Patent Application 18010419 - TURF MANAGEMENT SYSTEMS AND METHODS

Title: TURF MANAGEMENT SYSTEMS AND METHODS
Application Information

Invention Title: TURF MANAGEMENT SYSTEMS AND METHODS
Application Number: 18010419
Submission Date: 2025-05-16T00:00:00.000Z
Effective Filing Date: 2022-12-14T00:00:00.000Z
Filing Date: 2022-12-14T00:00:00.000Z
National Class: 700
National Sub-Class: 275000
Examiner Employee Number: 91601
Art Unit: 2116
Tech Center: 2100
Rejection Summary

102 Rejections: 0
103 Rejections: 2
Cited Patents

No patents were cited in this rejection.
Office Action Text


    Notice of Pre-AIA  or AIA  Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .

DETAILED ACTION
Priority
Acknowledgment is made of applicant's claim for domestic benefit based on a provisional application 63/044,016 filed on June 25, 2020.   

DETAILED ACTION
Claims 16 - 30 and 36 are pending in the application.
Claim 16 is independent. 
Claims 16 – 35 were subject to a restriction requirement; and given the election by the applicant, claims 31 – 35 are cancelled. 
Clam 36 is new. 
This action is first action non-final action.

	
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:

A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.


Claims 16 - 28 and 36 are rejected under 35 U.S.C. 103 as being unpatentable over Silva et al. (US PG Pub. No. 20210279867) herein “Silva,” in view of Bentwich (US PG Pub. No. 20160157446), herein “Bentwich.”


