Computer-implemented method and non-transitory computer-readable storage medium for seed product selection and recommendation
Through the computer decision support system, the genetic characteristics of seeds and field characteristics are used to generate target seed sets and planting strategies, solving the problem of matching seed selection and field, and improving the stability and adaptability of agricultural output.
Patent Information
- Application Number
- CN201980070460.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2019-04-10
- Filing Date
- 2019-10-24
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2039-10-24
AI Technical Summary
The decisions of seed selection and planting strategies in agriculture are difficult to effectively match field conditions, resulting in a decrease in yield under instability and environmental fluctuations.
A computer-implemented decision support system was developed to utilize the genetic characteristics of seeds and agricultural characteristics of the fields, and by generating target success yield groups and target seed sets, appropriate seed selection and planting strategies were recommended to improve yield and consistency.
Through data processing and machine learning models, precise matching of seeds and field conditions is improved, yield stability and adaptability are reduced, and the impact of environmental fluctuations on yield is reduced.
Smart Images

Figure CN112930544B_ABST
Abstract
Description
[0001] Copyright Notice
[0002] A portion of the disclosure of this patent document contains material that is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by anyone of the patent document or the patent disclosure, as it appears in the Patent and Trademark Office patent file or records, but otherwise reserves all copyright rights whatsoever. 2015-2019 Climate Corporation. Technical Field
[0003] One technical area of the present disclosure is computer-implemented decision support systems for agriculture, particularly in seed selection and planting strategies. Another technical area is a computer system programmed to generate predicted and comparative yield data for one or more fields using genetic characteristics of seeds and agricultural characteristics of fields. Another technical area is a computer system programmed to recommend seed selection and placement in one or more unique target fields to help improve yield and consistency. Background Art
[0004] The approaches described in this section are approaches that could be pursued, but are not necessarily approaches that have been previously conceived or pursued. Therefore, unless otherwise indicated, it should not be assumed that any of the approaches described in this section qualify as prior art merely by virtue of their inclusion in this section.
[0005] Successful harvesting depends on many factors, including seed selection, soil fertilization, irrigation, pest control and management practices, each of which contributes to the growth rate of a plant (e.g., a corn or soybean plant). One of the most important agricultural management factors is the selection of seeds to be planted on the target field. Seed varieties or hybrids range from seeds suitable for a short growing period to seeds with a longer growing period, seeds suitable for high temperatures or seeds suitable for low temperatures, seeds suitable for dry climates or seeds suitable for wet climates, and different seeds suitable for specific soil compositions. Achieving optimal performance of a particular seed hybrid or variety depends on whether field conditions coincide with the optimal growing conditions for the particular seed. For example, a particular corn hybrid may be rated to produce a specific amount of yield for a grower, but if field conditions do not match the optimal conditions used to evaluate the particular corn hybrid, the corn hybrid will not likely continue to meet the grower's expectations for yield.
[0006] Once a seed set has been selected for planting, the grower must determine a planting strategy. The planting strategy includes determining the amount and location of each seed in the selected seed. The strategy for determining the amount and location can determine whether the yield of the harvest meets expectations. For example, if conditions are favorable, planting seeds with similar strength and fragility can lead to high yields. However, if conditions fluctuate (such as, rainfall is less than expected or the temperature is higher than normal), the total yield of similar seeds can be reduced. In order to overcome unforeseen environmental fluctuations, it may be necessary to adopt a diversified planting strategy.
[0007] The technology described herein helps alleviate some of these problems and helps growers determine which seeds to plant in which fields. BRIEF DESCRIPTION OF THE DRAWINGS
[0008] Figure 1 An example computer system configured to perform the functions described herein is illustrated, the example computer system being shown in a field environment along with other devices with which the system can interact.
[0009] Figure 2 Illustrated are two views of an example logical organization of a set of instructions in main memory when an example mobile application is loaded for execution.
[0010] Figure 3 Illustrated is a programmed process by which an agricultural intelligence computer system generates one or more pre-configured agronomic models using agronomic data provided by one or more data sources.
[0011] Figure 4 is a block diagram illustrating a computer system upon which embodiments of the present disclosure may be implemented.
[0012] Figure 5 An example embodiment of a timeline view for data entries is depicted.
[0013] Figure 6 An example embodiment of a spreadsheet view for data entry is depicted.
[0014] Figure 7 An example flow chart for generating a target successful yield group of seeds identified for optimal yield performance on a target field based on agricultural data records for the seeds and geographic location data associated with the target field is depicted.
[0015] Figure 8 Depicted are examples of different regions within a state having different assigned relative maturity levels based on the duration of the growing season.
[0016] Fig. 9 Depicted is a graph describing the range of standardized yield values for seeds within categorized relative maturity.
[0017] Fig.10 Depicted is an example flow chart for generating a target set of seeds identified for optimal yield performance and managing risk for a target field based on agricultural data records of the seeds and geographic location data associated with the target field.
[0018] Fig.11 Depicted are example graphs of yield value versus risk value for one or more seeds.
[0019] Fig.12 An example flow chart for utilizing genetics to fill data gaps in historical agricultural data is illustrated.
[0020] Fig.13 An example of received agricultural data records and further processing to fill in data gaps is illustrated.
[0021] Fig.14 Another example of received agricultural data recording and further processing to fill data gaps is illustrated.
[0022] Fig.15 An example of genetic signature data including genomic marker data is illustrated.
[0023] Fig.16 An example pedigree-based affinity matrix is illustrated that identifies pairwise relationships between seeds based on their pedigrees.
[0024] Fig.17 An example of organizing seeds into genetic cluster relationships is shown.
[0025] Fig.18 An example genetic marker-based kinship matrix is illustrated, which identifies pairwise relationships between seeds based on SNP markers.
[0026] Fig.19 An exemplary inbreeding code for an inbred parental line of captured product is illustrated.
[0027] Fig. 20 The diagram shows the use of feature engineering to classify feature data and Fig.12 Example flow chart of preparing agricultural data records using the recommended model. DETAILED DESCRIPTION
[0028] In the following description, for the purpose of explanation, numerous specific details are set forth in order to provide a thorough understanding of the present disclosure. However, it will be apparent that embodiments can be practiced without these specific details. In other instances, well-known structures and devices are illustrated in block diagram form to avoid unnecessarily obscuring the present disclosure. According to the following summary, embodiments are disclosed in various sections:
[0029] 1. General Overview
[0030] 2. Example Agricultural Intelligent Computer System
[0031] 2.1 Structural Overview
[0032] 2.2 Application Overview
[0033] 2.3 Data entry into computer system
[0034] 2.4 Process Overview—Agronomic Model Training
[0035] 2.5 Seed classification subsystem
[0036] 2.6 Seed Recommendation Subsystem
[0037] 2.7 Implementation Example—Hardware Overview
[0038] 3. Functional Overview - Generate and display target successful yield groups of seeds
[0039] 3.1 Data Input
[0040] 3.2 Agricultural data processing
[0041] 3.3 Existing target success yield group
[0042] 4. Functional Overview - Generate and display target seeds for planting
[0043] 4.1 Data Input
[0044] 4.2 Seed Selection
[0045] 4.3 Generating risk values for seeds
[0046] 4.4 Generate target seed dataset
[0047] 4.5 Seed profile analysis
[0048] 4.6 Existing Set of Target Seeds
[0049] 5. Functional Overview - Generate and display yield improvement recommendations by field
[0050] 5.1 Data Input
[0051] 5.2 Data Attribution
[0052] 5.3 Determine expected yield performance
[0053] 5.4 Seed Optimization and Recommendation Generation
[0054] 5.5 Validating and Tuning the Model
[0055] 6. Functional Overview—Including examples of feature engineering to enhance data for recommendation modeling
[0056] 6.1 Original Features and Feature Classification
[0057] 6.2 Data Preparation
[0058] 1. General Overview
[0059] Disclosed herein is a computer system and a computer-implemented method for generating a set of target successful yield groups of hybrids or seed varieties that have a high probability of successful yield on one or more target fields. In an embodiment, a server computer system may be used to generate a target successful yield group of seeds, the server computer system being configured to receive, via a digital data communication network, one or more agricultural data records representing crop seed data describing seed and yield attributes of one or more seeds, and first field geographic location data of one or more agricultural fields where the one or more seeds are planted. The server computer system then receives second geographic location data of the one or more target fields where the seeds are to be planted.
[0060] The server computer system includes seed standardization instructions, which are configured to generate a data set of seed attributes, and the data set of the seed attributes describes the representative yield value and environmental classification of each seed from one or more agricultural data records. Then, the instruction of the probability of success generation on the server computer system is configured to generate a data set of success probability scores, and the data set of the probability of success describes the probability of successful yield on one or more target fields. Successful yield can be defined as an estimated yield value for a specific seed for environmental classification, and the estimated yield value of the seed exceeds a specific yield number than the average yield of the same environmental classification. The success probability value of each seed is based on the data set of seed attributes, and the second geographical location data of one or more target fields.
[0061] The server computer system includes yield classification instructions, which are configured to generate a target successful yield group consisting of subsets of one or more seeds and success probability values associated with each of the subsets of the one or more seeds, wherein the generation of the target successful yield group is based on a data set of success probability scores for each seed and a configured successful yield threshold, and if the success probability value of the seed exceeds the successful yield threshold, the seed is added to the target successful yield group.
[0062] The server computer system is configured to cause display of the target successful yield group and the yield value associated with each seed in the target successful yield group on a display device communicatively coupled to the server computer system.
[0063] In an embodiment, a target successful yield group (or another set of seeds and fields) can be used to generate a target seed set selected for planting on one or more target fields. The server computer system is configured to receive a target successful yield group of candidate seeds, which can be a candidate for planting on one or more target fields. The target successful yield group includes one or more seeds, a success probability value associated with each of the one or more seeds describing the probability of successful yield, and historical agricultural data associated with each of the one or more seeds. Then, the server computer receives attribute information related to the one or more target fields.
[0064] The seed filtering instructions within the server computer system are configured to select a subset of hybrids or seed varieties having a probability of success value greater than a target probability filtering threshold. The server computer system includes seed standardization instructions configured to generate representative yield values for seeds in the subset of one or more seeds based on historical agricultural data.
[0065] The server computer system includes risk generation instructions configured to generate a data set of risk values for a subset of one or more seeds. The data set of risk values describes the risk associated with each seed based on historical agricultural data. The server computer system includes optimization classification instructions configured to generate a data set of target seeds for planting on one or more target fields based on the risk values, representative yield values for a subset of one or more seeds, and a data set of one or more attributes for one or more target fields. The data set of target seeds includes target seeds whose representative yield values meet a specific target threshold of a risk value range from the data set of risk values across the one or more target fields.
[0066] The server computer system is configured to display a dataset of target seeds on a display device communicatively coupled to the server computer system, the dataset of target seeds including representative yield values and risk values from a dataset of risk values associated with each target seed in a dataset of target seeds and one or more target fields.
[0067] In another embodiment, a computer-implemented method includes: receiving an agricultural data record via a digital data communication network at a server computer system, the agricultural data record including a first yield attribute set of a first seed set grown in a first set of environments; and further receiving genetic characteristic data related to a second seed set via the digital data communication network, wherein the second seed set includes the first seed set. The method also includes using the server computer system to generate a second yield attribute set for a second seed set associated with a second set of environments by applying the genetic characteristic data to the agricultural data record. In this example, the second yield attribute set fills in the data gap from the first yield attribute set. The server computer system can then be used to determine the predicted yield performance of one or more seeds (such as a third seed set) on one or more target fields, the third seed set being the same or different from the first seed set and / or the second seed set. The predicted yield performance can be based on one or more of the following: an absolute or relative yield value, a yield grade, a probability of a success score, and / or other factors. In one example, the server computer determines the predicted yield performance of a second seed set associated with a second set of environments by applying the estimated yield attributes, and generates a yield improvement recommendation based on the predicted yield performance for the second seed set. The method may also include displaying the yield improvement recommendation on a display device communicatively coupled to the server computer system.
[0068] In another embodiment, a computer-implemented method includes receiving, via a digital data communications network at a server computer system, an agricultural data record including a set of yield attributes for a set of seeds grown in a set of environments, wherein the set of yield attributes includes yield attributes generated by applying genetic relationship data between the seeds. The method further includes receiving, via the digital data communications network, characteristic data for one or more target fields in which the seeds are to be planted. The server computer system may then be used to generate seed recommendations for the one or more target fields based on the yield attributes and the set of characteristic data. Furthermore, the method may also include displaying the seed recommendations on a display device coupled to the server computer system.
[0069] 2. Example Agricultural Intelligent Computer System
[0070] 2.1 Structural Overview
[0071] Figure 1An example computer system is illustrated that is configured to perform the functions described herein, and is shown in a field environment along with other devices with which the system can interact. In one embodiment, a user 102 owns, operates, or has a field manager computing device 104 in a field location or associated with a field location, such as a field for agricultural activities or a management location for one or more agricultural fields. The field manager computer device 104 is programmed or configured to provide field data 106 to an agricultural intelligence computer system 130 via one or more networks 109.
[0072] Examples of field data 106 include: (a) identification data (e.g., area, field name, field identifier, geographic identifier, boundary identifier, crop identifier, and any other suitable data that can be used to identify a farm, such as common land unit (CLU), lot and block number, parcel number, geographic coordinates and boundaries, farm serial number (FSN), farm number, lot number, field number, region, township and / or range); (b) harvest data (e.g., crop type, crop variety, crop rotation, whether the crop is grown organically, harvest date, actual production history (APH), expected yield, yield, crop price, crop income, grain moisture, tillage practices, and previous growing season information); (c) soil data (e.g., type, composition, pH, organic matter (OM), cation exchange capacity (CEC)); (d) planting data (e.g., planting date, seed type, relative maturity (RM) of the seed planted, seed population); (e) fertilizer data (e.g., nutrient type (nitrogen, phosphorus, potassium), (e.g., type of application, date of application, amount, source, method); (f) chemical application data (e.g., pesticide, herbicide, fungicide, other substance or mixture of substances intended for use as a plant regulator, defoliant or desiccant, date of application, amount, source, method); (g) irrigation data (e.g., date of application, amount, source, method); (h) weather data (e.g., precipitation, rainfall rate, forecast rainfall, water runoff area, temperature, wind, forecast, pressure, visibility, clouds, heat index, dew point, humidity, snow depth, air quality, sunrise, sunset); (i) imagery data (e.g., images and spectral information from agricultural device sensors, cameras, computers, smartphones, tablets, unmanned aerial vehicles, aircraft, or satellites); (j) exploration observations (photos, videos, free-form annotations, audio recordings, speech transcriptions, weather conditions (temperature, precipitation (current or over time), soil moisture, crop growth stage, wind speed, relative humidity, dew point, terra firma)); and (k) soil, seed, crop phenology, pest and disease reports, and forecast sources and databases.
[0073] The data server computer 108 is communicatively coupled to the agricultural intelligent computer system 130 and is programmed or configured to send external data 110 to the agricultural intelligent computer system 130 via (one or more) networks 109. The external data server computer 108 may be owned or operated by the same legal person or entity as the agricultural intelligent computer system 130, or owned or operated by a different person or entity (such as a government agency, a non-governmental organization (NGO)) and / or a private data service provider. Examples of external data include weather data, image data, soil data, or statistics related to crop yields, etc. The external data 110 may include the same type of information as the field data 106. In some embodiments, the external data 110 is provided by an external data server 108 owned by the same entity that owns and / or operates the agricultural intelligent computer system 130. For example, the agricultural intelligent computer system 130 may include a data server that is specifically focused on data types (such as weather data) that can be obtained from third-party sources. In some embodiments, the external data server 108 may actually be incorporated into the system 130.
