 40 can be achieved locally, such as with an on-board storage medium other or database 40. Various alternative storage and computing media would be known and recognized by those of skill in the art. The storage media 34/40, processor 32, and/or soil sensor 30 are in direct and/or indirect wired or wireless communication with a display 36 or displays 36 (described elsewhere herein in reference to a single display 36).

(22) In certain implementations, the display 36 is disposed within a tractor 12 for use during planting operations. In various alternative implementations, the display 36 is remote of the tractor 12 and is controlled and operated off-site during planting operations. In still further implementations, the display 36 is portable and is constructed and arranged to operate both from within the tractor 12 and outside of the tractor 12. A display 36 may be used in both planting and harvesting operations, as would be readily appreciated by those of skill in the art. According to certain implementations, the display 36A may be a secondary display 36A, or may also be a mobile device such as an iPad or other mobile device 36A, and in certain implementations an ISOBUS Universal Terminal (shown in FIG. 1 at 38) is further provided such that various components of the system 10 from the OEM or other manufacturers are able to communicate with the system 10 processors 32, displays 36 and the like.

(23) In these and other implementations, the display 36 can display the seeding rate and the organic matter readings to the user/operator in real time. It is understood that various factors can affect the ideal seeding rate and/or other planting parameters. These factors may include the amount of organic matter in the soil, historic yield, terrain characteristics, geo-spatial data, and various additional factors as would be recognized by those of skill in the art. Utilizing variable rate seeding can result in high yields while minimizing waste, ultimately increasing profits.

(24) However, previous attempts at using organic matter have resulted in inconsistent results. The implementations of the system 10 disclosed herein improve the performance of variable rate seeding applications by weighing organic matter readings with other augmented data to establish a weighted yield potential reading that can, in turn, be used to prescribe a specified seeding, hybrid selection and/or soil treatment amount.

(25) Returning to the drawings, in implementations like that of FIG. 4A, the system 10 is constructed and arranged to collect and weight two or more field attribute data types 100 to determine yield potential (box 120). In the implementation of FIG. 4A, the field attribute data 100 includes organic matter data (box 110) that is weighed with one or more additional augmenting field attribute data 100, such as normalized yield data (box 112), terrain data (box 114), geo-spatial data (box 116) and/or user input data (box 125). In alternate implementations, organic matter data (box 110) is not collected, but two or more of the other field attribute data types (boxes 112-116 and 125) are weighted (box 118).

(26) Continuing with the implementations of FIG. 4A, the system 10 weights (box 118) the two or more field attribute data inputs (boxes 110-116 and 125) in order to determine the yield potential (box 120) on-the-go and prescribe an output 130, such as one or more of the seed variety (box 121), the variable rate seeding rate (box 122), a soil treatment (box 124) and/or other treatment parameters according to the determined yield potential (box 120).

(27) That is, the system 10 according to these implementations is constructed and arranged to augment a first field attribute data 100, such as organic matter data (box 110), with another layer of field attribute data via a variety of optional steps, such that the system 10 can prescribe 130 a planting/treatment (boxes 121-124) so as to maximize the actual yield of an area is by adjusting the seeding rate, hybrid selection and/or soil treatment rate in real time.

(28) More specifically, the system 10 according to the implementation of FIG. 4A is constructed and arranged to collect organic matter data (box 110) from a soil sensor 30 mounted on the row unit (such as is shown in FIGS. 3A-3B) and augment that organic matter data with one or more of various additional field attribute data 100 sources (boxes 112-116). The additional field attribute data sources shown in FIG. 4A include normalized yield data (box 112), terrain data (box 114) and geo-spatial data (box 116), though there are others that would be readily appreciated by those of skill in the art, such as historical data. In further implementations, the additional field attribute data sources may be in the form of spatial data layers.