Regarding claim 16,
Silva teaches a method of managing turf (crop or plants in a crop) in a work region, (Par. 0012: “Knowing the field conditions and how representative plants are growing allows farmers to predict the yield for a crop and to make adjustments to thereby improve yields and make arrangements for the harvest of the crop ( e.g., determine silo space needed, plan for time of harvest, locate distributors, plan for next seeding).”  Examiner’s Note – Silva teaches a method of managing plants in a crop or fields for farms, but the same method could be used for a turf; and specification paragraph 0032 states that the turf may be subject to “cultivation” which is a process used on a farm or agriculture field that contains beans, corn or other tillable crops.) 
the method comprising: monitoring multiple sets of zone sensor data, (data collection points and sensors) each set associated with a different zone of a plurality of zones dividing the work region, (Par. 0036: “In various embodiments, the information for the pathing 320 is sent to the data collector 330 to navigate to an initial or subsequent data collection point 310 and to collect specific data at that data collection point 310. For example, the data collector 330 may be instructed to take a photograph of the plants located at the data collection point 310, which the yield predictor analyzes to extract various data about how the plants are growing at the data collection point 310 (e.g., leaf size, number of leaves, number of fruiting bodies, size of fruiting bodies, estimated content of fruiting bodies, density of plants per area, plant height). In other embodiments, the data collector 330 may specify data values regarding the plants (e.g., leaf size, number of leaves, number of fruiting bodies, size of fruiting bodies, estimated content of fruiting bodies, density of plants per area, plant height), the soil (e.g., pH, grain size, moisture content, temperature), air (e.g., temperature, pollen/particulate count, humidity), and treatments applied to the data collection point 310 (e.g., fertilizers, pesticides), and the like.” Par. 0016: “FIG. 2 illustrates cluster identification in a crop growing area 100, according to embodiments of the present disclosure. As illustrated, the fields 110a-c are subdivided into several regions 220a-g (generally, region 220) that represent similar growing conditions; ignoring the non-analyzed areas 210 (including the various sectors 120 in the crop growing area 100).” See also Par. 0013, 0019, and 0021.) 
each zone (region 220x) being sized and shaped to capture homogeneous environmental conditions in a same zone; (Par. 0012: “The present disclosure therefore provides systems and methods for identifying regions in which plants are expected to experience similar growing conditions,…” Par. 0016: “FIG. 2 illustrates cluster identification in a crop growing area 100, according to embodiments of the present disclosure. As illustrated, the fields 110a-c are subdivided into several regions 220a-g (generally, region 220) that represent similar growing conditions; ignoring the non-analyzed areas 210 (including the various sectors 120 in the crop growing area 100).” Par. 0019: “The yield predictor includes a machine learning model that is trained to use the characteristics of the field 110 to identify clusters of characteristics that define regions 220 in which similar growing conditions are expected. In various embodiments, the region-identifying machine learning model may be further constrained by the yield predictor to produce regions of a certain number, size, or general shape, to thereby aid data collection and differentiation. For example, a field 110 may be subdivided into two or more regions 220 of at least n hectares that describe the total area of the field 110, where different regions 220 describe portions of the field 110 where the plants are expected to have different growth patterns.”  See also Par. 0018.)
providing a set of zone sensor data as an input to a predictive turf model for each zone; (Par. 0027: “…the yield predictor may use an average (mean) value for the data collected from the multiple data collection points 310 or may select one data collection point 310 to provide the data to be used in estimating yield for the region 220 (e.g., using additional data collection points 310 to verify remotely gathered data or provide additional feedback/training for the machine learning models).” Par. 0032: “At block 420, the yield predictor identifies regions 220 within the fields 110. The yield predictor uses known data for the field 110 to identify several regions 220 therein that are predicted to produce similar yields for the crop. The yield predictor uses a machine learning model trained to identify clusters of relevant criteria that affect crop yield. These criteria may be weighted differently from one another in the machine learning model, and can include one or more of: vegetation indexes visible in images of the field 110, soil conditions throughout the field 110 from previously collected data or soil maps, fertilizer application levels throughout the field 110 from previously collected data or operator reporting, pest control application levels throughout the field 110 from previously collected data or operator reporting, and previously collected vegetative growth and yield metrics from an earlier data collection iteration in the field 110.”  Examiner’s Note – The yield predictor uses data collection points as input and uses the data for a learning model.) 
determining an estimated turf condition (plant yield or yield or growth) for each zone based on an output of the predictive turf model for each zone; (Par. 0039: “At block 460, the yield predictor refines the machine learning models and crop simulation models used in analyzing the crop yields. In some embodiments, the yield predictor performs a regression analysis of growing condition data (from the in-field data collected from the data collection points 310, from image analysis of the fields 110, and from mapped data sources) to the yield related data (i.e., the data directly related to crop production per unit area in a region 220) to identify relationships and relative effects of soil conditions, elevation, irrigation types and volumes, fertilizer types and levels, pest control types and levels, vegetation indices, daily temperatures, sun intensity, etc., on the output of the plants in the region 220. In additional embodiments, the yield predictor uses one or more intermediate models to output intermediate crop development metrics for use by other models when predicting the yield, and refines those models when in-field data are available to compare against. For example, intermediate models may be used to identify one or more of a leaf area index, a development stage for the crop, a dry weight of the crop (e.g., of a grain or legume), a number of seeds/pods/ears/fruiting bodies per plant, etc. Using the identified characteristics from the regression analysis, the yield predictor refines the models used to identify regions 220 and data collection points 310 by adjusting the weights assigned to the various inputs. By refining the models, the yield predictor improves the outputs of the machine learning models so that in a next iteration the models will identify new regions 220 to describe the fields 110, new data collection points 310 in the fields 110, and/or how to model the growth conditions in the field 110.” See also paragraph 0042. Examiner’s Note – The specification of the instant application paragraph describes that plant growth and yield may indicate a turn condition such as paragraph 0037 and ) 
Bentwich does not teach generating a recommended action. However, Bentwich does teach generating a recommended action for each zone based on the associated estimated turf condition for each zone. (Par. 0036: “…it would be possible to divide the field into effective irrigation zones, and monitor soil moisture in each of these zones, knowing that the same soil moisture is expected to be found everywhere within this zone. Irrigation could then be guided accordingly.” Par. 0099: “The compute irrigation 400 compares each sensor reading received, with the soil propertied of the soil of the irrigation zone, and the irrigation goal defined by the user, and calculates accordingly the recommended irrigation for that zone. Next step, present to user via app 420, preferably presents a tentative irrigation map, for each of the zones of the field 105 of FIG. 1, preferably via an app on a mobile device, or a computer, or a web browsing device.”  See also Par. 0079 and 0080 that teach predictive modelling; and several other paragraphs teach generating a recommended action such as paragraphs 0100, 0122, 0132, 0142, 0143, etc.) 
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have combined the system and method that allows adjustments to irrigation, pest control, etc. that uses sensors and data points and develops a machine learning model and predicts a yield or growth for each divided region of a field that have similar growing conditions as in Silva with method of controlling irrigation by predictive modeling of divided irrigation zones and calculate a recommend action for a particular zone as in Bentwich in order to create an efficient irrigation plan which ascertains a correct amount of water that represents water savings. (Par. 0010)