[0074] The agricultural device 111 may have one or more remote sensors 112 fixed thereon, which are directly or indirectly communicatively coupled to the agricultural intelligent computer system 130 via the agricultural device 111, and which are programmed or configured to send sensor data to the agricultural intelligent computer system 130. Examples of agricultural devices 111 include tractors, combines, harvesters, planters, trucks, fertilizer equipment, aircraft including unmanned aerial vehicles, and any other physical machinery or hardware, which is generally mobile machinery and can be used in tasks related to agriculture. In some embodiments, a single unit of the device 111 may include: multiple sensors 112, which are locally coupled in a network on the device; a controller area network CAN is an example of such a network that can be installed in a combine, a harvester, a sprayer, and a cultivator. The application controller 114 is communicatively coupled to the agricultural intelligent computer system 130 via (one or more) networks 109, and is programmed or configured to receive one or more scripts for controlling the operating parameters or implementations of the agricultural vehicle from the agricultural intelligent computer system. For example, a controller area network (CAN) bus interface may be used to enable communication from the agricultural intelligence computer system 130 to the agricultural device 111, such as how to use a climate vision field driver obtained from The Climate Corporation of San Francisco, California. The sensor data may include the same type of information as the field data 106. In some embodiments, the remote sensor 112 may not be fixed to the agricultural device 111, but may be remotely placed in the field and may communicate with the network 109.
[0075] The apparatus 111 may include a cab computer 115 programmed with a cab application, which may include a version or variation of a mobile application for the device 104 further described elsewhere herein. In an embodiment, the cab computer 115 includes a compact computer, typically a tablet computer or smart phone, with a graphical screen display, such as a color display, mounted in the cab of the apparatus 111. The cab computer 115 may implement some or all of the operations and functions further described herein for the mobile computer device 104.
[0076] The network(s) 109 broadly represents any combination of one or more data communication networks including a local area network, a wide area network, an internet or the Internet using wired or wireless links (including terrestrial or satellite links). Figure 1 Any medium or mechanism for data exchange between the various elements of the network(s) may be used. Figure 1 The various elements of the system may also have direct (wired or wireless) communication links. The sensors 112, controller 114, external data server computer 108 and other elements of the system each include an interface compatible with the network(s) 109 and are programmed or configured to use standardized protocols for communication across networks such as TCP / IP, Bluetooth, CAN protocols and higher layer protocols such as HTTP, TLS, etc.
[0077] The agricultural intelligence computer system 130 is programmed or configured to receive field data 106 from the field manager computing device 104, external data 110 from the external data server computer 108, and sensor data from the remote sensor 112. In a manner further described in other parts of this disclosure, the agricultural intelligence computer system 130 can be further configured to host, use or execute one or more computer programs, other software elements, digital programming logic (such as, FPGA or ASIC) or any combination thereof to perform conversion and storage of data values, construction of digital models of one or more crops on one or more fields, generation of recommendations and notifications, and generation and transmission of scripts to the application controller 114.
[0078] In an embodiment, the agricultural intelligence computer system 130 is programmed as or includes a communication layer 132, a presentation layer 134, a data management layer 140, a hardware / virtualization layer 150, and a model and field data repository 160. As used herein, a "layer" refers to any combination of electronic digital interface circuits, microcontrollers, firmware (such as drivers), and / or computer programs or other software elements.
[0079] The communication layer 132 may be programmed or configured to perform input / output interface functions, including sending requests for field data, external data, and sensor data, respectively, to the field manager computing device 104, the external data server computer 108, and the remote sensors 112. The communication layer 132 may be programmed or configured to send received data to the model and field data repository 160 for storage as the field data 106.
[0080] The presentation layer 134 may be programmed or configured to generate a graphical user interface (GUI) to be displayed on the field manager computing device 104, the cab computer 115, or other computer coupled to the system 130 via the network 109. The GUI may include controls for inputting data to be sent to the agricultural intelligence computer system 130, generating requests for models and / or recommendations, and / or displaying recommendations, notifications, models, and other field data.
[0081] The data management layer 140 may be programmed or configured to manage read operations and write operations involving the repository 160 and other functional elements of the system, including queries and result sets communicated between the system and the functional elements of the repository. Examples of the data management layer 140 include JDBC, SQL server interface code, and / or HADOOP interface code, etc. The repository 160 may include a database. As used herein, the term "database" may refer to a data body, a relational database management system (RDBMS), or both. As used herein, a database may include any collection of data, including: hierarchical databases, relational databases, flat file databases, object-relational databases, object-oriented databases, distributed databases, and any other structured collection of records or data stored in a computer system. Examples of RDBMS include, but are not limited to MYSQL, DB2, SQL SERVER, and POSTGRESQL databases. However, any database that enables the systems and methods described herein may be used.
[0082] When field data 106 is not provided directly to the agricultural intelligent computer system by one or more agricultural machinery or agricultural machinery equipment that interacts with the agricultural intelligent computer system, the user may be prompted to enter such information via one or more user interfaces on a user device (served by the agricultural intelligent computer system). In an example embodiment, the user can specify identification data by accessing a map on a user device (served by the agricultural intelligent computer system 130) and selecting a specific CLU that has been graphically shown on the map. In an alternative embodiment, the user 102 can specify identification data by accessing a map on a user device (served by the agricultural intelligent computer system 130) and drawing the boundaries of the field on the map. This CLU selection or map drawing represents a geographic identifier. In an alternative embodiment, the user can formulate identification data by accessing field identification data from the USDA Farm Service Agency or other sources (provided in a shape file or similar format) via a user device, and provide such field identification data to the agricultural intelligent computer system.
[0083] In an example embodiment, the agricultural intelligence computer system 130 is programmed to generate and display a graphical user interface including a data manager for data input. After one or more fields have been identified using the above method, the data manager can provide one or more graphical user interface widgets that, when selected, can identify changes in fields, soils, crops, tillage or nutrient practices. The data manager can include a timeline view, a spreadsheet view, and / or one or more editable programs.
[0084] Figure 5 Depicted is an example embodiment of a timeline view for data entry. Figure 5 , the user computer can input a selection of a specific field and a specific date for the addition of an event. The events depicted at the top of the timeline can include nitrogen, planting, practices, and soil. In order to add a nitrogen application event, the user computer can provide input to select a nitrogen tag. The user computer can then select a location on the timeline for a specific field to indicate the application of nitrogen on the selected field. In response to the received selection of a location on the timeline for a specific field, the data manager can display a data entry overlay, allowing the user computer to enter data related to nitrogen application, planting procedures, soil application, tillage procedures, irrigation practices, or other information related to the specific field. For example, if the user computer selects a portion of the timeline and indicates the application of nitrogen application rate, the data entry overlay can include a field for entering the nitrogen application rate, the application date, the type of fertilizer used, and any other information related to the application of nitrogen application rate.
[0085] In an embodiment, the data manager provides an interface for creating one or more programs. As used herein, a "program" refers to a collection of data relating to nitrogen applications, planting procedures, soil applications, tillage procedures, irrigation practices, or other information relating to one or more fields, and a program may be stored in a digital data store for reuse as a collection in other operations. After a program has been created, it may be conceptually applied to one or more fields, and a reference to the program may be stored in the digital storage in association with the data identifying the fields. Thus, instead of manually entering the same data relating to the same nitrogen application for multiple different fields, a user computer may create a program indicating a particular application of nitrogen rates, and then apply the program to multiple different fields. For example, in Figure 5 In the timeline view of FIG. 1 , the top two timelines have selected a "Spring Application" program that includes applying 150 lbs. N / ac in early April. The data manager may provide an interface for editing a program. In an embodiment, when editing a particular program, each field for which the particular program has been selected is edited. For example, in Figure 5 If the "Spring Application" program is edited to reduce the nitrogen application rate to 130 lbs.N / ac, the top two fields can be updated by reducing the nitrogen application rate based on the edited program.
[0086] In an embodiment, in response to receiving an edit to a field having a selected program, the data manager removes the correspondence between the field and the selected program. Figure 5 If a nitrogen application is added to the top field of the , the interface can be updated to indicate that the "Spring Application" program is no longer applied to the top field. Although the early April nitrogen application can be retained, the update of the "Spring Application" program will not change the April application of nitrogen application rates.
[0087] Figure 6 Depicted is an example embodiment of a spreadsheet view for data entry. Figure 6 The display depicted in FIG. 10 allows a user to create and edit information for one or more fields. The data manager may include a spreadsheet for inputting Figure 6 The nitrogen, planting, practice and soil related information depicted in the spreadsheet. To edit a specific entry, the user computer can select the specific entry in the spreadsheet and update the value. For example, Figure 6 The ongoing update of the target yield value for the second field is depicted. Additionally, the user computer can select one or more fields for application of one or more programs. In response to receiving a selection of a program for a particular field, the data manager can automatically complete the input for the particular field based on the selected program. Figure 1Similarly, the data manager can update the inputs of each field associated with a particular program in response to a received update to the program. Additionally, the data manager can remove the correspondence of the selected program to the field in response to receiving an edit to one of the inputs to the field.
[0088] In an embodiment, the model and field data are stored in the model and field data repository 160. The model data includes a data model created for one or more fields. For example, a crop model may include a digitally constructed model of the development of crops on one or more fields. In this article, "model" refers to a collection of executable instructions and data values associated with each other that are stored electronically and digitally, and these executable instructions and data values are able to receive and respond to programmed calls or other digital calls, calls or parsing requests to generate one or more stored output values or calculated output values, which can be used as the basis for computer-implemented recommendations, output data displays or machine control, etc. Those skilled in the art find it convenient to express the model using mathematical formulas, but this form of expression does not limit the models disclosed herein to abstract concepts; on the contrary, each model herein has practical applications in a computer in the form of stored executable instructions and data, which use a computer to implement the model. The model may include a model of past events on one or more fields, a model of the current state of one or more fields, and / or a model of predicted events on one or more fields. The model and field data may be stored in a data structure in memory, in rows in a database table, in a flat file or spreadsheet, or other form of stored digital data.
[0089] In an embodiment, the seed classification subsystem 170 includes specially configured logic, including but not limited to seed standardization instructions 172, success probability or predicted yield performance generation instructions 174, and yield classification instructions 176, the yield classification instructions 176 include a collection of one or more pages of main memory (such as, RAM), executable instructions have been loaded in the agricultural intelligent computer system 130, and when the executable instructions are executed, the agricultural intelligent computing system performs the functions or operations described herein with reference to those modules. In an embodiment, the seed recommendation subsystem 180 includes specially configured logic, including but not limited to seed filtering instructions 182, risk generation instructions 184, and optimized classification instructions 186, the optimized classification instructions 186 include a collection of one or more pages of main memory (such as, RAM), executable instructions have been loaded in the agricultural intelligent computer system 130, and when the executable instructions are executed, the agricultural intelligent computing system performs the functions or operations described herein with reference to those modules. For example, the seed standardization instructions 172 may include a collection of pages in the RAM containing instructions, which when executed, cause the target identification function described herein to be performed. The instruction can be in a machine executable code in the instruction set of the CPU, and can be compiled based on source code written in JAVA, C, C++, OBJECTIVE-C or any other human-readable programming language or environment, either alone or in combination with scripts in JAVASCRIPT, other scripting languages and other programming source texts. The term "page" is intended to refer broadly to any area within the main memory, and the specific terms used in the system can vary according to the memory architecture or processor architecture. In another embodiment, each of the seed standardization instructions 172, the success probability or predicted yield performance generation instructions 174, the yield classification instructions 176, the seed filtering instructions 182, the risk generation instructions 184 and the optimized classification instructions 186 can also represent one or more files or source code projects digitally stored in a mass storage device (such as, non-volatile RAM or disk storage) in the agricultural intelligent computer system 130 or a separate repository system, which, when compiled or interpreted, result in the generation of executable instructions, which, when executed, result in the agricultural intelligent computing system performing the functions or operations described herein with reference to those modules. In other words, the figures may represent the manner in which a programmer or software developer organizes and arranges source code for later execution by the agricultural intelligence computer system 130 so that it may be compiled into an executable file or interpreted into byte code or equivalent code.
[0090] For example, combining Figure 4As shown and described, the hardware / virtualization layer 150 includes one or more central processing units (CPUs), memory controllers, and other devices, components, or elements of a computer system, such as volatile or non-volatile memory, non-volatile storage (such as disks), and I / O devices or interfaces. Layer 150 may also include programming instructions configured to support virtualization, containerization, or other technologies.
[0091] For the purpose of illustrating a clear example, Figure 1 A limited number of instances of certain functional elements are illustrated. However, in other embodiments, there may be any number of such elements. For example, an embodiment may use thousands or millions of different mobile computing devices 104 associated with different users. In addition, the system 130 and / or the external data server computer 108 may be implemented using two or more processors, cores, clusters, or instances of physical or virtual machines, which may be configured in discrete locations or co-located with other elements in a data center, shared computing facility, or cloud computing facility.
[0092] 2.2 Application Overview
[0093] In an embodiment, the functions described herein are implemented using one or more computer programs or other software elements loaded into and executed by one or more general-purpose computers, which will cause the general-purpose computers to be configured as specific machines or computers that are particularly suitable for performing the functions described herein. In addition, each of the flowcharts further described herein can be used alone or in combination with the descriptions of the processes and functions herein as algorithms, plans or directions that can be used to program a computer or logic to implement the described functions. In other words, all narrative text and all appended Figure 1 It is intended to engage the skills and knowledge of a skilled person, and to enable him or her to have a skill level applicable to such invention and disclosure, to provide a disclosure of an algorithm, plan or direction sufficient to allow the skilled person to program a computer to perform the functions described herein.
[0094] In an embodiment, a user 102 interacts with an agricultural intelligence computer system 130 using a field manager computing device 104 configured with an operating system and one or more applications or applications; the field manager computing device 104 may also interoperate with the agricultural intelligence computer system independently and automatically under program control or logic control, and does not always require direct user interaction. The field manager computing device 104 broadly represents one or more devices in a smart phone, PDA, tablet computing device, laptop computer, desktop computer, workstation, or any other computing device capable of sending and receiving information and performing the functions described herein. The field manager computing device 104 may communicate via a network using a mobile application stored on the field manager computing device 104, and in some embodiments, the device may be coupled to a sensor 112 and / or a controller 114 using a cable 113 or connector. A particular user 102 may own, operate, or have and use multiple field manager computing devices 104 associated with the system 130 at one time.
[0095] The mobile application may provide client functionality to one or more mobile computing devices via a network. In an example embodiment, the field manager computing device 104 may access the mobile application via a web browser or a local client application or application. The field manager computing device 104 may send data to one or more front-end servers and receive data from one or more front-end servers using a web-based protocol or format (such as HTTP, XML and / or JSON) or an application-specific protocol. In an example embodiment, the data may take the form of a request and user information input (such as field data) into the mobile computing device. In some embodiments, the mobile application interacts with location tracking hardware and software on the field manager computing device 104, which uses standard tracking technology (such as multilateration of radio signals, global positioning system (GPS), WiFi positioning system or other methods of mobile positioning) to determine the location of the field manager computing device 104. In some cases, location data or other data associated with the device 104, the user 102 and / or (one or more) user accounts may be obtained by querying the device's operating system or by requesting an application on the device to obtain data from the operating system.