(29) In use according to one such implementation, in an optional field attribute data step, the system 10 collects/utilizes organic matter data (box 110) from the soil sensor 30. In another optional field attribute data step, the system 10 collects/utilizes normalized yield data or other yield data (box 112). In a further optional field attribute data step, the system 10 collects/utilizes terrain data (box 114), such as elevation data. In another optional field attribute data step, the system 10 collects/utilizes geo-spatial data (box 116). In a further optional field attribute data step, the system 10 collects/utilizes user inputs (box 125). In various implementations, the organic matter data (box 110) or any of the other optional augmenting field attribute data steps are omitted. It is readily appreciated that the collection/utilization each of the optional steps described herein can be performed in any order.

(30) In use according to the implementation of FIGS. 4A-4D, the collected field attribute data, such as optional organic matter data (box 110) and normalized yield data (box 112) are weighed (box 118) by the system 10 to determine yield potential (box 120). In various implementations, both the weighting (box 118) and the determination of yield potential (box 120) are done on the basis of user-inputted weighing values, while in other implementations the weighing is done via an artificial intelligence or machine learning protocol, as discussed in relation to FIG. 7.

(31) Continuing with the implementation of FIG. 4A, these data inputs may be replicated and displayed on the display 36. In various implementations, the various data inputs are displayed on the display 36 during planting operations. In these implementations, an algorithm executed on the processor 32 processes the various field attribute data inputs. That is, the processor 32 executes one or more algorithms to augment the organic matter data with the various other desired data inputs. In further implementations, the system algorithm(s) determine the yield potential for the soil being planted in real time for display 36.

(32) Following the determination of the yield potential (box 120), the system 10 prescribes 130 one or more actions, such as the seed variety (box 121), the seeding rate (box 122) and/or any soil treatments (box 124).

(33) As shown in FIGS. 4B-4D, in various implementations, the system 10 is constructed and arranged to weight (box 118 in FIG. 4A) two or more types of field attribute data. In various of these implementations, the system 10 applies various weighting zones to the utilized field attribute data types.

(34) In the example of FIG. 4B, the system 10 assigns an area having an organic matter (OM) percentage between 0 and 1%, a normalized yield between 0 and 80% or a terrain analysis of depression a yield potential value of 1. In this specific example, the seeding rate for yield potential 1 is 28,000 seeds/acre, as is shown in FIG. 4D. It is appreciated that these yield potential values represent values for an area based on a single field attribute data type, prior to weighting, as described herein.

(35) In another example, the system 10 defines assigns an area with an OM % between 1 and 2%, a normalized yield between 70 and 100%, or the terrain analysis is of ridge as yield potential value 2. As shown in FIG. 4B, additional attributes (1 and N) can also be included categorized (box 118 in FIG. 4A).

(36) In the implementation of FIGS. 4B-D, yield potential value 3 is assign by any of OM %: 3-4%, normalized yield: 100-110% or terrain: plain, seeding rate: 36,000. Yield potential value 4 is characterized by OM %: 4+%, normalized yield 100+% or terrain: side slope, for a seeding rate of 40,000.

(37) Critically, and as will be readily appreciated and as shown in FIG. 4C, any given specific area of the field (Area 1) can have multiple field attributes and data that fall into several yield potential values. For example, taking several different field attribute data points together, an area may have an OM % yield potential value of 3, a normalized yield value of 4, a terrain value of 4, and two other field attribute data values of 2 (shown at Attribute 1 and Attribute Nit is understood that these are included for illustrative purposes and can represent any types of field attribute data that are to be included).

(38) In these implementations, the system 10 according to certain implementations assigns different weights W to these different field attribute data inputs or parameters to determine weighted attribute scores, shown in FIGS. 4B-4C at X. The weights W may be adjusted manually or automatically by the system 10 and/or user, as would be appreciated.

(39) As shown in FIG. 4C, as a result of the weighs W, the weighted attribute scores X can be tabulated to calculate a weighted yield potential Y, which is directly correlated to the on-the-go yield potential (box 120 in FIG. 4A). As shown in FIG. 4D, the weighted yield potential Y (3.5625 out of a possible 4) can then be prescribed via the output table to determine an on-the-go augmented seeding rate Z: 38,250 seeds/acre. It is readily appreciated that any of the other prescribed outputs 130 of FIG. 4A or those understood in the art can be determined in the same way.