Regarding claim 17,
The previously cited references teach the limitations of claim 16 which claim 17 depends.  Silva also teaches that the predictive turf model is calibrated based on historical outcome data and historical sensor data of the work region. (Par. 0018: “The yield predictor identifies what qualifies as “similar growing conditions” based on the known characteristics of the field 110, which can include previously collected in-field data from the growth of the plants therein, physical features of the field 110, and image analysis of the field 110. For example, using one or more images of the fields 110, the yield predictor may identify a vegetation index using two or more spectral bands (e.g., the photosynthetically active radiation band and the near-infrared band) to identify where plants are growing and in what density. In another example, the yield predictor uses soil condition maps, topology maps, flood maps, climate/weather maps, and irrigation system schematics to identify areas that are expected to experience similar water and nutrient collection/retention characteristics. An operator may also provide data related to fertilization levels, pest control levels, and previously collected growth analysis data (e.g., at time to the density of plants per area, density of pods per plant, and density of seeds per pod were x, y, and z, respectively).” See also Par. 0023, 0032, 0034, 0040, and 0042.) 

Regarding claim 18,
The previously cited references teach the limitations of claim 16 which claim 18 depends.  Silva also teaches that updating the predictive turf model in response to the zone sensor data and a measured turf condition. (Par. 0039: “By refining the models, the yield predictor improves the outputs of the machine learning models so that in a next iteration the models will identify new regions 220 to describe the fields 110, new data collection points 310 in the fields 110, and/or how to model the growth conditions in the field 110.”  Par. 0021: “For example, the sections of the first region 220a and the second region 220b that border one another may exhibit the same yields as one another, and the sections of the second region 220b and third region 220c that border one another may also exhibit the same yields as one another, which are different than the yields at the border of the first and second regions 220a-b. Stated differently, yields may vary over the course of an individual region 220 more significantly than yields vary between adjacent portions of different regions; however, the yields within the individual region 220 are predicted to be within a given range of the average yield for that region 220 despite any variability therein. As data are collected for the regions 220, the machine learning model may update how different features are weighted and move the borders of the second region 220b (e.g., expanding the first region 220a and shrinking the second region 220b or vice versa) to better reflect the actual growth conditions in future predictions for regional yield.” Examiner’s Note – See also Bentwich Paragraph 0079 and 0080.) 