[0096] In an embodiment, the field manager computing device 104 sends the field data 106 to the agricultural intelligence computer system 130, which includes or includes, but is not limited to, data values representing one or more of the following information: geographic location of one or more fields; farming information for one or more fields; crops planted in one or more fields; and soil data extracted from one or more fields. The field manager computing device 104 can send the field data 106 in response to user input from the user 102 specifying data values for one or more fields. Additionally, the field manager computing device 104 can automatically send the field data 106 when one or more of the data values become available to the field manager computing device 104. For example, the field manager computing device 104 can be communicatively coupled to the remote sensor 102 and / or the application controller 114, which includes an irrigation sensor and / or an irrigation controller. In response to receiving data indicating that the application controller 114 released water onto one or more fields, the field manager computing device 104 may send field data 106 indicating that water was released onto the one or more fields to the agricultural intelligence computer system 130. The field data 106 identified in the present disclosure may be input and communicated using electronic digital data that is communicated between computing devices using parameterized URLs via HTTP or another suitable communication or messaging protocol.
[0097] A commercial example of a mobile application is CLIMATE FIELDVIEW, commercially available from The Climate Corporation of San Francisco, California. The CLIMATE FIELDVIEW application or other applications may be modified, expanded, or adapted to include features, functions, and programs that have not been disclosed prior to the filing date of this disclosure. In one embodiment, the mobile application includes an integrated software platform that allows growers to make fact-based decisions for their operations because it combines historical data about the grower's fields with any other data the grower wishes to compare. The combination and comparison can be done in real time and is based on scientific models that provide possible scenarios so that the grower can make better and more informed decisions.
[0098] Figure 2 Two views of an example logical organization of a set of instructions in main memory when an example mobile application is loaded for execution are illustrated. Figure 2In one embodiment, in view (a), the mobile computer application 200 includes account field data ingestion sharing instructions 202, overview and alert instructions 204, digital map manual instructions 206, seed and planting instructions 208, nitrogen application instructions 210, weather instructions 212, field health instructions 214, and performance instructions 216.
[0099] In one embodiment, the mobile computer application 200 includes account, field, data ingestion, sharing instructions 202, which are programmed to receive, translate and ingest field data from a third-party system via manual upload or API. Data types may include field boundaries, yield maps, planting maps, soil test results, application maps and / or management zones, etc. Data formats may include shape files, third-party native data formats and / or farm management information system (FMIS) exports, etc. Receiving data may occur via manual upload, email with attachments, external APIs that push data to the mobile application, or instructions that call an external system's API to pull data into the mobile application. In one embodiment, the mobile computer application 200 includes a data inbox. In response to a received selection of the data inbox, the mobile computer application 200 may display a graphical user interface for manually uploading data files and importing the uploaded files into the data manager.
[0100] In one embodiment, the digital map book instructions 206 include a field map data layer stored in the device memory, and the field map data layer is programmed with data visualization tools and geographic field annotations. This provides growers with convenient information for reference, record keeping, and intuitive insights into field performance. In one embodiment, the overview and alert instructions 204 are programmed to provide a full operational view of what is important to the grower and provide timely recommendations to take action or focus on specific issues. This allows growers to focus their time on where attention is needed to save time and maintain yields throughout the season. In one embodiment, the seed and planting instructions 208 are programmed to provide tools for seed selection, seed placement, and script creation based on scientific models and empirical data, including variable rate (VR) script creation. This allows growers to maximize yield or return on investment by optimizing seed purchases, placement, and populations.
[0101] In one embodiment, the script generation instructions 205 are programmed to provide an interface for generating scripts including variable rate (VR) productivity scripts. The interface enables growers to create scripts for field implementations, such as nutrient application, planting and irrigation. For example, the planting script interface may include a tool for identifying the seed type used for planting. After receiving the selection of the seed type, the mobile computer application 200 may display one or more fields subdivided into management zones, such as a field map data layer created as part of the digital map manual instructions 206. In one embodiment, the management zones include soil zones and a panel identifying each soil zone and soil name, texture, drainage or other field data for each zone. The mobile computer application 200 may also display tools for editing or creation on a map of one or more fields, such as a graphical tool for drawing management zones (such as soil zones). Planting processes may be applied to all management zones, or different planting processes may be applied to different subsets of management zones. When creating a script, the mobile computer application 200 may make the script available for download in a format readable by the application controller, such as an archive format or a compressed format. Additionally and / or alternatively, the script may be sent from the mobile computer application 200 directly to the cab computer 115 and / or uploaded to one or more data servers and stored for further use.
[0102] In one embodiment, the nitrogen application instructions 210 are programmed to provide tools to inform nitrogen application decisions by visualizing the availability of nitrogen application in the crop. This enables growers to maximize yield or return on investment by optimizing nitrogen application during the season. Example programming functions include: displaying images (such as SSURGO images) at high spatial resolution (which can be millimeters or less, depending on the proximity and resolution of the sensor) to enable the mapping of fertilizer application zones and / or images generated from sub-field soil data (such as data acquired from sensors); uploading existing grower-defined zones; providing charts and / or maps of plant nutrient availability to enable adjustment of nitrogen application across (one or more) zones; outputting scripts to drive machinery; tools for large data entry and adjustment; and / or maps for data visualization, etc. In this article, "large data entry" can mean entering data once and then applying the same data to multiple fields and / or zones that have been defined in the system; example data can include the same nitrogen application data for many fields and / or zones for the same grower, but such large data entry is applicable to entering any type of field data into the mobile computer application 200. For example, nitrogen application instructions 210 may be programmed to accept definitions of nitrogen application and practice programs and to accept user input specifying those programs to be applied across multiple fields. As used herein, a "nitrogen application program" refers to a named collection of stored data associated with: a name, color code or other identifier, one or more application dates, material or product type for each of the dates and amounts, a method of application or incorporation (such as injection or planting), and / or the date that is the subject of the application, an application amount or application rate for each of the crops or hybrids / varieties, etc. As used herein, a "nitrogen practice program" refers to a named collection of stored data associated with: a practice name; a previous crop; a tillage system; a primary tillage date; one or more previous tillage systems used; an indicator of one or more application types (such as fertilizers) used. Nitrogen application instructions 210 may also be programmed to generate and cause the display of a nitrogen application map indicating a prediction of plant use of a specified nitrogen application amount and whether a surplus or deficit is predicted; in some embodiments, different color indicators may represent a surplus or deficit. In one embodiment, a nitrogen application map includes: a graphical display in a computer display device, the computer display device including a plurality of rows, each row associated with and identifying a field; data specifying which crop is planted in the field, field size, field location, and a graphical representation of the field perimeter; in each row, a timetable for each month with a graphical indicator specifying each nitrogen application and amount associated with the name of the month; and a numerical and / or colored surplus or deficit indicator, wherein the color indicates the magnitude.
[0103] In one embodiment, the nitrogen application map may include one or more user input features (such as a dial or a slider) to dynamically change the nitrogen application planting and practice program so that the user can optimize his nitrogen application map. The user can then use his optimized nitrogen application map and the associated nitrogen application planting and practice program to implement one or more scripts, including a variable rate (VR) productivity script. The nitrogen application instructions 210 can also be programmed to generate and display a nitrogen application map that indicates the forecast of the use of the specified planting nitrogen application amount and whether a surplus or shortage is predicted; in some embodiments, different color indicators can represent the surplus or shortage. The nitrogen application map can display the forecast of the use of the specified planting nitrogen application amount and whether a surplus or shortage is predicted at different times in the past and future (such as daily, weekly, monthly, or annually), where the color indicates the magnitude. In one embodiment, the nitrogen application map may include one or more user input features, such as a dial or a slider, to dynamically change the nitrogen application planting and practice program so that the user can optimize his nitrogen application map, such as to obtain a preferred surplus to a preferred shortage. The user can then use his optimized nitrogen application map and associated nitrogen application planting and practice programs to implement one or more scripts, including variable rate (VR) productivity scripts. In other embodiments, instructions similar to nitrogen application instructions 210 can be used for the application of other nutrients (such as phosphorus and potassium), the application of pesticides, and irrigation programs.
[0104] In one embodiment, weather instructions 212 are programmed to provide field-specific recent weather data and forecast weather information. This allows growers to save time and have an efficient integrated display in daily operational decisions.
[0105] In one embodiment, field health instructions 214 are programmed to provide timely remote sensing images that highlight timely crop changes and potential problems. Example programmed functions include: cloud checks to identify possible clouds or cloud shadows; determining nitrogen index based on field images; graphical visualization of survey layers, such as those related to field health, and viewing and / or sharing survey notes; and / or downloading satellite imagery from multiple sources and prioritizing images for growers, etc.
[0106] In one embodiment, the performance instructions 216 are programmed to use the data on the farm to provide reporting, analysis and insight tools for evaluation, insight and decision making. This enables growers to seek improved results for the next year based on factual conclusions (why the return on investment is at a previous level) and insights into yield limiting factors. The performance instructions 216 can be programmed to communicate with the back-end analysis program executed at the agricultural intelligence computer system 130 and / or the external data server computer 108 via (one or more) networks 109, and the performance instructions 216 are configured to analyze metrics such as yield, yield differences, hybrids / variety, populations, SSURGO zones, soil test attributes or altitude, etc. Programmatic reporting and analysis can include: yield variability analysis, treatment effect estimation, benchmarking yield and other indicators against other growers based on anonymous data collected from many growers or data for seeds and plantings, etc.
[0107] An application with instructions configured in this manner can be implemented for different computing device platforms while retaining the same general user interface appearance. For example, a mobile application can be programmed for execution on a tablet computer, a smart phone, or a server computer accessed using a browser at a client computer. In addition, a mobile application configured for a tablet computer or a smart phone can provide a complete application experience or a cockpit application experience that is suitable for the display and processing capabilities of the cockpit computer 115. For example, referring now to Figure 2For view (b), in one embodiment, the cab computer application 220 may include: map cab instructions 222, remote viewing instructions 224, data collection and transmission instructions 226, machine alarm instructions 228, script transmission instructions 230, and exploration cab instructions 232. The code base for the instructions for view (b) may be the same as the code base for view (a), and the executable files implementing the code may be programmed to detect the type of platform on which they are executed and expose only those products that are applicable to the cab platform or the complete platform's functionality through a graphical user interface. This approach enables the system to recognize the distinct user experiences that are applicable to the different technical environments of the in-cab environment and the cab. The map cab instructions 222 may be programmed to provide a map view of a field, farm, or region that can be used to guide machine operations. The remote viewing instructions 224 may be programmed to connect, manage, and provide machine activity views to other computing devices connected to the system 130 in real time or near real time via a wireless network, a wired connector or adapter, etc. Data collection and transmission instructions 226 may be programmed to connect, manage, and transmit data collected at sensors and controllers to the system 130 via a wireless network, a wired connector or adapter, etc. Machine alert instructions 228 may be programmed to detect operational issues with a machine or implement associated with the cab and generate an operator alert. Script transmission instructions 230 may be configured to transmit in the form of an instruction script configured to direct the collection of machine operations or data. Survey cab instructions 232 may be programmed to display location-based alerts and information received from the system 130 based on the location of the field manager computing device 104, agricultural device 111, or sensor 112 in the field, and to ingest, manage, and provide location-based survey observations to the system 130 based on the location of the agricultural device 111 or sensor 112 in the field.
[0108] 2.3 Data ingested into computer systems
[0109] In an embodiment, the external data server computer 108 stores external data 110, which includes soil data representing the soil composition of one or more fields and weather data representing the temperature and precipitation of one or more fields. The weather data may include past and current weather data and predictions of future weather data. In an embodiment, the external data server computer 108 includes multiple servers hosted by different entities. For example, a first server may contain soil composition data, while a second server may include weather data. Additionally, the soil composition data may be stored in multiple servers. For example, one server may store data representing the percentage of sand, silt, and clay in the soil, while a second server may store data representing the percentage of organic matter (OM) in the soil.
[0110] In an embodiment, the remote sensor 112 includes one or more sensors programmed or configured to generate one or more observations. The remote sensor 112 can be an aerial sensor, such as a satellite, a vehicle sensor, a planting equipment sensor, a tillage sensor, a fertilizer or pesticide application sensor, a harvester sensor, and any other device capable of receiving data from one or more fields. In an embodiment, the application controller 114 is programmed or configured to receive instructions from the agricultural intelligent computer system 130. The application controller 114 can also be programmed or configured to control the operating parameters of agricultural vehicles or appliances. For example, the application controller can be programmed or configured to control the operating parameters of a vehicle (such as a tractor), a planting equipment, a tillage equipment, a fertilizer or pesticide equipment, a harvester equipment, or other farmland appliances (such as a water valve). Other embodiments can use any combination of sensors and controllers, and the following embodiments are only selected examples.
[0111] The system 130 may acquire or ingest data on a large scale from a large number of growers that have shared data to the shared database system under the control of the user 102. This form of acquiring data may be referred to as "manual data ingestion" when one or more user-controlled computer operations are requested or triggered to acquire data for use by the system 130. For example, the Climate Horizon application purchased from the Climate Corporation of San Francisco, California may be operated to export data to the system 130 for storage in the repository 160.
[0112] For example, a seed monitor system can both control planter assembly components and acquire planting data including signals from seed sensors via a signal harness including a CAN backbone and point-to-point connections for registration and / or diagnostics. The seed monitor system can be programmed or configured to display seed spacing, population, and other information to a user via the cab computer 115 or other device within the system 130. Examples are disclosed in U.S. Patent No. 8,738,243 and U.S. Patent Publication 20150094916, and the present disclosure assumes knowledge of those other patent disclosures.
[0113] Likewise, the yield monitor system may include yield sensors for the harvester assembly that transmit yield measurement data to the cab computer 115 or other device within the system 130. The yield monitor system may utilize one or more remote sensors 112 to obtain grain moisture measurements in a combine or other harvester and transmit these measurements to a user via the cab computer 115 or other device within the system 130.
[0114] In embodiments, examples of sensors 112 that may be used with any type of mobile vehicle or device described elsewhere herein include kinematic sensors and position sensors. Kinematic sensors may include any speed sensor, such as a radar or wheel speed sensor, an accelerometer, or a gyroscope. Position sensors may include: a GPS receiver or transceiver; or a WiFi-based location or mapping application that is programmed to determine location based on nearby WiFi hotspots, etc.
[0115] In an embodiment, examples of sensors 112 that may be used with a tractor or other moving vehicle include: an engine speed sensor, a fuel consumption sensor, an area counter or distance counter that interacts with a GPS or radar signal, a PTO (power take-off) speed sensor, a tractor hydraulic sensor configured to detect a hydraulic parameter such as pressure or flow and / or hydraulic pump speed, a wheel speed sensor, or a wheel slip sensor. In an embodiment, examples of controllers 114 that may be used with a tractor include: a hydraulic directional controller, a pressure controller, and / or a flow controller; a hydraulic pump speed controller; a speed controller or governor; a suspension position controller; or a wheel position controller that provides automatic steering.