(40) In a further implementation, the system 10 is constructed and arranged to include or exclude field attribute data types during use. In various alternative implementations, one or more field attribute data types may not be measured and or included in the system 10 processing. In some implementations, a user may manually choose to exclude or include a specific field attribute data type, such as by temporarily suspending or activating the field attribute data, such as through the display 36. In various alternative implementations, the system 10 may automatically exclude or include a field attribute data type, such as by detecting a malfunctioning sensor and excluding the field attribute data type measured by that sensor. Many other examples will be readily appreciated by the skilled artisan.

(41) Normalized Yield Data

(42) Turning to the various optional field attribute data types in detail, in certain implementations, the system 10 augments organic matter data (box 110) with normalized yield data (box 112). In some implementations, normalized yield data is historical yield data for the field over multiple years, which has been normalized into one data set. As would be appreciated, historical yield data can be an indicator of future yield potential.

(43) FIG. 5 shows an exemplary organic matter map 50 and a corresponding normalized yield map 52. In various implementations, the system 10 is constructed and arranged to augment the organic matter map 50 with the normalized yield map 52 (box 112 of FIG. 4A) and created an overlaid map 54. It is appreciated that in the organic matter map 50, lighter areas indicate high organic matter content while darker areas indicate lower organic matter content. In the normalized yield map 52, the darker area indicate higher yields while lighter areas indicate lower yields. As can be seen by comparing the maps 50, 52 the amount of organic matter does not always directly correlate to the yield value (where it is generally understood that higher organic matter indicates a higher yield). As would be readily appreciated, alternate implementations are of course possible.

(44) As would be further appreciated, another factor impacting yield can be weather. Including and weighting historical yield data, reflected in the normalized yield map 52, takes into account year-to-year weather variability. As such, in various implementations, the system 10 can use the normalized yield data to determine the yield potential of various areas of the field.

(45) Terrain Data

(46) In various implementations, the system 10 augments the organic matter data with terrain data (box 114). In some implementations, terrain data includes elevation data. In further implementations, the terrain data includes other terrain analysis such as soil type, drainage intensity, landscape position (depression, valley, side slope, ridge, plain), slope, grade, wetness potential. In some implementations, terrain data may be utilized by the system 10 to recognize areas of field or terrain that are prone to ponding or poor draining leading to chronic wetness. In these implementations, augmenting organic matter data with terrain data (box 114) allows the system 10 to prescribe a seeding rate that corresponds to the yield potential of the soil. In further implementation, the system 10 can prescribe or alter other planting parameters corresponding to the yield potential of the soil.

(47) In various implementations, the augmenting the soil sensor data with terrain data prevents the system 10 from incorrectly identifying chronically wet areas as high yield zones, despite the presence of high organic matter. For example, organic matter data when augmented with terrain data (box 114) may allow the system 10 to recognize areas that have high organic matter contentindicative of high yield potentialbut are chronically wet and/or drain poorly such that the yield potential is lower.

(48) The system 10 may evaluate and detect terrain characteristics via a variety of techniques. In some implementations, the system 10 uses a prerecorded spatial map of elevation and/or slope. In certain implementations, the system 10 uses a tilt or inertial sensor affixed to the tractor 12, planter 14, and/or row units 16 during planting operations. In some implementations, the measured tilt of the tractor 12, planter 14 and/or row unit 16 is correlated to the terrain slope. In further implementations, the system 10 includes a GPS system. In some implementations, the GPS system is affixed to the tractor 12, planter 14, or row unit 16 and terrain slope is calculated using the change in GPS elevation along the tractor 12, planter 14, or row unit 16 path.

(49) In some implementations, a combination of techniques for determining terrain data are used. In one implementation, the system 10 uses both a GPS system and an inertial sensor on the tractor 12 and/or planter 14. In these implementations, the system 10 utilizes GPS elevation data and a tilt or inertial sensor in combination to determine where the system 10 is located in a fieldon the top of a hill, on the side slope, in a depression at the bottom of the hill or the like. Of course, other techniques for measuring terrain characteristics are possible and would be recognized by those of skill in the art.

