Using machine learning-based seed harvest moisture predictions to improve a computer-assisted agricultural farm operation

Patent metadata

Field	Value
Publication number	US-11017306-B2
Application number	US-201916661860-A
Country	US
Kind code	B2
Filing date	Oct 23, 2019
Priority date	Oct 24, 2018
Publication date	May 25, 2021
Grant date	May 25, 2021

Abstract

Embodiments generate digital plans for agricultural fields. In an embodiment, a model receives digital inputs including stress risk data, product maturity data, field location data, planting date data, and/or harvest date data. The model mathematically correlates sets of digital inputs with threshold data associated with the stress risk data. The model is used to generate stress risk prediction data for a set of product maturity and field location combinations. In a digital plan, product maturity data or planting date data or harvest date data or field location data can be adjusted based on the stress risk prediction data. A digital plan can be transmitted to a field manager computing device. An agricultural apparatus can be moved in response to a digital plan.

First claim

What is claimed is:

1. A computer-implemented method for generating an improved digital plan for agricultural fields associated with a grower operation during a growing season, the method comprising:
receiving, by a predictive model, a set of digital inputs relating to a digital plan;
wherein the set of digital inputs comprises, for products to be planted in the agricultural fields, stress risk data and at least one of product maturity data, field location data, planting date data, harvest date data;
wherein the predictive model has been trained to mathematically correlate sets of digital inputs with predictive threshold data that is associated with the stress risk data;
using the predictive model, generating, as digital output in response to the set of digital inputs, stress risk prediction data for a set of product maturity and field location combinations;
wherein the stress risk prediction data indicates a mathematical likelihood of actual harvest data matching desired harvest data on a particular date during the growing season;
wherein the actual harvest data includes product moisture measured at harvest resulting from planting of a product in accordance with the digital plan;
wherein the desired harvest data indicates moisture associated with a desired product yield;
creating and digitally storing an improved digital plan by adjusting the product maturity data or the planting date data or the harvest date data or the field location data based on the stress risk prediction data;
transmitting the improved digital plan to a field manager computing device associated with the grower operation and causing display of the improved digital plan on a display of the field manager computing device or causing movement of an agricultural apparatus in response to the improved digital plan.

2. The method of claim 1 , wherein the stress risk prediction data indicates a prediction of any one or more of yield, harvest moisture, field readiness for planting.

3. The method of claim 1 , wherein the digital plan identifies a distribution of product maturities across the agricultural fields, and the improved digital plan is created by changing the distribution of product maturities across the agricultural fields based on the stress risk prediction data.

4. The method of claim 1 , wherein the digital plan indicates an assignment of a product maturity to an agricultural field and the improved digital plan is created by changing the assignment of the product maturity to the agricultural field based on the stress risk prediction data.

5. The method of claim 1 , wherein the digital plan indicates an assignment of a planting date or a harvest date to an agricultural field and the improved digital plan is created by changing the assignment of the planting date or the harvest date to the agricultural field based on the stress risk prediction data.

6. The method of claim 1 , further comprising, using the predictive model, wherein the stress risk data is associated with at least two different stages of a grower operation for a particular agricultural field.

7. The method of claim 1 , further comprising adjusting the improved digital plan in response to changes in weather forecast data obtained after the digital plan has been created.

8. The method of claim 1 , wherein at least some of the set of digital inputs are received using electronic communication with an agricultural apparatus.

9. The method of claim 1 , wherein the improved digital plan comprises a digital visualization of product maturity allocations to agricultural fields.

10. The method of claim 1 , wherein the predictive model is trained using a computer-implemented supervised machine learning algorithm comprising a random forest algorithm or a gradient boosting algorithm.

11. One or more non-transitory storage media storing instructions which, when executed by one or more computing devices, cause performance of operations recited in claim 1 .

12. A computer system configured to cause performance of operations recited in claim 1 .

13. A computer-implemented method for generating an improved digital plan for an agricultural field associated with a grower operation and a product planted in the agricultural field during a growing season, the method comprising:
receiving, by a predictive model, a set of digital inputs relating to a digital plan and a prediction date that is after the product has been planted in the agricultural field during the growing season;
wherein the set of digital inputs comprises, for the agricultural field, observed weather data collected during the growing season to the prediction date and forecasted weather data computed for a future date range after the prediction date during the growing season, and for the product planted in the agricultural field, planting date data and product maturity data;
wherein the predictive model has been trained to mathematically correlate sets of digital inputs with threshold data associated with harvest moisture;
using the predictive model, generating, as digital output in response to the set of digital inputs, harvest moisture prediction data for the future date range for the product planted in the agricultural field;
wherein the harvest moisture prediction data indicates a mathematical likelihood of actual harvest moisture data matching desired harvest moisture data during the future date range;
wherein the actual harvest data includes product moisture measured at harvest resulting from planting of the product in accordance with the digital plan;
wherein the desired harvest moisture data indicates moisture associated with a desired product yield;
creating and digitally storing an improved digital planting plan including a harvest date recommended based on the harvest moisture prediction data;
transmitting the improved digital plan to a field manager computing device associated with the grower operation and causing display of the improved digital plan on a display of the field manager computing device or causing movement of an agricultural apparatus in response to the improved digital plan.

14. The computer-implemented method of claim 13 , further comprising:
in response to a new prediction date following the prediction date, receiving, by a predictive model, a new set of digital inputs relating to the digital plan and the new prediction date;
using the predictive model, generating, as digital output in response to the new set of digital inputs, new harvest moisture prediction data for a new future date range for the product planted in the agricultural field;
creating and digitally storing a new improved digital plan by adjusting the harvest date based on the new harvest moisture prediction data;
transmitting the new improved digital plan to a field manager computing device associated with the grower operation and causing display of the new improved digital plan on a display of the field manager computing device.

15. The computer-implemented method of claim 14 , further comprising repeating the receiving, generating, creating, and transmitting on a periodic basis or a daily basis.

16. The computer-implemented method of claim 13 , further comprising:
inputting the harvest moisture prediction data into a second predictive model;
using the second predictive model, generating, for a product and agricultural field combination, recommendation data comprising at least one of a relative maturity recommendation, a planting date recommendation, and a harvest date recommendation;
transmitting the recommendation data to a computing device associated with the grower operation and causing display of th

Assignees

• Climate Corp

Inventors

• Sood Shilpa
• Sorge Matthew
• Shah Nikisha
• Reich Timothy
• Ssegane Herbert
• Bull Jason Kendrick
• Ehlmann Tonya S
• Jacobs Morrison
• Macisaac Susan Andrea
• Schnicker Bruce J
• Xie Yao
• Trapp Allan
• Yang Xiao

Classifications

• G06N5/01

  Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound · CPC title

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06Q50/02

  Agriculture; Fishing; Forestry; Mining · CPC title

• G06Q10/0631

  Resource planning, allocation, distributing or scheduling for enterprises or organisations · CPC title

• G06Q10/04

  Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" (market predictions or forecasting for commercial activities G06Q30/0202) · CPC title