[0116] In an embodiment, examples of sensors 112 that can be used with seed planting equipment (such as, a planter, drill, or air seeder) include: a seed sensor, which can be an optical sensor, an electromagnetic sensor, or a collision sensor; a downforce sensor, such as a weighing pin, a weighing cell, a pressure sensor; a soil property sensor, such as a reflectivity sensor, a moisture sensor, a conductivity sensor, an optical residue sensor, or a temperature sensor; a component operation standard sensor, such as a planting depth sensor, a downforce cylinder pressure sensor, a seed tray speed sensor, a seed drive motor encoder, a seed delivery system speed sensor, or a vacuum sensor; or a pesticide application sensor, such as an optical sensor or other electromagnetic sensor; or a collision sensor. In an embodiment, examples of controllers 114 that may be used with such seed planting equipment include: a toolbar fold controller, such as a controller for a valve associated with a hydraulic cylinder; a down force controller, such as a controller for a valve associated with a pneumatic cylinder, an airbag, or a hydraulic cylinder, and programmed to apply down force to a single row unit or an entire planter frame; a planting depth controller, such as a linear actuator; a metering controller, such as an electric seed metering drive motor, a hydraulic seed metering drive motor, or a strip control clutch; a hybrid / variety selection controller, such as a seed metering drive motor or other actuator programmed to selectively allow or prevent seed or an air-seed mixture from delivering seed to a seed meter or a central bulk hopper; a metering controller, such as an electric seed metering drive motor, or a hydraulic seed metering drive motor; a seed delivery system controller, such as a controller for a belt seed delivery motor; or a pesticide application rate controller, such as a metering drive controller, an orifice size or position controller.
[0117] In an embodiment, examples of sensors 112 that can be used with a tillage device include: a position sensor for a tool (such as a nozzle bar or disc); a tool position sensor for such a tool configured to detect depth, engagement angle, or lateral spacing; a down force sensor; or a traction force sensor. In an embodiment, examples of controllers 114 that can be used with a tillage device include a down force controller or a tool position controller, such as a controller configured to control tool depth, engagement angle, or lateral spacing.
[0118] In embodiments, examples of sensors 112 that may be associated with an apparatus for applying fertilizers, pesticides, fungicides, and the like (such as a starter fertilizer system on a planter, a subsoil fertilizer applicator, or a fertilizer sprayer) include: a fluid system standard sensor, such as a fluid sensor or a pressure sensor; a sensor indicating which sprinkler valve or fluid line valve is open; a sensor associated with a tank, such as a fill level sensor; a segment or system-wide supply line sensor, or a row-specific supply line sensor; or a kinematic sensor (such as an accelerometer disposed on a sprayer boom). In embodiments, examples of controllers 114 that may be used with such an apparatus include: a pump speed controller; a valve controller programmed to control pressure, flow, direction, PWM, and the like; or a position actuator, such as boom height, subsoil depth, or boom position.
[0119] In embodiments, examples of sensors 112 that may be used with a harvester include: a yield monitor, such as a strike plate strain gauge or position sensor, a capacitive flow sensor, a load sensor, a weight sensor, or a torque sensor associated with an elevator or auger, or an optical or other electromagnetic grain height sensor; a grain moisture sensor, such as a capacitive sensor; a grain loss sensor, including a collision, optical, or capacitive sensor; a header operation standard sensor, such as a header height, header type, cover clearance, feed speed, and reel speed sensor; a separator operation standard sensor, such as a chock clearance, rotor speed, shoe clearance, or shroud clearance sensor; an auger sensor for position, operation, or speed; or an engine speed sensor. In embodiments, examples of controllers 114 that may be used with a harvester include: a header operation standard controller for an element, such as header height, header type, cover clearance, feed speed, or reel speed; a separator operation standard controller for a feature, such as a chock clearance, rotor speed, shoe clearance, or shroud clearance; or a controller for auger position, operation, or speed.
[0120] In an embodiment, examples of sensors 112 that may be used with a grain cart include: a weight sensor, or a sensor for auger position, operation, or speed. In an embodiment, examples of controllers 114 that may be used with a grain cart include: a controller for auger position, operation, or speed.
[0121] In an embodiment, examples of sensors 112 and controllers 114 may be installed in an unmanned aerial vehicle (UAV) device or "drone." Such sensors may include: a camera with a detector that is effective in any range of the electromagnetic spectrum including visible light, infrared, ultraviolet, near infrared (NIR), etc.; an accelerometer; an altimeter; a temperature sensor; a humidity sensor; a pitot tube sensor or other airspeed or wind speed sensor; a battery life sensor; or a radar transmitter and a reflected radar energy detection device; other electromagnetic radiation transmitters and reflected electromagnetic radiation detection devices. Such controllers may include: a guidance or motor control device, a control surface controller, a camera controller, or a controller programmed to turn on, operate, obtain data from, manage, and configure any of the aforementioned sensors. Examples are disclosed in U.S. Patent Application No. 14 / 831,165, and the present disclosure assumes knowledge of the other patent disclosures.
[0122] In an embodiment, the sensor 112 and the controller 114 may be secured to a soil sampling and measuring device that is configured or programmed to sample soil and perform soil chemical tests, soil moisture tests, and other tests related to the soil. For example, the devices disclosed in U.S. Pat. No. 8,767,194 and U.S. Pat. No. 8,712,148 may be used, and the present disclosure assumes knowledge of those patent disclosures.
[0123] In an embodiment, the sensor 112 and the controller 114 may include a weather device for monitoring weather conditions of the field. For example, the devices disclosed in U.S. Provisional Application No. 62 / 154,207 filed on April 29, 2015, U.S. Provisional Application No. 62 / 175,160 filed on June 12, 2015, U.S. Provisional Application No. 62 / 198,060 filed on July 28, 2015, and U.S. Provisional Application No. 62 / 220852 filed on September 18, 2015 may be used, and the present disclosure assumes knowledge of those patent disclosures.
[0124] 2.4 Process Overview - Agronomic Model Training
[0125] In an embodiment, the agricultural intelligent computer system 130 is programmed or configured to create an agronomic model. In this article, an agronomic model is a data structure in the memory of the agricultural intelligent computer system 130, and the data structure includes field data 106, such as identification data and harvest data for one or more fields. The agronomic model can also include a calculated agronomic attribute, which describes the conditions or attributes of one or more crops that can affect the growth of one or more crops on the field or both. Additionally, the agronomy model can include recommendations based on agronomic factors, such as crop recommendations, irrigation recommendations, planting recommendations, fertilizer recommendations, fungicide recommendations, pesticide recommendations, harvest recommendations and other crop management recommendations. Agronomic factors can also be used to estimate the results related to one or more crops, such as agronomic yield. The agronomic yield of a crop is an estimate of the quantity of the crop produced, or in some examples, is the income or profit obtained from the crop produced.
[0126] In an embodiment, the agricultural intelligence computer system 130 can use a preconfigured agronomic model to calculate agronomic attributes related to the location and crop information of one or more fields currently received. The preconfigured agronomic model is based on previously processed field data, including but not limited to identification data, harvest data, fertilizer data, and weather data. The preconfigured agronomic model may have been cross-validated to ensure the accuracy of the model. Cross-validation may include a comparison with ground truth that compares the predicted results with the actual results on the field, such as comparing rainfall estimates with sensors that use rain gauges or provide weather data at the same location or nearby locations, or comparing nitrogen content estimates with soil sampling measurements.
[0127] Figure 3 Illustrated is a programmed process by which an agricultural intelligence computer system generates one or more pre-configured agronomic models using field data provided by one or more data sources. Figure 3 Algorithms or instructions may be used to program the functional elements of the agricultural intelligence computer system 130 to perform the operations now described.
[0128] At box 305, the agricultural intelligent computer system 130 is configured or programmed to perform agronomic data preprocessing on the field data received from one or more data sources. The field data received from one or more data sources can be preprocessed to eliminate noise, distortion effects, and confounding factors within the agronomic data including outliers that may adversely affect the received field data values. Examples of agronomic data preprocessing include, but are not limited to: removing data values that are typically associated with outlier data values, specific measurement data points that are known to unnecessarily skew other data values, data smoothing, aggregation, or sampling techniques used to remove or reduce additive or multiplicative effects of noise, and other filtering or data derivation techniques for clearly distinguishing between positive data inputs and negative data inputs.
[0129] At box 310, the agricultural intelligent computer system 130 is configured or programmed to perform data subset selection using the preprocessed field data to identify a data set useful for initial agronomic model generation. The agricultural intelligent computer system 130 that can implement data subset selection techniques includes, but is not limited to: genetic algorithm methods, methods of all subset models, sequential search methods, stepwise regression methods, particle swarm optimization methods, and ant colony optimization methods. For example, the genetic algorithm selection technique is based on the evolutionary principles of natural selection and genetics, and uses an adaptive heuristic search algorithm to determine and evaluate data sets within the preprocessed agronomic data.
[0130] At box 315, the agricultural intelligent computer system 130 is configured or programmed to implement field data set evaluation. In an embodiment, a specific field data set is evaluated by creating an agronomic model and using a specific quality threshold of the created agronomic model. One or more comparison techniques can be used to compare and / or verify the agronomic model, such as, but not limited to, root mean square error and leave-one-out cross validation (RMSECV), mean absolute error, and mean percentage error. For example, RMSECV can cross-validate the agronomic model by comparing the predicted agronomic attribute values created by the agronomic model with the historical agronomic attribute values collected and analyzed. In an embodiment, the agronomic data set evaluation logic is used as a feedback loop, where an agronomic data set that does not meet the configured quality threshold is used during a future data subset selection step (box 310).
[0131] At block 320, the agricultural intelligence computer system 130 is configured or programmed to implement agronomic model creation based on the cross-validated agronomic data set. In an embodiment, the agronomic model creation may implement a multivariate regression technique to create a pre-configured agronomic data model.
[0132] At box 325 , the agricultural intelligence computer system 130 is configured or programmed to store preconfigured agronomic data models for use in future field data evaluations.
[0133] 2.5 Seed classification subsystem
[0134] In an embodiment, the agricultural intelligent computer system 130 includes, among other components, a seed classification subsystem 170. The seed classification subsystem 170 is configured to generate a target successful yield group of seeds, which is specifically identified for the best attributes of the target field. As used herein, the term "optimal" and related terms (e.g., "optimized," "optimization," etc.) are broad terms that refer to improvements in "best or most effective" ("general optimization") and "better or more effective" ("relative optimization") for any results, systems, data, etc. The target successful yield group includes: a subset of one or more seeds, an estimated yield prediction for each seed, and the probability of successfully exceeding the average estimated yield prediction of similarly classified seeds.
[0135] In an embodiment, seeds that will perform best on a target field are identified based on input received by the agricultural intelligence computer system 130, the input including, but not limited to, agricultural data records for a plurality of different seeds and geographic location data related to the field where the agricultural data records were collected. For example, if agricultural data records for one hundred seeds are received, the agricultural data records will include growth and yield data for the one hundred seeds and geographic location data about the field where the one hundred seeds were planted. In an embodiment, the agricultural intelligence computer system 130 also receives geographic location and agricultural data for a second set of fields. The second set of fields is the target field where the grower intends to plant the selected seeds. Information about the target field is particularly relevant to matching a particular seed to the environment of the target field.
[0136] Seed standardization instructions 172 provide instructions to generate a data set of seed attributes that describes representative yield values and environmental classifications associated with preferred environmental conditions for each of the seeds received by the agricultural intelligence computer system 130. Probability of successful generations instructions 174 provide instructions to generate a data set of success probability scores associated with each of the seeds. The success probability score describes the probability of a successful yield on a target field. Yield classification instructions 176 provide instructions to generate a target successful yield group of seeds that have been identified as having the best attributes on a target field based on the success probability scores associated with each of the seeds.
[0137] In an embodiment, agricultural intelligence computer system 130 is configured to present target successful yield groups of selected seeds and their normalized yield values and success probability scores via presentation layer 134 .
[0138] Seed classification subsystem 170 and related instructions are additionally described elsewhere herein.
[0139] 2.6 Seed Recommendation Subsystem
[0140] In an embodiment, the agricultural intelligence computer system 130 includes, among other components, a seed recommendation subsystem 180. The seed recommendation subsystem 180 is configured to generate a target seed set that is specifically selected for optimal performance to minimize or reduce risk on a target field. The target seed set includes a subset of one or more seeds whose estimated yield forecasts are above a specific yield threshold and whose associated risk values are below a specific risk target.
[0141] In an embodiment, a target seed set that will perform best on a target field is identified based on a seed input set that has been identified as having a specific probability of producing a successful yield on the target field. The agricultural intelligent computer system 130 can be configured to receive a seed set as part of a target successful yield group generated by the seed classification subsystem 170. The target successful yield group can also include agricultural data specifying the probability of success for each seed and other agricultural data (such as yield values, relative maturity, and environmental observations in previously observed harvests). In an embodiment, the agricultural intelligent computer system 130 also receives the geographic location and agricultural data for the set of target fields. A "target field" is a field where a grower is considering or intends to plant target seeds.
[0142] Seed filtering instructions 182 provide instructions to filter and identify a subset of seeds having a probability of success value that exceeds a specified success yield threshold. Risk generation instructions 184 provide instructions to generate a data set of risk values associated with each of the seeds. The risk value describes the amount of risk associated with each seed relative to the estimated yield value of each seed. Optimize classification instructions 186 provide instructions to generate a data set of target seeds from the data set of risk values, the average yield value of the target seeds being above a target threshold for a range of risk values.
[0143] In an embodiment, agricultural intelligence computer system 130 is configured to present the target seed set via presentation layer 134 and include their average yield values.
[0144] Seed recommendation subsystem 180 and related instructions are additionally described elsewhere herein.
[0145] 2.7 Implementation Example—Hardware Overview
[0146] According to one embodiment, the technology described herein is implemented by one or more special-purpose computing devices. Special-purpose computing devices can be hard-wired to perform the technology, or can include digital electronic devices, such as one or more application-specific integrated circuits (ASICs) or field programmable gate arrays (FPGAs) that are permanently programmed to perform the technology, or can include one or more general-purpose hardware processors that are programmed to perform the technology according to program instructions in hardware, memory, other storage or combination. This special-purpose computing device can also combine customized hard-wired logic, ASIC or FPGA with customized programming to implement the technology. Special-purpose computing devices can be desktop computer systems, portable computer systems, handheld devices, network devices, or any other devices that combine hard-wired and / or program logic to implement the technology.
[0147] For example, Figure 4 4 is a block diagram illustrating a computer system 400 on which embodiments of the present disclosure may be implemented. Computer system 400 includes a bus 402 or other communication mechanism for communicating information, and a hardware processor 404 coupled to bus 402 for processing information. Hardware processor 404 may be, for example, a general purpose microprocessor.
[0148] The computer system 400 also includes a main memory 406, such as a random access memory (RAM) or other dynamic storage device, coupled to the bus 402 for storing information and instructions to be executed by the processor 404. The main memory 406 may also be used to store temporary variables or other intermediate information during the execution of instructions to be executed by the processor 404. When these instructions are stored in a non-transitory storage medium accessible to the processor 404, they render the computer system 400 into a special-purpose machine dedicated to performing the operations specified in the instructions.
[0149] Computer system 400 further includes a read only memory (ROM) 408 or other static storage device coupled to bus 402 for storing static information and instructions for processor 404. A storage device 410, such as a magnetic disk, optical disk, or solid state drive, is provided and coupled to bus 402 for storing information and instructions.