(50) By way of example, FIG. 6 shows a side view of an exemplary hill 3 or field area with varying elevation. In this example, at the top of the hill 3A, the system 10 detects the higher elevation and a low level of organic matter. As a tractor 12 or planter 14 traverses the hill 3 the system 10 detects the changes in elevation and organic matter content. For example, the bottom of the hill 3B has a lower elevation and higher level of organic matter than the top of the hill 3A. When the organic matter data is viewed on its own the bottom of the hill 3B has the highest yield potential due to its high level of organic matter. But the system 10 is constructed and arranged to overlay the organic matter data with the terrain data (box 114 of FIG. 4A)here elevation datato recognize area 3B as being the bottom of a hill that is prone to pooling water and/or ponding thereby decreasing the yield potential for area 3B.

(51) In the example of FIG. 6, the area 3C just above the bottom of the hill 3B and downhill of the top of the hill 3A has a medium organic matter level and a favorable slope for water flow, and therefore has the highest yield potential. As would be recognized by those of skill in the art, despite having a lower level of organic matter than the bottom of the hill 3B, the field at area 3C does not have the ponding/pooling issue that the bottom of the hill 3B has, and therefore has a greater yield potential, and so it is preferable to prescribe yield-improving outputs to that area 3C.

(52) Continuing with the illustrative example of FIG. 6, the top of the hill 3A does not have the pooling or ponding issues that the bottom on the hill 3B has because water flows downhill. Yet, due to the steeper slope of the hill the top soil may erode away or flow downhill thereby decreasing the amount of organic matter in the soil, which decreases the yield potential of the area 3A.

(53) Geo-Spatial Data

(54) Turning back to FIG. 4A, in various implementations, the system 10 utilizes geo-spatial data (box 116) as one of the two or more field attribute datum. In various implementations, this geo-spatial data may include the location of field drainage tile lines and other physical characteristics well understood in the art. The system 10 augmenting the organic matter data with such geo-spatial data (box 116) may indicate area of the field with higher or lower yield potentials. For example, areas close to the drainage tiles have a higher yield potential than areas farther away from the drainage tiles. As would be readily appreciated, such geo-spatial data may be integrated via a map overlay, as shown in relation to FIG. 5.

(55) User Inputs

(56) In various implementations, the system 10 augments the organic matter data with various user inputs (box 125). In some implementations, user inputs may include a user defined map of low and high yield potential areas. For example, a user may manually mark and/or otherwise identify yield potentials for various locations in a field.

(57) Turning back to FIG. 4A, the system 10 according to certain implementations is constructed and arranged to weight (box 118) and determine yield potential (box 120) in real time. That is, the system 10 is constructed and arranged to use the field attribute data inputs (boxes 110-116) to determine the yield potential for the current location of the tractor 12, planter 14, row unit 16, or other implement in real time during planting operations.

(58) In another optional step, the system 10 can use the yield potential (box 120) to prescribe a seeding rate (box 112) for that zone. In a further implementation, the system 10 automatically adjusts the seeding rate to the prescribed seeding rate in real time, as has been previously described. In further implementations, the system 10 alerts a user to the prescribed seeding rate and the user may then manually adjust the seeding rate.

(59) In a further optional step, the system 10 can prescribe a soil treatment (box 124) or other adjustment to planting parameters based on either or both of the yield potential (120) and the prescribed seeding rate (box 122). As noted above, adjustments or changes may be made manually or automatically be the system 10. That is, the algorithm may establish correlated relationships between selected hybrid (box 121) and the seeding rate (box 122). For example, for a yield potential of 3.5 or higher, a specific hybrid may be selected, while an alternate hybrid is selected for yield potentials below 3.5, and the seeding rates (box 122) and/or treatment regimens (box 124) may vary depending on the selected hybrid. While this example presumes dependency on the selected hybrid, the relationship between the various selections/rates can be independent or operate in alternate order, such that the seeding rate (box 122) can be used in prescribing the hybrid (box 121) and treatment (box 124), or the treatment (box 124) can be used in prescribing the seeding rate (box 122) and hybrid (box 121). Many examples are of course possible and would be readily appreciated by those of skill in the art.