Regarding claim 19,
The previously cited references teach the limitations of claim 18 which claim 19 depends.  Silva also teaches that wherein updating the predictive turf model comprises: measuring the measured turf condition for at least one zone of the plurality of zones; and comparing the estimated turf condition and the measured turf condition for the at least one zone.  (Par. 0039: “At block 460, the yield predictor refines the machine learning models and crop simulation models used in analyzing the crop yields. In some embodiments, the yield predictor performs a regression analysis of growing condition data (from the in-field data collected from the data collection points 310, from image analysis of the fields 110, and from mapped data sources) to the yield related data (i.e., the data directly related to crop production per unit area in a region 220) to identify relationships and relative effects of soil conditions, elevation, irrigation types and volumes, fertilizer types and levels, pest control types and levels, vegetation indices, daily temperatures, sun intensity, etc., on the output of the plants in the region 220. In additional embodiments, the yield predictor uses one or more intermediate models to output intermediate crop development metrics for use by other models when predicting the yield, and refines those models when in-field data are available to compare against. For example, intermediate models may be used to identify one or more of a leaf area index, a development stage for the crop, a dry weight of the crop (e.g., of a grain or legume), a number of seeds/pods/ears/fruiting bodies per plant, etc. Using the identified characteristics from the regression analysis, the yield predictor refines the models used to identify regions 220 and data collection points 310 by adjusting the weights assigned to the various inputs. By refining the models, the yield predictor improves the outputs of the machine learning models so that in a next iteration the models will identify new regions 220 to describe the fields 110, new data collection points 310 in the fields 110, and/or how to model the growth conditions in the field 110.” Par. 0040: “Method 400 may conclude after block 460, or may return to block 420, where the newly refined models identify a new set of regions 220 that more accurately represent the average production of the crop growing in those fields 110 than in previous iterations. Method 400 may iterate several times throughout a growing season at periodic intervals (e.g., every d days), at an operator's request, (using one data set) until the machine learning models converge on a stable set of weights for the input features, or until another end condition is satisfied.” See also Par. 0054, ) 

Regarding claim 20,
The previously cited references teach the limitations of claim 19 which claim 20 depends.  Silva also teaches that wherein measuring the measured turf condition comprises inputting the measured turf condition based on observed turf condition at the at least one zone of the plurality of zones. (Par. 0027: “The yield predictor extrapolates the observed data from the data collection points 310 to the rest of the region 220, and may use kriging or a data-driven model to blend the estimates from neighboring regions 220 to arrive at a yield estimate for a field 110.” Par. 0022 - 0023: “A data collector 330 may be an automated system (such as the first data collector 330a, which is illustrated as a flying drone) or a person (such as the second data collector 330b). The yield predictor identifies at least one data collection point 310 per region 220, and may select the data collection point 310 to be the most representative point in the region.” “See also Par. 0024.  Examiner’s Note – see also Bentwich in numerous paragraphs that teach user input (observed) and also paragraphs 0026) 

Regarding claim 21,
The previously cited references teach the limitations of claim 19 which claim 21 depends.  Silva also teaches that the measured turf condition is measured by one or more turf condition sensors. (Par. 0045: “In one instance, the vegetation index may have been gathered when farm equipment was present in the field 110, and the outlier location may be an outlier due to the farm equipment obscuring the crop from the view of the remote sensor 130, and can be included in the area of the region 220 used for yield predictions. In another instance, the variance in vegetation index may be due to an ongoing factor that may reduce the area of the region 220 that is used for yield predictions (e.g., the location corresponds to a well head, a pivot for an irrigation system, an area around a salt lick, the footprint of a wind turbine or other permanent fixture in the field 110, etc.). Accordingly, the field analyzer may identify all the locations in the region 220 are outliers as candidate data collection points 310 for the associated region 220; however, these candidate outlier data points are treated differently than the candidate typical data points (identified per block 520) by the field analyzer and the yield predictor.” Par. 0014: “Three remote sensors 130a-c (generally, remote sensor 130), are illustrated in relation to the crop growing area 100 that provide remote sensing of various features of the fields 110 and sectors 120 therein. A remote sensor 130 may be a satellite imager (as in the first remote sensor 130a), a fully or semi-autonomous aerial imager or “drone” (as in the second remote sensor 130b), a piloted aerial imager (as in the third remote sensor 130c), or another system capable of capturing images of the crop growing area 100 that can be used to identify individual fields 110 or sectors 120 and/or conditions affecting or evidencing plant growth therein. These data may be combined to form a composite image from several remote sensors 130 that are collected as substantially the same time, or at substantially different times. For example, a swarm of several drone-type remote sensors 130 may be directed to collect remotely sensed data over the course of one day, which can be combined into a composite image and data set related to the day, which can further be combined with an earlier (or later) collected composite image or data set (e.g., collected by the swarm or a satellite on a different day).”) 