[0150] The computer system 400 may be coupled to a display 412, such as a cathode ray tube (CRT), via the bus 402 for displaying information to a computer user. An input device 414, including alphanumeric and other keys, is coupled to the bus 402 for communicating information and commands to the processor 404. Another type of user input device is a cursor control 416, such as a mouse, trackball, or cursor direction keys, for communicating direction information and command selections to the processor 404 and for controlling cursor movement on the display 412. The input device typically has two degrees of freedom in two axes, a first axis (e.g., an x-axis) and a second axis (e.g., a y-axis), which allows the device to specify a position in a plane.
[0151] Computer system 400 may implement the techniques described herein using custom hardwired logic, one or more ASICs or FPGAs, firmware, and / or program logic that is combined with the computer system to make computer system 400 a special purpose machine or to program it. According to one embodiment, the techniques herein are performed by computer system 400 in response to processor 404 executing one or more sequences of one or more instructions contained in main memory 406. Such instructions may be read into main memory 406 from another storage medium, such as storage device 410. Execution of the sequences of instructions contained in main memory 406 causes processor 404 to perform the process steps described herein. In alternative embodiments, hardwired circuitry may be used in place of or in combination with software instructions.
[0152] As used herein, the term "storage medium" refers to any non-transitory medium that stores data and / or instructions that cause a machine to operate in a particular manner. Such storage media may include non-volatile media and / or volatile media. For example, non-volatile media include: optical disks, magnetic disks, or solid-state drives (such as, storage device 410). Volatile media include dynamic memory, such as main memory 406. For example, common forms of storage media include: floppy disks, floppy disks, hard disks, solid-state drives, magnetic tape or any other magnetic data storage medium, CD-ROM, any other optical data storage medium, any physical medium with a pattern of holes, RAM, PROM and EPROM, FLASH-EPROM, NVRAM, any other memory chip or cassette tape.
[0153] Storage media are distinct from transmission media, but may be used in conjunction with transmission media. Transmission media participate in the transmission of information between storage media. For example, transmission media include coaxial cables, copper wires, and optical fibers, including the wires that make up bus 402. Transmission media may also take the form of acoustic or light waves, such as those generated during radio wave and infrared data communications.
[0154] When one or more sequences of one or more instructions are carried to the processor 404 for execution, various forms of media may be involved. For example, the instructions may initially be carried on a disk or solid-state drive of a remote computer. The remote computer may load the instructions into its dynamic memory and send the instructions over a telephone line using a modem. A modem local to the computer system 400 may receive data over a telephone line and convert the data into an infrared signal using an infrared transmitter. An infrared detector may receive the data carried in the infrared signal, and appropriate circuitry may place the data on the bus 402. The bus 402 carries the data to the main memory 406, from which the processor 404 retrieves the instructions and executes them. The instructions received by the main memory 406 may optionally be stored on the storage device 410 before or after being executed by the processor 404.
[0155] Computer system 400 also includes a communication interface 418 coupled to bus 402. Communication interface 418 provides bidirectional data communication coupled to network link 420, which is connected to local network 422. For example, communication interface 418 can be an integrated services digital network (ISDN) card, a cable modem, a satellite modem, or a modem that provides a data communication connection with a corresponding type of telephone line. As another example, communication interface 418 can be a local area network (LAN) card to provide a data communication connection to a compatible LAN. Wireless links can also be implemented. In any such implementation, communication interface 418 sends and receives electrical signals, electromagnetic signals, or optical signals that carry digital data streams representing various types of information.
[0156] The network link 420 typically provides data communication to other data devices through one or more networks. For example, the network link 420 may provide a connection to a host computer 424 through a local network 422, or to a data device operated by an Internet Service Provider (ISP) 426. The ISP 426, in turn, provides data communication services through a global packet data communication network, now commonly referred to as the "Internet" 428. Both the local network 422 and the Internet 428 use electrical, electromagnetic, or optical signals that carry digital data streams. The signals through the various networks and the signals on the network link 420 and through the communication interface 418 are example forms of transmission media that carry the digital data to and from the computer system 400.
[0157] Computer system 400 can send information and receive data (including program code) through the network(s), network link 420 and communication interface 418. In the Internet example, server 430 can send a request code for an application through Internet 428, ISP 426, local network 422 and communication interface 418.
[0158] The received code may be executed by processor 404 as it is received, and / or stored in storage device 410 or other non-volatile storage for later execution.
[0159] 3. Functional Overview - Generate and display target successful yield groups of seeds
[0160] Figure 7 Depicted is a detailed example of generating a target successful yield group of seeds identified for optimal yield performance on a target field based on agricultural data records of the seeds and geographic location data associated with the target field.
[0161] 3.1 Data Input
[0162] At step 705, agricultural intelligent computer system 130 receives agricultural data records for multiple different seeds from one or more fields. In an embodiment, agricultural data records may include crop seed data for one or more seeds. Crop seed data may include historical agricultural data related to specific seeds planted, grown and harvested on one or more fields. Examples of crop seed data may include, but are not limited to, historical yield values, harvest time information, and relative maturity of seeds, and any other observational data about plant life cycle. For example, agricultural data records may include seed data for available corn hybrids of two hundred (or more) different types. Crop seed data associated with each corn hybrid in corn hybrids will include: historical yield values associated with observed harvest, harvest time information relative to planting, and relative maturity of observations for each corn hybrid in the corn hybrid on each observed field in the observed field. For example, corn hybrid 001 may have agricultural data records including historical yield data collected from twenty (or more) different fields in the past ten years (or longer).
[0163] In an embodiment, the agricultural data record may include field-specific data related to the field in which the crop seed data was observed. For example, field-specific data may include, but is not limited to, geographic location information, observed relative maturity based on the field geographic location, historical weather index data, observed soil properties, observed soil moisture and water level, and any other environmental observations that may be specific to the field where the historical crop seed data was collected. Field-specific data may be used to further quantify and classify crop seed data because it is related to each seed in the seed. For example, based on the relative maturity of the seed and the length of the growing season, different fields in different geographic locations may be more suitable for different seeds. Fields within specific regions and sub-regions may have specific relative maturity of the growing season, which is based on the climate associated with the specific geographic location and the growing degree days (GDD) available during the growing season.
[0164] Figure 8 Depicted are examples of different regions within a state having different assigned relative maturity based on the duration of the growing period. State 805 is Illinois and is divided into a number of different regions and sub-regions. Examples of sub-regions may include regions based on county boundaries, city boundaries, or town boundaries. Each of regions 810, 815, 820, 825, and 830 represents a geographically specific area having a different duration of the growing period. For example, region 810 represents a region of fields having a shorter growing period based on its geographical location and associated climate due to a cool climate. As a result, region 810 may be classified as suitable for fields having seeds having a relative maturity of 100 days (in Figure 8 815 is located south of region 100 and therefore may have a warmer overall climate. Fields in region 815 may be classified as fields suitable for seeds having a relative maturity of 105 days. Similarly, regions 820, 825, and 830 are located further south than regions 810 and 815 and therefore are classified with relative maturity classifications of 110 days, 115 days, and 120 days, respectively. The relative maturity classifications for different regions may be used together with historical yield data for seeds to assess how seeds will perform in the field based on rated relative maturity.
[0165] In an embodiment, the specific field data within the agricultural data record may also include crop rotation data. Soil nutrient management for a field may depend on factors such as establishing different crop rotations and managing the amount of soil tillage. For example, some historical observations indicate that the "rotation effect" of rotating between different crops on a field can increase crop yields by 5% to 15% over growing the same crop every year. As a result, the crop rotation data within the agricultural data record may be used to help determine more accurate yield estimates.
[0166] In an embodiment, specific field data may include tillage data and management practices used during a crop season. Tillage data and management practices refer to the manner and schedule in which tillage is performed on a specific field. Soil quality and the amount of available nutrients in the soil vary depending on the amount of topsoil. Soil erosion refers to the removal of topsoil, which is the layer of soil richest in organic matter and nutrient value. One practice that causes soil erosion is tillage. Tillage breaks up soil aggregates and increases soil aeration, which can speed up the decomposition of organic matter. Therefore, tracking tillage management practices can help understand the extent of the impact on soil erosion, which can affect the overall yield of crops grown.
[0167] In an embodiment, agricultural data records include historical crop seed data and field-specific data from a collection of test fields for determining seed attributes by manufacturers. For example, Monsanto Corporation produces several commercial hybrid seeds (e.g., corn hybrids) and seed varieties (e.g., soybean varieties), and tests their crop growth on multiple test fields. Monsanto's test fields can be used as examples of collections of test fields, where agricultural data records are collected and received by agricultural intelligent computer system 130. In another embodiment, agricultural data records can include historical crop seed data and field-specific data from individual growers' ownership and operation. The collection of these fields for collecting agricultural data records can also be the same fields designated as the target fields for planting newly selected crops. In other embodiments, when determining the target successful yield group of seeds, the collection of fields owned and operated by growers can provide agricultural data records used by other growers.
[0168] Reference again Figure 7 At step 710, the agricultural intelligence computer system 130 receives geographic location information for one or more target fields. The target fields represent fields where the grower is considering planting or planning to plant a seed set selected from a target success yield group. In an embodiment, the geographic location information of the one or more target fields can be used in conjunction with agricultural data records for the specific fields to determine which seeds are best suited for the target fields based on relative maturity and climate.
[0169] 3.2 Agricultural data processing
[0170] At step 715, seed standardization instructions 172 provide instructions to generate a data set of seed attributes that describes a representative yield value and environmental classification for each seed received as part of an agricultural data record. In an embodiment, the agricultural data record associated with the seed is used to calculate a representative yield value and environmental classification for each seed in the seed. The representative yield value is the expected yield value of a particular seed if it is planted in a field based on historical yield values and other agricultural data observed from past harvests.
[0171] In an embodiment, standardized yield values can be calculated by standardizing multiple different yield observations from different fields across different observed growing years. For example, the first-year average growth cycle yield of a specific seed can be calculated using the field where the specific seed is first planted. The first-year average growth cycle yield of a specific seed can include the yield values observed by merging different fields in different years. For example, a specific seed may have been planted on a field that was tested during the product phase (2011 to 2017) of a Monsanto commercial product cycle (PS3, PS4, MD1, and MD2). However, the first cycle of a specific seed may have been planted on each field in the fields of different years. The following table shows such an example:
[0172] 2011 2012 2013 2014 2015 2016 2017 Cycle 1 PS3 PS4 MD1 MD2 Cycle 2 PS3 PS4 MD1 MD2 Cycle 3 PS3 PS4 MD1 MD2 Cycle 4 PS3 PS4 MD1 MD2
[0173] The columns in the table represent harvest years and the rows in the table represent Monsanto commercial product development cycles, where cycle 1 represents 4 years of seeds being planted on various fields, and cycle 2 represents a second cycle of 4 years for another set of seeds being planted on the same field environment, and so on.
[0174] In an embodiment, standardized yield values may be calculated based on similar cycles of seeds planted at multiple fields. For example, the standardized yield value for cycle 1 may be calculated as the average of the yield values observed on fields PS3 (2011), PS4 (2012), MD1 (2013), and MD2 (2014). By doing so, the yield values may be averaged based on the common characteristics of how many growth cycles have occurred on a particular field. In other embodiments, standardized yield values may be calculated based on other agricultural attributes from agricultural data records (such as, the same year or the same region / field).
[0175] In an embodiment, the environmental classification of each of the seeds may be calculated using the relative maturity field attributes associated with the agricultural data record for the seed. For example, a particular seed may have been planted on multiple fields within region 820. Each of the fields within region 820 is classified as having an observed growth period consistent with a relative maturity of 110 days. Therefore, based on the fields associated with the particular seed, a relative maturity may be assigned to the environmental classification of the particular seed that is equal to the relative maturity of region 820, i.e., 110 days. In other embodiments, if the fields associated with the historical observations of the particular seed include fields classified within multiple regions, the environmental classification may be calculated as an average of the different assigned relative maturity values.
[0176] In an embodiment, the data set of seed attributes includes standardized yield values for each seed and environmental classifications describing relative maturity values associated with the standardized yield values. In other embodiments, the data set of seed attributes may also include attributes associated with the seed growth cycle and field attributes, such as crop rotation, tillage, weather observations, soil composition, and any other agricultural observations.
[0177] Reference again Figure 7 , at step 720, the success probability generation instruction 174 provides instructions to generate a data set of success probability scores for each seed in the seed, and the data set of the success probability score describes the probability of successful yield as a probability value for obtaining a successful yield relative to the average yield of other seeds with the same relative maturity. In an embodiment, the success probability score of the seed is based on a data set of seed attributes relative to the geographical location associated with the target field. For example, the relative maturity value associated with the geographical location of the target field is partially used to determine the seed set to be evaluated, so as to calculate the success probability score for a specific seed. For example, corn hybrid 002 can be a seed with a standardized yield calculated at 7.5 bushels per acre and a specified relative maturity of 100GDD. Then, corn hybrid 002 is compared with other seeds with similar relative maturity, so as to determine whether corn hybrid 002 is a good candidate for planting based on the standardized yield values of corn hybrid 002 and other seeds.
[0178] Machine learning techniques are implemented to determine the probability score of success for seeds at a geographic location associated with a target field. In an embodiment, standardized yield values and assigned relative maturity values are used as predictors of a machine learning model. In other embodiments, additional attributes (such as crop rotation, tillage, weather observations, soil composition) may also be used as additional predictors of a machine learning model. The target variable of the machine learning model is a probability value from 0 to 1, where 0 equals 0% probability of successful yield and 1 equals 100% probability of successful yield, and in other embodiments, the target variable may be a probability value that may be scaled from 0 to 10, 1 to 10, or any other measurement scale. Successful yield is described as the probability that the yield of a particular seed is higher than the average yield of similarly classified seeds by a certain value. For example, a successful yield may be defined as a yield of 5 bushels / acre higher than the average yield of seeds with the same assigned relative maturity values.
[0179] Fig. 9 Describe sampling diagram, this sampling diagram has described the scope of the standardized yield value of seed in the relative maturity of classification.Meanwhile 905 represents the average yield value of the calculation of the seed with the same relative maturity (such as, 100GDD).In an embodiment, can determine which seeds have the standardized yield significantly higher than meanwhile 905 by realizing minimum significant difference calculation.Minimum significant difference is the value at the specific level of statistical probability.If the difference between two means has exceeded this value, then two means are considered to be different.For example, if the difference between the yield value of seed and the average yield calculated exceeds the minimum significant difference, then the yield of seed is considered to be different.In other embodiments, can use any other statistical algorithm to determine the significant difference of determination between yield value and meanwhile 905.
[0180] Range 910 represents a range of yield values that are considered to be within the minimum significant difference and therefore not significantly different. Threshold 915 represents the upper limit of range 910. Then, standardized yield values above threshold 915 are considered to be significantly different from average value 905. In an embodiment, range 910 and threshold 915 can be configured to represent a threshold value that is used to determine which seed yield is considered to be significantly higher than average value 905 and therefore a successful yield value. For example, threshold 915 can be configured to be equal to a value that is 5 bushels / acre higher than average value 905. In an embodiment, threshold 915 can be configured to be a yield value that depends on average value 905, range 910, and the total range of yield values for specific seeds with the same relative maturity.