(60) In a still further optional step, the system 10 can use the yield potential (box 120), prescribed hybrid (box 121), prescribed seeding rate (box 122), and/or prescribed soil treatments (box 124) to select and locate population trials (box 126). That is the system 10 can automatically determine where to locate a population trial (box 126) based on the various inputs to the system 10 shown in boxes 110-116). For example, the system 10 may determine that a specified area or zone has a highly consistent (or inconsistent) yield potential, and determine that area zone is thus suitable for a population trial.

(61) As would be understood, population trials are areas, often small areas, in a field where different seeding rates are planted and analyzed during harvest for yield and/or other parameters of interest. In some implementations, these population trials are replicated in each zone of the field.

(62) In some of these implementations, the system 10 stores the locations of the population trials in the storage media 34. During harvest the storage media 34 can be accessed and the population trial areas can be displayed on the display 36. In various of these implementations, during harvest the yield for each population trial can be checked. In some implementations, the system 10 is constructed and arranged to compare the harvest yields of the various population trials to determine the best seeding rate and/or treatment parameters for each zone.

(63) In certain implementations, the system 10 can display the population trial yield information to a user during planting, harvesting, or any other time such that adjustments in seeding rates, treatments, seed type, and/or other parameters can be made during subsequent plantings and growing seasons.

(64) As shown in FIG. 7, certain implementations of the system 10 feature machine learning that comprise the implementation of a variety of optional steps. In these implementations, the system 10 processor 32 is in electrical communication with a storage medium, such as a database 40 and, optionally, a display 36, components through which the system 10 is constructed and arranged to execute a variety of optional machine learning algorithm steps (shown generally at box 140).

(65) In implementations like that of FIG. 7, the system 10 (via the processor 32 and storage 40) is optionally constructed and arranged to store field attribute data (box 100) for individual locations, optionally so as to be combined with actualized yield data (box 150) recorded by the harvester so as to perform pattern recognition, artificial intelligence, neural net technology, and/or machine learning to establish yield potential performance thresholds and predict future yield potentials and/or weighting parameters for use in subsequent planting cycles. The machine learning data may in turn be stored for access by the processor 32 and subsequently be optionally used to, for example, optionally alter the weighting of the field attributes (box 118), optionally calculate of yield potentials (box 120), and/or optionally revise the prescribed outputs (box 130) for defined yield potentials, that is, adjust stored tables like those shown in FIGS. 4B-4D. Additional implementations also monitor additional parameters such as historical weather data and the like for the application of statistical analysis, machine-learning and the like.

(66) Further implementations of the system featuring machine learning are constructed and arranged to store the locations of the population trials (box 126) in the storage media 40. During harvest, the storage media 40 can be accessed and actualized yield harvest data (box 150) for the population trials (box 126) are compared with at least one of the field attribute data (box 100), the weighted field attribute data (box 118), the determined yield potential (box 120), the prescribed outputs (box 130) and/or other factors such as user inputs (not shown) and the like. In various of these implementations, during harvest the yield for each population trial can be analyzed via machine learning (box 140) to optimize the various components of the system 10. In some implementations, the system 10 is thereby constructed and arranged to compare the actualized harvest yields of the various location trials to determine the best prescribed outputs. Further implementations of the machine learning processes are of course possible and would be readily appreciated by those of skill in the art.

(67) It is further appreciated by those of skill in the art that in implementations featuring machine learning and those without, the system 10 may display one or more of the calculated yield potential and/or prescribed rate/hybrid/treatment in the display 36 to be used during harvesting to compare calculated yield potential (box 120) to actualized yield (box 150). It is appreciated that in use according to these implementations, the user is thereby able to visualize which field attributes (box 100) have the largest impact on actualized yield in the specific field, such that the user can adjust the various weightings either manually or with the aid of machine learning. Further combinations and implementations are of course possible.

(68) Although the disclosure has been described with references to various embodiments, persons skilled in the art will recognized that changes may be made in form and detail without departing from the spirit and scope of this disclosure.