Regarding claim 22,
The previously cited references teach the limitations of claim 21 which claim 22 depends.  Silva also teaches that the one or more turf condition sensors comprise a light reflectance sensor. (Par. 0039: “At block 460, the yield predictor refines the machine learning models and crop simulation models used in analyzing the crop yields. In some embodiments, the yield predictor performs a regression analysis of growing condition data (from the in-field data collected from the data collection points 310, from image analysis of the fields 110, and from mapped data sources) to the yield related data (i.e., the data directly related to crop production per unit area in a region 220) to identify relationships and relative effects of soil conditions, elevation, irrigation types and volumes, fertilizer types and levels, pest control types and levels, vegetation indices, daily temperatures, sun intensity, etc., on the output of the plants in the region 220.”) 

Regarding claim 23,
The previously cited references teach the limitations of claim 18 which claim 23 depends.  Bentwich also teaches determining a subsequent estimated turf condition for each zone based on an output of the updated predictive turf model for each zone and generating a subsequent recommended action for each zone based on the associated subsequent estimated turf condition for each zone. (Par. 0132: “Within informational graphical illustration 930, two iconized elements are displayed: a bar graphic representation 950 and a drop graphic representation 960. The bar graphic representation 950 graphically displays a relation between a current soil moisture level in soil-topography zone 910 relative to a field-capacity value and a refill-point value, both of soil-topography zone 910. The drop graphic representation 960 graphically displays an amount of irrigation recommended by the system for a next irrigation event of soil-topography zone 910 relative to a maximal amount of irrigation in an irrigation event, as determined by the user. Numeral 955 designates the current soil moisture level in soil-topography zone 910. In one embodiment, numeral 955 represents an integration of a plurality of soil moisture measurements taken at different depths, such as two readings taken at two depths, or three readings taken at 3 depths. In another embodiment, numeral 955 is a number of volume units of water within a predetermined length of depth of soil, such as square millimetres or square inches, in a meter or another predetermined length measurement unit, of soil. In another preferred embodiment, numeral 955 represents an integration of soil moisture readings taken by several sensors, in order to increase the measurement accuracy. Numeral 365 designates the amount of irrigation recommended by the system for the next irrigation event of soil-topography zone 910. Bar graphic representation 970, drop graphic representation 980, and numerals 975 and 985 are similar to elements 950, 960, 955 and 965, respectively with the exception that they are related to soil-topography zone 920 and are displayed within informational graphical illustration 940.”) 

Regarding claim 24,
The previously cited references teach the limitations of claim 16 which claim 24 depends.  Silva also teaches monitoring the multiple sets of zone sensor data comprises: collecting the sets of zone sensor data for the associated zone from at least one of connected equipment, user inputs, and sensor nodes; transmitting the sets of zone sensor data to a network; and verifying the sets of zone sensor data collected. (Par. 0014 and 0015: “These data may be combined to form a composite image from several remote sensors 130 that are collected as substantially the same time, or at substantially different times. For example, a swarm of several drone-type remote sensors 130 may be directed to collect remotely sensed data over the course of one day, which can be combined into a composite image and data set related to the day, which can further be combined with an earlier (or later) collected composite image or data set (e.g., collected by the swarm or a satellite on a different day). A yield predictor identifies the crop growing area 100 in one or more images (e.g., a composite image) that are captured via the remotes sensors 130. In various embodiments, the yield predictor identifies defined zones in the crop growing area 100 by machine image processing, by geolocation coordination, or by manual selection. For example, the yield predictor may use coloration differences between portions of the crop growing field 110 used to grow plants and roadways, fences, barren areas, wild areas, residential areas and the like to identify different fields 110 and sectors 120 from the image of the crop growing area 100.”   Par. 0053: “Further, the computing system 600 is included to be representative of a physical computing system as well as virtual machine instances hosted on a set of underlying physical computing systems. Further still, although shown as a single computing system, one of ordinary skill in the art will recognize that the components of the computing system 600 shown in FIG. 6 may be distributed across multiple computing systems connected by a data communications network.”  Par. 0060: “Computer readable program instructions described herein can be downloaded to respective computing/processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and/or a wireless network. The network may comprise copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers. A network adapter card or network interface in each computing/processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium within the respective computing/processing device.” See also Par. 0061. See also Figure 6 that shows a computing device that may be a server on a network that processes the data and comprises machine learning models and the yield predictor.) 