[0181] Range 920 represents a range of yield values for seeds that are considered to be successful yields. Seed 925 represents a specific hybrid seed or seed variety within range 920 that has a standardized yield value above threshold 915. In an embodiment, the machine learning model can be configured to use range 910 and threshold 915 when calculating the probability of a success score between 0 and 1. Different machine learning models can include, but are not limited to: logistic regression, random forest, vector machine modeling, and gradient boosting modeling.
[0182] In an embodiment, logistic regression can be implemented as a machine learning technique to determine the probability of success score for each seed for a target field. For logistic regression, the input values for each seed in the seed are the standardized yield value and the environmental classification, which is specified as relative maturity. The functional form of logistic regression is:
[0183] in
[0184] P(y=1|x1= yld i , x2= RM j )
[0185] is the probability of success (y=1) of product i with standardized yield value and in target field j with given relative maturity; constants a, b and c are regression coefficients estimated from historical data. For each seed, the output of logistic regression is a set of probability scores between 0 and 1 that specify success at the target field based on the relative maturity assigned to the geographic location associated with the target field.
[0186] In another embodiment, a random forest algorithm can be implemented as a machine learning technique to determine the probability of a success score for each of the seeds for a target field. The random forest algorithm is an integrated machine learning method that operates by constructing multiple decision trees during a training cycle and then outputs a class that is a mean regression of the individual trees. The input values for each seed are a standardized yield value and an environmental classification as a relative maturity. The output is a collection of probability scores between 0 and 1 for each seed.
[0187] In another embodiment, support vector machine (SVM) modeling can be implemented as a machine learning technique to determine the probability of the success score of each seed in the seed for the target field. Support vector machine modeling is a supervised learning model that is used to classify whether the input uses classification and regression analysis. The input value for support vector machine modeling is the standardized yield value and the relative maturity value of the environmental classification for each seed. The output is a set of probability scores between 0 and 1 for each seed. In yet another embodiment, gradient boosting (GBM) modeling can be implemented as a machine learning technique, wherein the input value is the standardized yield value and the relative maturity value of the environmental classification for each seed. Gradient boosting is a technique for regression and classification problems that produces a prediction model in the form of a set of weak prediction models (such as, decision trees).
[0188] refer to Figure 7 , at step 725, the yield classification instruction 176 generates a target successful yield group, which is composed of a subset of seeds that have been identified as having a high probability to produce a yield significantly higher than the average yield of other seeds within the same relative maturity classification for the target field. In an embodiment, the target successful yield group contains seeds with a probability value of success that is higher than a specific success probability threshold. The success probability threshold can be a configured probability value associated with a yield that is significantly higher than the average yield of other seeds. For example, if at step 720, the yield threshold for successful yield is equal to 5 bushels / acre above the average, the success probability threshold can be associated with the probability value of success equal to the probability value of success of the yield threshold. For example, if the yield threshold is equal to 5 bushels / acre above the average yield and the probability value of success is 0.80, the success probability threshold can be specified as 0.80. In this example, the target successful yield group will contain seeds with a probability value of success equal to or greater than 0.80.
[0189] In other embodiments, the success probability threshold may be configured to be higher or lower depending on whether the grower desires a smaller or larger target successful yield group, respectively.
[0190] 3.3 Current target success output group
[0191] In an embodiment, the target successful yield group contains seeds with specified relative maturity values, which are equal to the relative maturity associated with the target field. At step 730, the presentation layer 134 of the agricultural intelligent computer system 130 is configured to display or cause the display of the target successful yield group and the standardized yield value for each seed in the target successful yield group on a display device on the field manager computing device 104. In another embodiment, the presentation layer 134 can communicate the display of the target successful yield group to any other display device that can be communicatively coupled to the agricultural intelligent computer system 130, such as a remote computer device, an in-cab display device, or any other connected mobile device. In yet another embodiment, the presentation layer 134 can communicate the target successful yield group to other systems and subsystems with the agricultural intelligent computer system 130 for further processing and presentation.
[0192] In an embodiment, the presentation layer 134 may display additional seed attribute data and other agricultural data that may be relevant to the grower. The presentation layer 134 may also sort the seeds in the target successful yield group based on the probability of success value. For example, the display of seeds may be sorted in descending order of probability of success, so that the grower can first view the most successful seeds for their target fields.
[0193] In some embodiments, after receiving the displayed information, the grower can act on the information and plant the recommended seeds. In some embodiments, the grower can work as part of an organization that determines a target successful yield group and / or can work separately. For example, the grower can be a customer of the organization that determines the target successful yield group and can plant seeds based on the target successful yield group.
[0194] 4. Functional Overview - Generate and display target seeds for planting
[0195] Fig.10 Depicted is a detailed example of generating a target set of seeds identified for optimal yield performance and managed risks on a target field based on agricultural data records of the seeds and geolocation data associated with the target field.
[0196] 4.1 Data Input
[0197] At step 1005, the agricultural intelligence computer system 130 receives a data set of candidate seeds, the data set including: one or more seeds suitable for planting on a target field, a probability value of success associated with each seed, and historical agricultural data associated with each seed. In an embodiment, the data set of candidate seeds may include one or more sets of seeds identified by the seed classification subsystem 170 as having a high probability of producing a successful yield value on the target field and historical agricultural data associated with each seed in the candidate seed set. Figure 7 The target successful yield group generated at step 725 in can represent a data set of candidate seeds.
[0198] In an embodiment, historical agricultural data may include agricultural data related to the planting, growth, and harvesting of specific seeds on one or more fields. Examples of agricultural data may include, but are not limited to, historical yield values, harvest time information, and relative maturity of seeds, as well as any other observational data related to the life cycle of the plant. For example, if the data set for the candidate seeds is a target successful yield group from the seed classification subsystem 170, the agricultural data may include the average yield value and relative maturity assigned to each seed.
[0199] At step 1010, the agricultural intelligence computer system 130 receives data about target fields in which the grower plans to plant a set of target hybrid seeds and / or seed varieties. In an embodiment, the data about the target fields is attribute information, which includes, but is not limited to, geographic location information for the target fields and size and dimension information for each of the target fields. In an embodiment, the geographic location information of the target fields can be used in conjunction with historical agricultural data to determine the optimal set of target seeds and the amount of each of the target seeds to be planted in each of the target fields based on the relative maturity and climate of the target fields.
[0200] 4.2 Seed Selection
[0201] At step 1015, the seed filtering instructions 182 provide instructions to select a subset of one or more seeds from a candidate set of seeds having a probability of success value greater than or equal to a target probability filtering threshold. In an embodiment, the target probability filtering threshold is a configured threshold of the probability of success value associated with each seed in the seed candidate set of seeds. The target probability filtering threshold can be used to further narrow the selection pool of seeds based only on selecting seeds with a certain probability of success. In an embodiment, if the candidate set of seeds represents the target successful yield group generated at step 725, the seed set may have been filtered to include only seeds with high probability of success values. In one example, the target probability filtering threshold can have the same threshold as the successful yield threshold used to generate the target successful yield group. If this is the case, the subset of one or more seeds can include the entire set of seeds. In another example, a grower can expect a narrower list of seeds, which can be achieved by filtering out seeds with a probability of success value lower than the expected value for the target probability filtering threshold.
[0202] At step 1020, seed standardization instruction 172 provides instructions to generate representative yield values of each seed in a subset of one or more seeds based on the yield values of historical agricultural data of each seed in the seed. In an embodiment, if specific seeds are planted in a field based on historical yield values and other agricultural data observed from past harvests, the representative yield value is the expected yield value for specific seeds. In an embodiment, the representative yield value is the calculated mean value of the yield of the growth period from multiple different observations on multiple fields. For example, the representative yield value can be calculated as the mean value of the growth cycle year of different observations, wherein the average first year growth cycle yield for specific seeds can be combined with a combination of yield values observed from different fields in different years. After calculating the average growth cycle yield for different growth cycle years, each mean value in the mean value can be combined to generate a representative average yield for each specific seed. In another embodiment, the representative yield value can be the standardized yield value calculated at step 715.
[0203] 4.3 Generating risk values for seeds
[0204] At step 1025, the risk generation instructions 184 provide instructions to generate a data set of risk values for each hybrid seed or seed variety in a subset of one or more seeds based on historical agricultural data associated with each of the seeds. The risk value is based on a representative yield value, describing the amount of risk for each seed according to the yield variability. For example, if the representative yield of corn hybrid 002 is 15 bushels per acre, but the variability of corn hybrid 002 is high, so that the yield can vary from 5 bushels per acre to 25 bushels per acre, and the representative yield of corn hybrid 002 may not be a good representation of the actual yield because the yield may vary between 5 bushels per acre and 25 bushels per acre. A high risk value is associated with a high variability in yield returns, while a low risk value is associated with a low variability in yield returns and yield results that are more closely consistent with the representative yield.
[0205] In an embodiment, the risk value for seed is based on the variability between the annual yield returns of specific seeds in two or more years. For example, the risk value calculated for hybrid 002 includes the variability of calculating the yield value from the multi-year yield output of historical agricultural data. The variance in the yield output from 2015 to 2016 for corn hybrid 002 can be used to determine the risk value associated with the representative yield value for corn hybrid 002. The variance determining the yield output is not limited to using the yield output of the previous two years, and the yield output data of many years can also be utilized to calculate the variance. In an embodiment, the calculated risk value can be represented by the standard deviation of bushels / acre, wherein the standard deviation is calculated as the square root of the calculated risk variance.
[0206] In an embodiment, a risk value for a seed may be based on the variability of yield output from field observations for a particular year. For example, calculating a risk value associated with field variability may include determining the variability of yield from each field observed for a particular seed in a particular year. If the observed yield output across multiple fields for a particular seed is 5 bushels per acre to 50 bushels per acre, the particular seed may have high field variability. As a result, a high risk factor may be assigned to a particular seed based on field variability because the expected output for any given field may vary between 5 bushels per acre to 50 bushels per acre, rather than being close to a representative yield value.
[0207] In another embodiment, the risk value for seed can be based on the variability between yield returns between years and the variability between field observations. Both the year-to-year risk value and the field risk value can be combined to represent a risk value that incorporates the variability of yield output across multiple observed fields and multiple observed seasons. In other embodiments, the risk value can incorporate other observed crop seed data associated with historical crop growth and yield.
[0208] 4.4 Generate target seed dataset
[0209] At step 1030, the optimize classification instructions 186 provide instructions to generate a dataset of target seeds for planting on the target field based on the risk value, the representative yield value for the seed, and the dataset of one or more attributes for the target field. In an embodiment, the target seed in the dataset of target seeds is selected based on the representative yield value of the target seed and the associated risk value from the dataset of risk values.
[0210] Determining the combination of seeds to include in the target seed data set involves determining the relationship between the representative yield for a particular seed and the risk value associated with the particular seed. Selecting seeds with a high representative yield may not result in an optimal set of seeds if a high yield seed also has a high risk level. Conversely, selecting seeds with a low risk value may not result in a sufficiently high yield reward.
[0211] In an embodiment, seeds from a subset of one or more seeds may be graphed based on their respective representative yield values relative to their associated risk values. Fig.11 Describe the example curve graph 1105 of the yield of the subset of one or more seeds relative to risk.Y-axis 1110 represents the representative yield (as expected yield) for seed, and x-axis 1115 represents the risk value (expressed as standard deviation) for seed.By expressing the risk value as standard deviation, the unit of risk value can be the same as the unit representing yield, i.e. bushels / acre.The points on the curve graph 1105 represented by group 1125 and group 1130 represent each seed in the seed of the subset of one or more seeds.For example, curve graph 1105 shows that the representative yield value of seed 1135 is 200 bushels / acre, and the standard deviation of risk value is 191 bushels / acre.In other embodiments, different units can be used to generate curve graph 1105, such as profit per acre measured in US dollars or any other derived measurement unit.
[0212] In an embodiment, determining which seeds belong to the target seed dataset involves determining the expected yield return for a specified amount of risk. In order to generate a target seed set that may be resilient to various environments and other factors, it is preferred to generate a diverse set of seeds that contain seeds with lower risk values and higher risk values and medium to high yield outputs. Fig.10, step 1032 represents generating a target threshold value of a representative yield value for a range of risk values. In an embodiment, the optimization classification instructions 186 provide instructions to calculate an optimal boundary curve, which represents a threshold value of an optimal yield output with a manageable risk tolerance within a range of risk values. A boundary curve is a fitted curve that represents the optimal output relative to the graphical input values taking into account optimal efficiency. For example, graph 1105 contains seeds based on representative yields relative to risk values, where it can be inferred that a particular seed with a higher yield may also have a higher risk. Conversely, seeds with lower risk values may have lower representative yield values. Boundary curve 1120 represents an optimal curve that tracks optimal yield based on a range of risk values.
[0213] At step 1034, the optimization classification instructions 186 provide instructions to select seeds that make up the target seed set by selecting seeds with representative yields and risk values that meet the thresholds defined by the boundary curve 1120. Seeds that fall on or near the boundary curve 1120 provide the best yield level at the desired risk level. The target seeds 1140 represent the best set of seeds for the data set of the target seeds. The seeds that fall below the boundary curve 1120 have a suboptimal yield output for the risk level, or have a higher risk than expected for the level of the yield output produced. For example, seed 1135 is below the boundary curve 1120 and can be interpreted as having an optimal yield below its risk amount, as shown by placing seed 1135 vertically below the boundary curve 1120. Moreover, seed 1135 can be interpreted as having an expected risk higher than its yield output, as shown by placing seed 1135 horizontally on the right side of the boundary curve 1120 for the representative yield. Seeds 1135 that are not on or near boundary curve 1120 have suboptimal representative yields for their associated risk values and are therefore not included in the target seed set. Additionally, seeds 1135 represent seeds that have higher than desired risk values and are therefore not included in the target seed set.
[0214] In an embodiment, the optimization classification instruction 186 provides instructions to generate allocation instructions for each target seed in the target seed set. The allocation instruction describes the seed allocation amount for each target seed in the target seed set, and the target seed set provides the best allocation strategy for growers based on the quantity and position of the target field. For example, the allocation instruction for the target seed set including seeds (CN-001, CN-002, SOY-005, CN-023) may include 75% of CN-001, 10% of CN-002, 13% of SOY-005 and 2% of CN-023. The embodiment of the allocation instruction may include but is not limited to: the number of seed bags, the percentage of the total number of seeds to be planted across the target field, or the number of acres allocated for each target seed to be planted. In an embodiment, a third-party optimization solver product (such as, IBM's CPLEX optimizer) can be used to calculate and determine the allocation amount. The CPLEX optimizer is a mathematical programming solver for linear programming, mixed integer programming and quadratic programming. An optimization solver (such as a CPLEX optimizer) is configured to evaluate representative yield values and risk values associated with target seeds, and determine a set of allocation instructions for allocating seed quantities for each of the target seeds in the target seed set. In an embodiment, the optimization solver can use the sum of the representative yield values of the target seeds and the calculated sum of the risk values of the target seeds to calculate a configured total risk threshold, which can be used to determine an upper limit on the allowable risk and yield output for the target seed set.
[0215] In another embodiment, the optimization solver may also input target field data describing the size, shape, and geographic location of each of the target fields in order to determine allocation instructions including placement instructions for each of the allocations of target seeds. For example, if a particular target field is shaped or sized in a particular manner, the optimization solver may determine that it is preferred to allocate one target seed on that particular field, as opposed to planting multiple target seeds on that particular field. The optimization solver is not limited to the CPLEX optimizer, and other embodiments may implement other optimization solvers or other optimization algorithms to determine a set of allocation instructions for a set of target seeds.
