Training method for an agricultural vehicle automation system

Patent metadata

Field	Value
Publication number	US-12588592-B2
Application number	US-202418432757-A
Country	US
Kind code	B2
Filing date	Feb 5, 2024
Priority date	Feb 6, 2023
Publication date	Mar 31, 2026
Grant date	Mar 31, 2026

Abstract

An automation system is configured to automatically adjust an operational setting of an agricultural vehicle while performing an operation in a field in order to adjust an operational output parameter of the agricultural vehicle. A method of training the automation system comprises receiving operational information including at least a location of the agricultural vehicle and georeferenced field information. The method further comprises, based on the location of the agricultural vehicle and the georeferenced field information, deciding to run a training sequence. The training sequence comprises measuring the operational output parameter, changing the operational setting, monitoring a subsequent change of the operational output parameter, and updating the automation system based on the changing of the operational setting and the subsequent change of the operational parameter.

First claim

The invention claimed is:

1 . A method of training an automation system for an agricultural vehicle, the method comprising:
receiving operational information including a fuel tank filling level and fuel consumption, and further including at least a location of the agricultural vehicle and georeferenced field information, and deciding to run a training sequence for the automation system while performing an agricultural operation in a field based on the location of the agricultural vehicle and the georeferenced field information; wherein the training sequence comprises:
measuring an operational output parameter of the agricultural vehicle,
changing an operational setting of the agricultural vehicle,
actively controlling an operation of the agricultural vehicle based on the changing of the operational setting,
monitoring a subsequent change of the operational output parameter, and
updating the automation system based on the changing of the operational setting and the subsequent change of the operational output parameter; and
wherein the deciding to run the training sequence comprises:
determining a predicted fuel tank filling level based on the fuel consumption, and
deciding to run the training sequence based on the predicted fuel tank filling level.

2 . The method of training then automation system as claimed in claim 1 , wherein the deciding to run the training sequence comprises:
based on the operational information, determining a stable training area that is suitable for running the training sequence, determining a size of the stable training area in front of the agricultural vehicle, and deciding to run the training sequence based on the size of the stable training area in front of the agricultural vehicle.

3 . The method of training the automation system as claimed in claim 2 , wherein the determining the size of the stable training area comprises:
determining a predicted variance of an agricultural vehicle parameter, or of a field parameter, the predicted variance pertaining to a traversal of a selected area of the field by the agricultural vehicle while performing the agricultural operation, and adding the selected area to the stable training area if the predicted variance is below a predetermined threshold variance.

4 . The method of training the automation system as claimed in claim 3 , wherein the automation system is updated based on the changing of the operational setting and the subsequent change of the operational output parameter, weighed with a factor inversely proportional with the predicted variance.

5 . The method of training the automation system as claimed in claim 1 , wherein the georeferenced field information includes at least one of:
a predicted field inclination, a predicted field condition, or a distance to an end of the field.

6 . The method of training the automation system as claimed in claim 1 , wherein the georeferenced field information includes at least one of:
a predicted crop yield per unit area, a predicted crop density, a predicted crop height, a predicted crop moisture level, or a predicted weed density and/or height.

7 . The method of training the automation system as claimed in claim 1 , wherein the operational information further includes a current and/or predicted position of a second agricultural vehicle.

8 . An automation system for an agricultural vehicles, the automation system comprising:
an agricultural vehicle that performs an agricultural operation within a field; and a controller that actively controls an operation of the agricultural vehicle during the performance of the agricultural operation, the controller: receives operational information including a fuel tank filling level and fuel consumption, and further including at least a location of the agricultural vehicle and georeferenced field information, and decides to run a training sequence while performing the agricultural operation in the field based on the location of the agricultural vehicle and the georeferenced field information; wherein the training sequence comprises:
measuring an operational output parameter of the agricultural vehicle,
changing an operational setting of the agricultural vehicle,
actively controlling the operation of the agricultural vehicle based on the changing of the operational setting,
monitoring a subsequent change of the operational output parameter, and
updating the automation system based on the changing of the operational setting and the subsequent change of the operational output parameter
wherein the deciding to run the training sequence comprises:
determining a predicted fuel tank filling level based on the fuel consumption, and
deciding to run the training sequence based on the predicted fuel tank filling level.

9 . The automation system as claimed in claim 8 wherein the agricultural vehicle is a tractor, a combine harvester, a forage harvester, or a tractor-baler combination.

10 . The automation system as claimed in claim 8 , wherein, when deciding to run the training sequence, the controller:
based on the operational information, determines a stable training area that is suitable for running the training sequence, determines a size of the stable training area in front of the agricultural vehicle, and decides to run the training sequence based on the size of the stable training area in front of the agricultural vehicle.

11 . The automation system as claimed in claim 10 , wherein, when determining the size of the stable training area, the controller:
determines a predicted variance of an agricultural vehicle parameter or of a field parameter, the predicted variance pertaining to a traversal of a selected area of the field by the agricultural vehicle while performing the agricultural operation, and adds the selected area to the stable training area if the predicted variance is below a predetermined threshold variance.

12 . The automation system as claimed in claim 11 , wherein the controller updates the automation system based on the changing of the operational setting and the subsequent change of the operational output parameter, weighed with a factor inversely proportional with the predicted variance.

13 . The automation system as claimed in claim 8 , wherein the georeferenced field information includes at least one of:
a predicted field inclination, a predicted field condition, or a distance to an end of the field.

14 . The automation system as claimed in claim 8 , wherein the georeferenced field information includes at least one of:
a predicted crop yield per unit area, a predicted crop density, a predicted crop height, a predicted crop moisture level, or a predicted weed density and/or height.

15 . The automation system as claimed in claim 14 , wherein the georeferenced field information includes the predicted crop yield per unit area, and, when deciding to run the training sequence, the controller:
determines a predicted grain tank filling level based on the predicted crop yield per unit area, and decides to run the training sequence based on the predicted grain tank filling level.

16 . The automation system as claimed in claim 8 , wherein the operational information further includes a current and/or predicted position of a second agricultural vehicle.

17 . A method of training an automation system for an agricultural vehicle, the method comprising:
receiving operational information including at least a location of the agricultural vehicle and georeferenced field information, the georeferenced field informati

Assignees

• Cnh Ind Belgium Nv

Inventors

• Missotten Bart M A
• Maelegheer Pieter
• Van Campenhout Simon
• Vandewalle Bert
• Leenknegt Arno
• Mahieu Thomas

Classifications

• A01B79/02

  combined with other agricultural processing, e.g. fertilising, planting · CPC title

• A01B79/005

  Precision agriculture · CPC title

• A01B69/001

  Steering by means of optical assistance, e.g. television cameras (steering devices for road marking vehicles E01C23/163) · CPC title

• A01D41/1275 Primary

  for the level of grain in grain tanks · CPC title

• A01B76/00 Primary

  Parts, details or accessories of agricultural machines or implements, not provided for in groups A01B51/00 - A01B75/00 · CPC title