Method and system for estimating performance for position-specific control of an agricultural machine

Patent metadata

Field	Value
Publication number	US-12575475-B2
Application number	US-202318446056-A
Country	US
Kind code	B2
Filing date	Aug 8, 2023
Priority date	Apr 14, 2023
Publication date	Mar 17, 2026
Grant date	Mar 17, 2026

Abstract

In one embodiment, obtained moisture data, planting data, seeding rate and seeding depth, and planted seedbed/furrow quality (e.g., first through fourth data layers associated with select data spear zones distributed throughout the field to conserve data processing resources) are applied, inputted or transmitted wirelessly to a central server, with an electronic data processing device that is configured to estimate a yield of the crop at a row position basis throughout the field.

First claim

The following is claimed:

1 . A method for determining the yield of a crop in a field, the method comprising:
obtaining moisture data during a growing period based on sensor data applicable to the field; obtaining planting date data applicable to planting of a seed or plant in the field for the growing period; measuring at least one of a seeding rate or a seeding depth for each row of a planter indexed to a position of the row unit estimated by row position data provided by a location-determining receiver; measuring, by an imaging device in conjunction with an image processing system, a planted seedbed/furrow quality associated with the row position data; and applying or inputting the obtained moisture data, planting data, seeding rate or seeding depth, and planted seedbed/furrow quality to an electronic data processor of a central server, the electronic data processor that is configured to estimate a yield of the crop at a row position basis throughout the field; wherein a first data layer comprises the obtained moisture data; a second data layer comprises the obtained planting date data; a third data layer comprises the seeding rate or seeding depth; a fourth data layer comprises a planted seedbed/furrow quality associated with the row position data; and selecting sample zones for obtaining the first data layer, the second data layer, the third data layer and the fourth data layer based on a variance, deviation or change in the first data layer exceeding a threshold during a sampling interval or a successive series of sampling intervals.

2 . The method according to claim 1 wherein the first data layer, the second data layer, the third data layer, and the fourth data layer are associated with selected sample zones spatially distributed and spaced apart throughout the field to conserve data processing resources.

3 . The method according to claim 2 further comprising: culling or deleting a first portion of non-selected data within the first data layer, the second data layer, the third data layer and the fourth data layer for areas in the field outside of the selected sample zones to conserve data processing resources, wherein a second portion of the non-selected data is retained or stored for average, mode, median or benchmark yield estimations.

Assignees

• Deere & Co

Inventors

• Casper Robert T
• Hartsock Nathaniel J
• Krogh Ryan M
• Sauder Doug S

Classifications

• G06T2207/30188

  Vegetation; Agriculture · CPC title

• G06T2207/20084

  Artificial neural networks [ANN] · CPC title

• G06T2207/20081

  Training; Learning · CPC title

• G06V20/188

  Vegetation · CPC title

• G05D1/0278

  using satellite positioning signals, e.g. GPS · CPC title