[0216] 4.5 Seed profile analysis
[0217] Step 1030 describes determining and generating a target seed set based on a target field using a boundary curve to determine the optimal yield output for a desired risk level. In an embodiment, the optimization classification instructions 186 provide instructions to configure a boundary curve to determine the overall optimal performance of a grower's seed profile relative to other growers in the same area or sub-area. For example, a representative yield output and an overall risk value can be calculated for each grower in a specific area. For example, using historical agricultural data for multiple growers, the representative yield values and associated risk values of the seeds planted by each grower can be summarized to generate a total yield output value and a total risk value associated with each grower. Then, the total value for each grower can be plotted on a seed profile graph, similar to the graph 1105, where each point on the graph can represent the total seed yield output and the total risk of the grower. In an embodiment, a boundary curve can be generated to determine the optimal total yield output and the total risk for growers in a specific area. A grower on or near the boundary curve can represent a grower whose seed profile produces the best yield with a manageable amount of risk. A grower below the boundary curve represents a grower who does not maximize its output based on its risk.
[0218] In an embodiment, if the aggregate yield output and aggregate risk for the grower's seed profile do not meet the optimal threshold for the seed profile described by the boundary curve on the seed profile graph, the optimization classification instructions 186 provide instructions to generate an alert message for a specific grower. The presentation layer 134 can be configured to present the alert message to the grower and send the alert message to the field manager computing device 104. The grower can then select a request for a target seed set that can provide the best yield output for the future planting season.
[0219] 4.6 Current Set of Target Seeds
[0220] In an embodiment, the dataset of target seeds may include a representative yield value and a risk value from the dataset of risk values associated with each target seed in the dataset of target seeds for the target field. Fig.10At step 1035, the presentation layer 134 of the agricultural intelligence computer system 130 is configured to communicate a display of a data set of target seeds including a representative yield value and an associated risk value for each target seed on a display device of the field manager computing device 104. In another embodiment, the presentation layer 134 may communicate the display of the data set of target seeds to any other display device that may be communicatively coupled to the agricultural intelligence computer system 130 (such as a remote computer device, an in-cab display device, or any other connected mobile device). In yet another embodiment, the presentation layer 134 may communicate the data set of target seeds to other systems and subsystems with the agricultural intelligence computer system 130 for further processing and presentation.
[0221] In an embodiment, the presentation layer 134 may display allocation instructions for each target seed, including seed allocation and placement information. The presentation layer 134 may also rank the target seeds based on allocation amounts, or may present the target seeds based on a placement strategy on the target field. For example, the display of the target seeds and allocation instructions may be superimposed on a map of the target field so that the grower can visualize the planting strategy for the upcoming season.
[0222] In some embodiments, a grower may receive information related to an allocation instruction and plant seeds based on the allocation instruction. The grower may work as part of an organization that determines the allocation instruction and / or may work separately. For example, a grower may be a client of an organization that determines the allocation instruction and may also plant seeds based on the allocation instruction.
[0223] 5. Functional Overview - Generate and display yield improvement recommendations by field
[0224] As described above, embodiments disclosed herein can be used to identify seed products that will perform best on a target field based on inputs received by an agricultural intelligence computer system 130. Such inputs can include: agricultural data and historical yield data for different seeds, and environmental data associated with the fields of growers who observe seed data. In addition to grower data, the agricultural intelligence computer system 130 can also utilize seed and environmental data observed during different breeding and development stages associated with seeds. These data are valuable, and as harvests are analyzed, environmental conditions change, unique field locations are added, and new and existing seeds are further developed and tested, these data will continue to grow over time. Even so, seeds cannot be tested at each field location or under each potential combination of environmental conditions.
[0225] In an embodiment, the data used in the disclosed machine learning model is enriched by using genetic data to generate agricultural data of seeds that have not been tested under specific environmental conditions. For example, the disclosed technology uses genetic data by obtaining and using germplasm (basic genetics + traits) and / or pedigree information, genetic cluster patterns and / or genomic marker relationships to estimate yield data in different environments. All such data are digitally stored, retrieved and converted using computer-implemented instructions.
[0226] Fig.12 An example flow chart is illustrated that includes using genetics to fill in data gaps in historical agricultural data. The resulting agricultural data will be enhanced with predicted estimated data that can provide a basis for improving seed placement calculation strategies in actual fields with specific environmental conditions. According to one example, Figure 1 The agricultural intelligence computer system 130 is programmed or configured to perform Fig.12 1200. For example, the seed classification subsystem 170 and / or the seed recommendation subsystem 180 may include genetic modeling instructions as further described herein.
[0227] 5.1 Data Input
[0228] At block 1202, for example, agricultural intelligence computer system 130 receives or otherwise accesses agricultural data records. In one example, computer system 130 receives agricultural data records via digital data communication network 109. For example, agricultural data records include: crop seed data and yield attributes of the seeds and environmental data in which the seeds are planted and / or tested.
[0229] Fig.13 An example of received agricultural data records and further processing to estimate data values is shown. Fig.13 In, the received agricultural data records include seed products G1, G2, G3, G4, G5, G6 and yield data (e.g., associated with different fields or environments E1, E2, E3, E4) provided in the form of bushels per acre (bu / ac). The yield data can be associated with a specific year or harvest, and additional data records for other years / harvests can be received, and / or the yield data can be an average yield or other representation of multi-year data. As an example, seed product G1 was planted in field E1 and was associated with an actual yield of 222 bu / ac. However, the received agricultural data record lacks the selected yield data that should be associated with a given seed product in a given field. For example, in Fig.13In the data set, no yield data is associated with seed product G3 in field E1. This data gap may be due to not planting a given seed in a given field or other reasons. In practice, it is impossible to conduct actual field testing or laboratory testing of each seed in every combination of unique environmental conditions.
[0230] Fig.14 Another example of received agricultural data records and further processing to fill in data gaps is shown. Fig.14 , data records for different product stages are provided. For example, data record 1402 is associated with an early product development or breeding stage, data record 1404 is associated with a subsequent product development or commercial testing stage, and data record 1406 is associated with a field use stage. The first column of each data record identifies different seeds (e.g., corn hybrids) at breeding stages H1, H2, H3, H4, and seed products at commercial testing and field use stages P1, P2, P3, P4. For the purposes of discussion, seeds H1, H2, H3, H4 are grown from the breeding stage and are renamed or subsequently identified as corresponding seed products P1, P2, P3, P4 in other stages, respectively.
[0231] The top row in data record 1402 identifies different test cycles PS3, PS4, which may be defined by a given time period (such as a year) and associated with unique environmental conditions. The top row in data record 1404 identifies different additional test cycles MD1, MD2, which may similarly be defined by a given time period (such as a year) and may be associated with other unique environmental conditions. The top row in data record 1406 identifies cycles associated with different fields or environments Fld1-FldX in which seed products were grown and harvested to provide yield data.
[0232] Similar to Fig.13 Agricultural data records received in Fig.14 The data records in also have data gaps, where no yield data are associated with a given seed product in a given field or test environment. However, even when data gaps exist, the Fig.13 and Fig.14The agricultural data record of expression also provides abundant information, test cycle and multi-year planting and harvest cycle for tens of thousands of field positions and test conditions and one thousand kinds or more seeds in multiple product stages.The present embodiment uses genetics relationship to further enhance and build on this abundant information.The agricultural data record received can be associated with extensive characteristic data, and this characteristic data is relevant with seed, environment and / or test conditions and yield attribute.The general category of this characteristic data is relevant with weather, soil condition, environmental classification, field management practice, pest risk, hereditary characteristics and the whole genome of capturing the non-cumulative interaction between gene and environmental characteristics by environmental characteristics (GxE characteristics).Other categories of this characteristic data comprise genome by management characteristics (GxM) and genome by environment by management characteristics (GxExM), and they have captured the interaction between the non-additive interaction between hereditary and management characteristics and hereditary, environment and management characteristics respectively.This paper provides various specific features in this category.
[0233] Reference Fig.12 , at frame 1204, computer system 130 receives or otherwise accesses the genetic characteristics data relevant to the seed. The genetic data can include the genetic relationship between the seeds. Although in one example, computer system 130 uses the original genetic characteristics data received to develop this genetic relationship between the seeds. In one example, the agricultural data record received or accessed at frame 1202 is associated with a first seed set, the genetic characteristics data received or accessed at frame 1204 is associated with a second seed set, and the second seed set includes the first seed set. In certain embodiments, genetic characteristics data and / or genetic relationship can be commercially obtained from the crop science department of Bayer AG in Leverkusen, Germany.
[0234] Fig.15 An example of genetic signature data including genomic marker data is shown. Genomic marker data is generally a gene or DNA sequence that can be used to identify a unique genetic signature. In one example, Fig.15 The genomic marker data, represented by genes 1-10 in , can be combined with genome-wide single nucleotide polymorphism (SNP) markers found in seeds.
[0235] Fig.16An example kinship matrix based on pedigree is illustrated, which identifies paired relationships between seeds based on seed pedigrees. The relationship is captured by a value between 0.0 and 1.0, where a value of 0.0 represents that two seeds are completely different and unrelated according to pedigree, and a value of 1.0 represents that two seeds have the same pedigree. The computer system 130 can receive the pedigree-based kinship matrix at box 1204, or can generate the matrix using genomic marker data by tracking female and male inbreeding marker data related to the original parental origin genotype (pedigree), to develop the matrix.
[0236] Fig.17 An example of the organization of seeds into genetic cluster relationships is shown. In general, a genetic cluster is a group of genes found within the DNA of a seed that encode similar polypeptides or proteins that share a common function. Fig.17 The middle and lower branches or individual ends represent different seeds, which are organized in the gene tree based on shared genomic marker data or genes. Fig.17 Illustrated is an example of further identifying seeds by different genetic clusters 1702, 1704, 1706, 1708, 1710. The computer system 130 may receive the genetic cluster data at block 1204, or may use the genomic marker data to organize the seeds into any number of suitable genetic cluster relationships.
[0237] Fig.18 An example genetic marker-based kinship matrix is illustrated that identifies pairwise relationships between seeds based on SNP markers. The relationships are captured by values between 0.0 and 1.0, where a value of 0.0 indicates that the two seeds are completely different and unrelated according to the SNP markers, and a value of 1.0 indicates that the two seeds are identical. The computer system 130 can receive the marker-based kinship matrix at block 1204, or can use genomic marker data to generate the matrix using an appropriate computational method (such as a squared Euclidean distance calculation). As shown by Fig.16 and Fig.18 As schematically shown, the marker-based kinship matrix provides more detailed relationship data between seed pairs compared to the pedigree-based matrix.
[0238] Fig.19 An example of an inbreeding coding matrix is shown that can be used to distinctively identify a seed product by capturing the inbred parental lines of the product. More specifically, Fig.19A "Female" column is included that identifies three example female seed products as FL1, FL2, and FL3. The "One-Click Examples" matrix for female parental lines provides an embedding method to encode the parental characteristics of a given product. More specifically, female product FL1 is encoded as 100, FL2 is encoded as 010, FL3 is encoded as 001, and so on if there are additional female products. Fig.19 A "Male" column is also provided, which identifies three example male seed products as ML1, ML2, and ML3. The corresponding "One-Click Example" matrix for male parental lines provides an embedded method to encode the parental characteristics of a given product. More specifically, male product ML1 is encoded as 100, ML2 is encoded as 010, ML3 is encoded as 001, and so on if there are additional male products.
[0239] In this example of inbred coding, a one-click matrix is used to convert a given product line into a code or ID that captures inbred parental line information. Typically, hybrid seed products are characterized by female parental lines and male parental lines. For a given hybrid seed, a one-click matrix is used to provide a code for each female and male line of the hybrid. For example, the first hybrid developed from FL1 and ML1 will be coded as 100+100, the second hybrid developed from FL2 and ML2 will be coded as 001+010, and the third hybrid developed from FL2 and ML3 will be coded as 010+001. As a result, hybrids can be uniquely coded by aggregating inbred parental line data, which is useful for distinguishing different seed products and more accurately associating GxE and yield characteristics with different products.
[0240] 5.2 Data Attribution
[0241] At block 1206, the computer system 130 generates predicted yield attributes for seeds associated with a particular field or environment. More specifically, the computer system 130 utilizes the received agricultural data records and genetic characteristic data (including genetic relationships between seeds) to provide estimated yield data to fill in the gaps, for example, in the Fig.13 and Fig.14In one example, a computer system combines various terms using a statistical mixed effects model using the following mathematical representation: yield (bu / ac) = f(G+E+GxE+error). The G term represents the genetic characteristic data of a list of hybrids / variety, and may include relative maturity, biotech traits, genomic marker data, pedigree-based kinship matrix, genetic cluster relationships, and genetic marker-based kinship matrix. The E term represents the environmental and management characteristics of a collection of fields, and may include precipitation, drought risk, heat stress, soil composition, soil texture, soil drainage, environmental zones, disease risk, crop rotation, farming practices, etc. The GxE term is a mathematical term for capturing the non-additive interactions between genetic characteristics and environmental / management characteristics. GxE captures the variability caused by the different performance of seeds under different environmental conditions, which can also take into account management characteristics. The error term helps to take into account yield variations not captured by the G, E, and GxE terms.
[0242] In general, genetic relationship data (such as, genomic marker data, pedigree-based kinship matrix, genetic cluster relationships and / or kinship matrix based on gene markers) help improve the data estimation process by identifying the genetic similarity between seeds tested under specific environmental conditions and seeds not tested under specific environmental conditions. Appropriate machine learning models (such as, statistical mixed effects models or best linear unbiased prediction (BLUP) models) and the genetic characteristics and relationships discussed herein, as well as other possible genetic characteristics and relationships, original environmental characteristics or filtered and engineered environmental characteristics and GxE interactions all use this degree of genetic similarity to fill data gaps to provide more reliable yield predictions. Fig.13 and Fig.14 Each of the graphs in provides an example of a received data record using genetic characteristics and the resulting data record with estimated yield values to fill in the data gaps (processing blocks 1310, 1410), respectively.
[0243] In one example, the estimated yield data can be calculated using a mixed genomic BLUP according to the following equation: yield = Xβ + Zu + error. The Xβ term represents a vector of fixed environmental effects, Zu represents a vector of relationships or correlations between hybrid genomic factors and environmental factors, and the error term is a vector of random residual effects that take into account other yield variations. In this example, the u term follows a random distribution with correlations captured by a variance-covariance matrix K (e.g., a kinship matrix between hybrids) and an identification matrix I, and can be determined according to the following equation:
[0244] 5.3 Determine expected yield performance
[0245] At block 1208, the computer system 130 determines a predicted yield performance for the one or more seeds. In one example, at block 1208, the computer system 130 generates a probability of success score for the one or more seeds based on the estimated data set, other genetic data, and field attributes for one or more target fields or environments generated at block 1206. Alternatively or in combination, the computer system 130 determines a predicted yield performance at block 1208 using the estimated data set, other genetic data, and field attributes to generate an absolute or relative yield value, a yield ranking, and / or other yield performance metric.