Regarding claim 25,
The previously cited references teach the limitations of claim 16 which claim 25 depends.  Silva also teaches that the sets of zone sensor data associated with a zone comprise information pertaining to previous action taken to manage conditions at the associated zone. (Par. 0032: “These criteria may be weighted differently from one another in the machine learning model, and can include one or more of: vegetation indexes visible in images of the field 110, soil conditions throughout the field 110 from previously collected data or soil maps, fertilizer application levels throughout the field 110 from previously collected data or operator reporting, pest control application levels throughout the field 110 from previously collected data or operator reporting, and previously collected vegetative growth and yield metrics from an earlier data collection iteration in the field 110.”  Par. 0018: “An operator may also provide data related to fertilization levels, pest control levels, and previously collected growth analysis data (e.g., at time to the density of plants per area, density of pods per plant, and density of seeds per pod were x, y, and z, respectively).” Examiner’s Note – See also Scheiner in the conclusion section.) 

Regarding claim 26,
The previously cited references teach the limitations of claim 16 which claim 26 depends.  Silva also teaches that the sets of zone sensor data associated with a zone comprise environmental conditions and trends. (Par. 0017: “In various embodiments, the regions 220 are differentiated from one another by identified conditions in the soil, growth patterns of plants therein, altitudes, or the like.” See also paragraph 0019. 

Regarding claim 27,
The previously cited references teach the limitations of claim 16 which claim 27 depends.  Bentwich also teaches that displaying the recommended action for each zone to a user.  (Par. 0132: “The drop graphic representation 960 graphically displays an amount of irrigation recommended by the system for a next irrigation event of soil-topography zone 910 relative to a maximal amount of irrigation in an irrigation event…” See also Par. 0142.) 

Regarding claim 28,
The previously cited references teach the limitations of claim 16 which claim 28 depends.  Bentwich also teaches that comprising providing a command to turf maintenance equipment to automatically carry out the recommended action for each zone.  (Par. 0005: “Topography Integrated Ground watEr Retention (TIGER) map generator includes: a computerized topographic feature processing functionality providing information relating to at least one of slope, aspect and catchment area features of said area to be irrigated; and a computerized topographic feature utilization functionality employing at least one of slope, aspect and catchment area features of the area to be irrigated for automatically ascertaining water retention at a plurality of different regions within the area to be irrigated; and a computerized computing functionality employing the Topography Integrated Ground watEr Retention (TIGER) map together with at least current outputs of wetness sensors located at the plurality of different regions within the area to be irrigated to generate a current irrigation plan; and a computerized irrigation control subsystem automatically utilizing the current irrigation map to control irrigation within the area to be irrigated based on the current irrigation instructions and to cause different amounts of water to be provided to the different regions within the area to be irrigated.” Par. 0103: “In accordance with another preferred embodiment, the differential irrigator 100 of FIG. 1 may automatically control differential irrigation of the field 105, through use of a drip irrigation system.” See also Par. 0106.) 