[0246] In one example, flowchart 1200 at block 1208 or elsewhere includes receiving or otherwise accessing characteristic data for one or more target fields where seeds are planned to be planted. A machine learning model is implemented to determine a predicted yield performance for the seeds at the target field(s). In an embodiment, the machine learning model uses estimated yield data, genetic relationship data, genomic marker data (e.g., Fig.15 related data), genetic cluster data (e.g. Fig.16 and Fig.17 data related to inbreeding (e.g. Fig.19 related codes) and / or genetic kinship matrices (e.g. Fig.16 and Fig.18 The target variable of a machine learning model can be a probability value ranging from 0 to 1, where 0 equals 0% probability of a successful yield and 1 equals 100% probability of a successful yield. In the example, a successful yield is described as the probability that the yield of a particular seed is a certain value higher than the average yield of similarly classified seeds. For example, a successful yield can be defined as a yield that is 5 bushels / acre higher than the average yield of seeds with the same assigned relative maturity value. For example, this article is relative to Figure 7 and Fig. 9 Additional details and techniques are described.
[0247] The present disclosure includes additional techniques for improving field layout models using genomic predictions and genetics-related traits. Additional genetic trait data (G) and interactions between genetic traits and environmental traits (GxE) can be used as data layers in training models, including, for example: hybrids and inbred genetic hybrid populations; genetic markers associated with key biotech traits and key quantitative trait loci (QTLs), genome-wide genetic markers and / or long-form haplotypes; inbred BLUP-GCA (general combining ability) yield and yield-related phenotypes; hybrid and inbred disease traits (GLS, NLB, SR, ASR, GW) and other genomic prediction traits and derived genetically-related traits.
[0248] According to an embodiment, the field layout model can use a data layer of genetic hybrid populations including inbred product lines and hybrid clusters. More specifically, the cluster model is configured to process genetic marker data and inbred and hybrid information to generate or estimate genetic hybrid populations of inbred product lines and hybrid clusters therefrom. Typically, a hybrid population is a group of related germplasms or unrelated germplasms from the same population or different populations that exhibit similar combining abilities and hybrid responses when hybridized with germplasms from other genetically different germplasm populations. Reference inbred and hybrid information is used to verify and derive hybrid populations or hybrid clusters.
[0249] The field layout model can also use genetic marker data, including genetic marker data of biotech traits, genetic marker data of key QTLs, whole genome genetic marker data and / or long haplotype data. In this example, the field layout model can use such genetic marker data, which is usually raw data, as an alternative or supplement to other genetic kinship matrix data, and the genetic marker data provides correlations derived from the above raw genetic marker data.
[0250] In addition, the field layout model can use a hybrid predictive model that is configured to process original research and market development disease and other phenotypes of hybrid and inbred products data and genetic marker data to develop inbred BLUP-GCA yield and yield-related phenotypes, hybrid and inbred disease traits, and other genomic predicted or derived characteristics.
[0251] 5.4 Seed Optimization and Recommendation Generation
[0252] At frame 1210, computer system 130 can use the unique features or attributes of one or more target fields and the data set of success probability score to generate field-specific seed recommendations for grower fields. Computer system 130 can receive the unique features or attributes of the target fields at frame 1210, or can receive these features at other times. This recommendation can include information such as the sowing rate in units of density. The sowing rate in units of density can be used to recommend selecting specific seeds, so as to obtain the target yield range of expectation. In an embodiment, the sowing rate in units of density can also be used to regulate seed population or seed density. Fig.12 The overall result of the process is that the estimated data provides useful yield information that can be matched to the unique characteristics of the target field, thereby tailoring each recommendation to those unique characteristics. This improves average yields compared to previous models, which could generalize recommendations by a larger scale, such as a region or zip code.
[0253] When generating recommendations, the computer system 130 may also perform feature selection to reduce redundancy from many field features. In general, feature selection helps avoid potential scalability issues, remove redundant features, eliminate non-predictive features or combinations of features, and enhance generalization capabilities by reducing overfitting, thereby simplifying the model and reducing the impact of missing feature data. The computer system 130 may perform feature selection using appropriate strategies, such as backward selection based on automated likelihood ratio tests.
[0254] Furthermore, similar to other examples discussed herein, computer system 130 may cause display of recommendations for each field.
[0255] 5.5 Validating and Tuning the Model
[0256] At box 1212, computer system 130 can verify and adjust machine learning model. In one example, verification process includes: receiving actual yield data of hybrids / variety planted in a specific field, and comparing the yield data with the estimated yield data. Verification process can also receive yield data of different seeds grown on fields of similar combinations of the same field, nearby fields or otherwise sharing attributes, and compare the yield data for different seeds with each other. The data provided by this verification at field level can be used to help improve recommendation, and provide results that are better than previous models that may perform verification at a level lower than the level. Then, computer system 130 can solve the difference between actual yield data and estimated yield data by modifying the corresponding model (such as, by modifying GxE, GxM and / or GxExM relationship and / or adjusting the above-mentioned error term). Then, the adjusted model can be used to generate estimated yield data and calculate the future iteration of recommendation.
[0257] At frame 1212, computer system 130 can also use actual yield data with other data input and machine learning techniques, to help identify specific environment and management attributes that can predict positive seed arrangement results.Usually, computer system 130 can apply machine learning techniques to identify the correlation between the combination of a single attribute and attribute and yield output.In one example, for example, by simplifying calculations and / or data inputs from different users, predicted attributes are used to generate the yield data of estimation and the future iteration of plan recommendation.Predicted attributes can also be used in different breeding and product development stages, to help enhance agricultural data and promote research and testing.
[0258] 6. Feature Overview—Including examples of feature engineering to enhance data for recommendation modeling
[0259] Fig. 20 Illustration of using feature engineering to classify feature data and prepare it for Fig.12 Example process of agricultural data records of recommendation models of the invention. The process disclosed herein can be extended and customized for specific products (such as corn). Generally, different environmental characteristics can be regarded as the main driving factors in terms of the yield of different products. In one embodiment, corn growth is mainly driven by heat units or growth degree units. Fig. 20 The process 2000 can be used to design features for a specific product to exploit knowledge about key features or otherwise develop and enhance data to provide quality results in recommendation modeling. For example, engineering features can be used at block 1210 to generate field-specific seed recommendations for a target field or environment.
[0260] 6.1 Original Features and Feature Classification
[0261] At box 2002, for example, the agricultural intelligent computer system 130 identifies the original features that are important drivers of yield, and further performs feature classification to convert continuous features into categorical features. The original features can be derived from general categories including topography and hydrology, weather, management practices, and soil properties. For example, the features derived from topography and hydrology include: altitude, slope, contour curve (concave and convex characteristics), aspect (the direction of the wheel facing the slope), distance from the water source, soil EC500, etc. The weather-derived features can be related to day length, temperature, evapotranspiration, rainfall, solar characteristics, drought index, etc. The management-derived features can include planting time, harvesting time, seeds planted per acre, seed product segmentation, seed MAC zone, seed position maturity group zone, etc. For example, the soil-derived features can quantify or characterize organic matter, texture category, sand / clay percentage, permeability and bulk density, CEC (cation exchange capacity), PAW (plant available water) and soil productivity index.
[0262] At block 2002, the agricultural intelligence computer system 130 identifies one or more features that drive yield for a particular product. According to an embodiment, the system 130 converts otherwise continuous features into categorical features by characterizing at least the identified key features into a smaller number of different feature categories. The system 130 can then use the original continuous features to identify the key features that drive yield for a particular product. Fig.12 This classification of key features is used in recommendation modeling to enhance the results.
[0263] Using corn as an example, key characteristics include soil and topographic features. Based on field data and scientific studies in different environments (e.g., Indiana, Illinois, Iowa, Minnesota, Missouri, and Wisconsin), soil and topographic features can be classified according to the examples in Table 1:
[0264]
[0265]
[0266]
[0267] Table 1 - Example feature classification
[0268] 6.2 Data Preparation
[0269] At block 2004, for example, the agricultural intelligence computer system 130 receives agricultural data records via the digital data communication network 109. For example, the agricultural data records include crop seed data and yield attributes of the seeds and environmental data in which the seeds were planted and / or tested. At block 2006, the system 130 further provides a machine learning model (e.g., Fig.12 According to an embodiment, the system 130 uses the key feature classification of block 2002 to characterize the received environmental features in preparation for use in the machine learning model. Illustratively, as a result of block 2004, the system 130 associates yield attributes of seeds from multiple fields with the key feature classification corresponding to specific field conditions.
[0270] At block 2006, the system 130 may also perform filtering to extract more important data for use in recommendation modeling. In the case of corn, fields where multiple products are tested or planted in the same field may find a large amount of yield data compared to fields where only one product or a relatively small number of products are tested. For example, at block 2006, the system 130 may extract agricultural data records only for fields with six or more products tested simultaneously, and perform recommendation modeling on only this extracted data using key feature classification.
[0271] According to an embodiment, the system 130 uses the processed agricultural data records to generate GxE relationships between genetic characteristics of seeds, field characteristics, and yield, for example, using some form of BLUP model (e.g., an environmental best linear unbiased prediction (eBLUP) model), T-stat, and / or kernel smoothing using Gaussian processes. For example, the system 130 may also use the processed agricultural data to generate GxE relationships between genetic characteristics of seeds, field characteristics, and yield, for example, using some form of BLUP model (e.g., an environmental best linear unbiased prediction (eBLUP) model), T-stat, and / or kernel smoothing using Gaussian processes. Fig.12 The data gaps can be filled in by using block 1206 of the present invention. As described above, the GxE relationships and / or estimation data can be used to generate predicted yield performance for one or more seeds for one or more specific target fields, thereby generating field-level yield improvement recommendations.
Claims
1. A computer-implemented method comprising: receiving, via a digital data communications network at a server computer system, an agricultural data record comprising a first set of yield attributes for a first set of seeds grown in a first set of environments; receiving, via the digital data communications network, genetic characteristic data associated with a second set of seeds, wherein the second set of seeds includes the first set of seeds; generating, using the server computer system, a second set of yield attributes for the second set of seeds associated with a second set of environments by applying a model using the genetic characteristic data and the agricultural data records, wherein the second set of yield attributes fills one or more data gaps from the first set of yield attributes; using the server computer system to generate a predicted yield performance for the third seed set by applying feature engineering to develop a genomic-by-environment signature for the third seed set and using the second set of yield attributes and the genomic-by-environment signature in a machine learning model to determine the predicted yield performance for the third seed set associated with one or more target environments; generating, using the server computer system, seed recommendations for the one or more target environments based on the predicted yield performance for the third set of seeds; as well as Displaying the seed recommendation on a display device is caused, the display device being communicatively coupled to the server computer system.
2. The computer-implemented method of claim 1 , wherein the genetic signature data comprises genomic marker data, and wherein generating the second set of yield attributes comprises applying the model using the genomic marker data.
3. The computer-implemented method of claim 1 , wherein the genetic characteristic data comprises a pedigree-based kinship matrix, and wherein generating the second set of yield attributes comprises applying the model using the pedigree-based kinship matrix.
4. The computer-implemented method of claim 1, wherein the genetic signature data comprises gene cluster relationship data, and wherein generating the second set of yield attributes comprises applying the model using the gene cluster relationship data.
5. The computer-implemented method of claim 1 , wherein the genetic characteristic data comprises a genetic marker-based kinship matrix, and wherein generating the second set of yield attributes comprises applying the model using the genetic marker-based kinship matrix. 6 . The computer-implemented method of claim 1 , wherein generating the predicted yield performance for the third set of seeds comprises applying inbreeding coding to associate genomic-by-environment characteristics with different seeds.
7. The computer-implemented method of claim 1, wherein generating seed recommendations for the one or more target environments is further based on one or more of the following factors: hybrid or inbred genetic hybrid populations, genetic markers associated with biotech traits or quantitative trait loci, genome-wide genetic markers, long-form haplotypes, best linear unbiased prediction of inbreeding-general combining ability yield, yield-related phenotypes, or hybrid or inbred disease characteristics.
8. The computer-implemented method of claim 1, wherein feature engineering Further including: Convert continuous environmental features into one or more distinct feature classes; characterizing environmental characteristics associated with the agricultural data record using the one or more different feature classes; The characterized environmental features are used in the machine learning model to generate the predicted yield performance.
9. The computer-implemented method of claim 8, wherein the feature The engineering further includes: for only one or more agricultural data records having a plurality of seeds grown in a given environment, characterizing the environmental features associated with the agricultural data record using the one or more different feature classes.
10. One or more non-transitory computer-readable storage media storing instructions that, when executed by one or more processors, cause operations to be performed, the operations comprising: receiving an agricultural data record comprising a first set of yield attributes for a first set of seeds grown in a first set of environments; receiving genetic characteristic data associated with a second set of seeds, wherein the second set of seeds includes the first set of seeds; generating a second set of yield attributes for the second set of seeds associated with a second set of environments by applying a model using the genetic characteristic data and the agricultural data records, wherein the second set of yield attributes fills one or more data gaps from the first set of yield attributes; generating a predicted yield performance for the third seed set by applying feature engineering to develop a genomic-by-environment signature for the third seed set and using the second set of yield attributes and the genomic-by-environment signature in a machine learning model to determine the predicted yield performance for the third seed set associated with one or more target environments; generating seed recommendations for the one or more target environments based on the predicted yield performance for the third set of seeds; and Causing display of the seed recommendation.
11. The one or more non-transitory computer-readable storage media of claim 10, wherein the genetic signature data comprises genomic marker data, and wherein the operation of generating the second set of yield attributes comprises applying the model using the genomic marker data.
12. One or more non-transitory computer-readable storage media according to claim 10, wherein the genetic characteristic data includes a pedigree-based kinship matrix, and wherein the operation of generating the second set of yield attributes includes applying the model using the pedigree-based kinship matrix.
13. The one or more non-transitory computer-readable storage media of claim 10, wherein the genetic signature data comprises gene cluster relationship data, and wherein the operation of generating the second set of yield attributes comprises applying the model using the gene cluster relationship data.
14. The one or more non-transitory computer-readable storage media of claim 10, wherein the genetic characteristic data comprises a genetic marker-based kinship matrix, and wherein the operation of generating the second set of yield attributes comprises applying the model using the genetic marker-based kinship matrix.
15. The one or more non-transitory computer-readable storage media of claim 10, wherein the operation of generating the predicted yield performance for the third set of seeds comprises the operation of applying inbreeding coding to associate genomic-by-environment characteristics with different seeds.
16. One or more non-transitory computer-readable storage media according to claim 10, wherein the operation of generating seed recommendations for the one or more target environments is further based on one or more of the following factors: hybrid or inbred genetic hybrid populations, genetic markers associated with biotech traits or quantitative trait loci, genome-wide genetic markers, long-form haplotypes, best linear unbiased prediction of inbreeding-general combining ability yield, yield-related phenotypes, or hybrid or inbred disease characteristics.
17. The one or more non-transitory computer-readable storage media of claim 10, wherein the operation of applying feature engineering further comprises the following operations: Convert continuous environmental features into one or more distinct feature classes; characterizing environmental characteristics associated with the agricultural data record using the one or more different feature classes; The characterized environmental features are used in the machine learning model to generate the predicted yield performance.
18. One or more non-transitory computer-readable storage media according to claim 10, wherein the operation of applying feature engineering further includes the following operation: only for one or more agricultural data records having multiple seeds grown in a given environment, using the one or more different feature classes to characterize the environmental characteristics associated with the agricultural data records.