Regarding claim 36,
The previously cited references teach the limitations of claim 16 which claim 36 depends.  Silva also teaches that each zone includes a single node providing a representation of a micro-climate of the respective zone, and wherein each single node within each zone includes one or more sensors. (Par. 0023: “The yield predictor identifies at least one data collection point 310 per region 220, and may select the data collection point 310 to be the most representative point in the region. Stated differently, a data collection point 310 is selected to represent what the average growth conditions in the region 220 will produce. The yield predictor may impose further constraints on the selection of data collection points 310, such as, for example, a minimum or maximum distance to one or more other points (e.g., another data  collection point 310, a border of a region 220, an irrigation system, a fence, a path), a minimum number of data collection points 10 per region 220, the relative locations of previously selected data collection points 310 (e.g., at least x meters away from where a previous measurement was gathered d days ago), and the like.” 


Claims 29 and 30 are rejected under 35 U.S.C. 103 as being unpatentable over Silva in view of Bentwich in further view of McEntire et al. (US Patent No. 11,704,581), herein “McEntire,” supported by four provisional patent applications filed before the effective filing date of the instant application. 

Regarding claim 29,
The previously cited references teach the limitations of claim 16 which claim 29 depends. They do not teach the elements of claim 29 wherein a recommended action comprises forming a ruleset and executing the ruleset and model via knowledge engine. However, McEntire does teach generating the recommended action comprises: forming a ruleset (list or response surfaces for each interaction or instruction blocks, See figure 14.) comprising a set of rules ordered by priority, the rules producing deductions as outputs; (Col. 61, lines 49 – 63: “ The illustrative method continues with relative accuracy testing using cross validation at decision diamond 14250 and tuning of the RS weighting coefficients at process block 14260 and model refitting at process block 14270 to achieve the desired accuracy and model estimation expectations for each of the interactions in the list. The illustrative regression process examines the estimated model output for the main-effect and the main co-occurring features from one or more set of interventions that have exhibited the highest co-occurrence frequency from the RF model. Once a pruned list of low order interactions is built by ranking and reduction analysis, the method iterates through at least one list of Response Surfaces starting with the surface that trends to be the closest to the main-effect, moving through the list adding and removing certain interventions from the list.” See also Col. 60 lines 43 – 65; Col. 61, line 37 – Col. 62, line 36) 
 forming one or more models that take the zone sensor data, the deductions, and initial conditions as inputs and provide new deductions as output; (Col. 23, lines 12 – 27: “For example, the crop prediction engine 4000 may prescribe nutrients, seeds or other farm management supplies and based on such predictions autonomously place purchase orders directly to manufacturers and suppliers. Additionally, a client device may be used to directly obtain bids from crop purchasers and/or crop brokers 3030 who can view estimated production volumes and prices directly. The prediction engine 4000 inputs data from multiple client devices to source information and derive the optimal soil and seed application costs for desired crop production in preparation for planting by allowing the farmer, agronomist, or crop consultant to rely on trained AI models to understand the multiplicity of soil and seed characteristics decisions. Thus, recommendations for optimal crop efficiency with precision application may be obtained through the crop prediction engine 4000 described herein.” Col. 35, lines 12 – 26: “At process block 5300, program instructions representing an illustrative training model are used by the crop prediction engine 4000. In the illustrative embodiment, a random forest (RF) training model is implemented as a computational model. In operation, each of the RF trees is built into a set of tree estimators and calibrated using first out-of-bag training data, which assesses initial conditions, parameter settings, and first pass model quality using techniques like R-squared error minimization prior to application of actual training data sets. Tuning of the RF training model may be iterative and may use different feature weights based on the desired properties inherent to the set of training data as known to one of skill in the art. Once the RF training model has been tuned, a data set of multi-dimensional set of features is applied to each tree in the random forest.” See also Col. 2, lines 43 – 67. See also Col. 22, line 10 – Col. 23, line 61; and Col. 61, line 37 – Col. 62, line 62; Col. 63, line 46 – Col. 66, line 36, and figures 4, 14, and 16, respectively, that describes the interactive (ruleset) of the crop prediction engine that uses modeling that results in recommendations) 
and executing the ruleset and models via a knowledge engine to determine the recommended action. (Col. 66, lines 50 – 62: “The final step performed by the programming software running on the Data Science Computing Cluster 4300 with support from the NFS software 4301 is to use all observations to formulate Reduced Order Surrogate model (RoSM) and subsequently a GAM model to determine the optimum seed and chemical nutrients to apply to achieve the maximum crop performance 14000. This “Prescription” is then sent to the Application Back-End 4001 for further processing and subsequent downloading via the Transactional Data Transport Services 4002 to the client browser 4003 for final output display and/or printed reports or manufacturer supply orders for seed and nutrient recommendations and purchases.”) 
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have combined the system and method that allows adjustments to irrigation, pest control, etc. that uses sensors and data points and develops a machine learning model and predicts a yield or growth for each divided region of a field that have similar growing conditions as in Silva with method of controlling irrigation by predictive modeling of divided irrigation zones and calculate a recommend action for a particular zone as in Bentwich with using interactive programming, containing rules, with machine learning that determines recommendations of chemical application for plants or crops as in McEntire in order to  build a generalized model running as a tool on scalable, client/server web based computing platforms in order to optimize agricultural efficacy. (Col. 4, lines 55 – 58).

Regarding claim 30,
The previously cited references teach the limitations of claim 29 which claim 30 depends.  McEntire also teaches displaying the ruleset to a user to explain why the recommended action was selected. (Col. 66, lines 1 – 23: “The Application Back-End 4001 software programming enables the cloud computing hardware 4100 along with the System Software NFS Layers 4101 to send the geographical representation of each Voxel. In the illustrative embodiment, the geographical representation is sent through one or more load balancers 4110, one or more Internet Networks 3000 and terminating at one or more local gateways 4050 prior to output display on one or more client browsers 4003. The output displayed may include: input variables, output dependent variables and other predicted crop or farm operational responses 4430. These output responses 4430 may be displayed visually, by text lists or be embedded into downloadable reports or supplier order forms. Calculated and predicted outputs sent to the application computing cluster from block 8500 for mapping and display may include: AGT data-set clusters 4430i, crop-yield predictions 4430h, soil carbon predictions 4430f, ROI on operations and harvest analysis 4430e, digital elevation and moisture indexes 4430g, various interactive maps 4430a, nutrient recommendations 4430b, seed recommendations 4430c, digital prescriptions 4430d, and ETL results for big data repository and data-completeness verification purposes.” See also Col. 67, lines 37 – 60.)

Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:

Jagyasi et al. (US PG Pub. No. 20170223900) may also teach that the elements of the independent claim of generating a recommended action for each zone based on the associated estimated turf condition for each zone in paragraph 0007: “…generic forecasting generation module (216); selecting at least one feature out of the feature set for generating a plurality of adaptive forecasting model based for ecological forecasting on the selected feature out of the feature set using an adaptive forecasting generation module (218); and recommending a plurality of control measures to a user based on said generated adaptive forecasting model using a recommendation generation module (220).” 

Scheiner et al. (US PG Pub. No. 20230165181) may also teach the elements of claim 25 in Par. 0275: “The analyzer/control system may also recommend adding a run or varying the protocol of a run to increase the sampling density of a certain area of the growing site in order to verify the effectivity of a specific treatment action previously carried out.” 

Any inquiry concerning this communication or earlier communications from the examiner should be directed to CHAD G ERDMAN whose telephone number is (571)270-0177. The examiner can normally be reached Mon - Fri 7am - 5pm EST; Off every other Friday.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Kamini S. Shah can be reached at (571) 272-2279. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.





/CHAD G ERDMAN/Primary Examiner, Art Unit 2116