Precision agriculture system

Patent metadata

Field	Value
Publication number	US-9959507-B2
Application number	US-201615149888-A
Country	US
Kind code	B2
Filing date	May 9, 2016
Priority date	Jan 14, 2015
Publication date	May 1, 2018
Grant date	May 1, 2018

Abstract

A device may receive sensor data from a sensor device located on a particular farm. The device may identify an alert, associated with the particular farm, based on the sensor data and using a model. The model may be created based on imagery data and numeric data relating to a group of farms. The device may determine, using the model, a recommended course of action to address the alert, and provide, to a user device associated with the particular farm, the recommended course of action.

First claim

What is claimed is:

1. A method, comprising:
capturing, by a processor of a first device, from a plurality of sensors, and via a network, first data regarding one or more farms,
the plurality of sensors including:
a ground-based sensor, and
an aerial-based sensor,

the first data including:
ground-related data provided by the ground-based sensor, and aerial-related data provided by the aerial-based sensor;

receiving, by the processor, from a plurality of sources, and via the network, second data regarding the one or more farms,
the plurality of sources being different than the plurality of sensors,
the second data being different than the first data, and
the second data including at least one of:
imagery data, or
numeric data;

processing, by the processor, the first data to create processed first data;
processing, by the processor, the second data to create processed second data;
creating, by the processor, a model,
the model being created using a machine learning technique, the processed first data, and the processed second data;

training, by the processor, the model based on training data,
the training data including:
information relating to an action taken, and
information identifying a result of taking the action;

receiving, by the processor and via the network, sensor data associated with a geographic location;
determining, by the processor and using the model and the sensor data, a recommended course of action in relation to a plot of the one or more farms,
the recommended course of action including gathering additional data regarding the plot;

providing, by the processor and for display, a map that visually identifies an alert relating to the plot,
the alert indicating that a potential issue exists regarding the plot or information regarding harvesting the plot;

receiving, by the processor, a selection of the alert;
providing, by the processor and for display, information identifying the recommended course of action based on receiving the selection of the alert;
receiving, by the processor, a selection relating to the recommended course of action; and
causing, by the processor and based on receiving the selection relating to the recommended course of action, a second device, associated with the geographic location, to perform the recommended course of action,
the second device being an unmanned aerial vehicle, and
the second device being different than the first device.

2. The method of claim 1 ,
where the geographic location is the plot,
where the gathering of the additional data includes capturing an image of the plot, and
where, when causing the second device to perform the recommended course of action, the method includes:
causing the unmanned aerial vehicle to capture the image of the plot.

3. The method of claim 1 ,
where the geographic location is the plot, and
where, when creating the model, the method includes:
creating a plurality of models for the plot,
the plurality of models including:
a first model that is associated with a first portion of the plot, and
a second model that is different than the first model and that is associated with a second portion of the plot,
 the first portion and the second portion corresponding to different portions of the plot or different crops of the plot.

4. The method of claim 1 , further comprising:
causing the second device or a third device, associated with the geographic location, to perform a different course of action,
where at least one of the recommended course of action or the different course of action relates to at least one of:
maintenance of a piece of farming equipment associated with the plot,
the plot relating to the geographic location,

malfunction of the piece of farming equipment associated with the plot,
harvest of a crop associated with the plot,
storage of the crop associated with the plot,
sale of the crop associated with the plot,
detection of a disease in the plot,
detection of an insect issue in connection with the plot, or
an irrigation issue of the plot.

5. The method of claim 1 ,
where, when determining the recommended course of action, the method includes:
determining a plurality of recommended courses of action;
determining, for each recommended course of action of the plurality of recommended courses of action, an impact of performing the recommended course of action or not performing the recommended course of action; and
ranking the plurality of recommended courses of action, based on determining the impact for each recommended course of action, to create a ranked list; and

where, when providing the information identifying the recommended course of action, the method includes:
providing information identifying the ranked list to a user device.

6. The method of claim 1 , further comprising:
causing a third device, associated with the geographic location, to perform a different course of action,
where the geographic location is the plot,
where the third device includes an irrigation system,
where the different course of action relates to watering of the plot, and
where, when causing the third device to perform the different course of action, the method includes:
causing the irrigation system to water the plot for a particular period of time.

7. The method of claim 1 ,
where the recommended course of action relates to an issue in connection with the plot, and
where the method further comprises:
providing, for display, a user interface,
the user interface including:
aerial imagery of the plot at a point in time, and
an interface that allows aerial imagery of the plot to be viewed at different points in time.

8. A non-transitory computer-readable medium storing instructions, the instructions comprising:
a plurality of instructions, which, when executed by a processor of a first device, cause the processor to:
capture, from a plurality of sensors and via a network, first data regarding one or more farms,
the plurality of sensors including:
a ground-based sensor, and
an aerial-based sensor,

the first data including:
ground-related data provided by the ground-based sensor, and
aerial-related data provided by the aerial-based sensor;

receive, from a plurality of sources and via the network, second data regarding the one or more farms,
the plurality of sources being different than the plurality of sensors,
the second data being different than the first data, and
the second data including at least one of:
imagery data, or
numeric data;

process the first data to create processed first data;
process the second data to create processed second data;
create a model,
the model being created using a machine learning technique,

the processed first data, and the processed second data;
receive, via the network, sensor data associated with a geographic location;
determine, using the model and the sensor data, a recommended course of action in relation to a plot of the one or more farms,
the recommended course of action including gathering additional data regarding the plot;

provide, for display, a map that visually identifies an alert relating to the plot,
the alert indicating that a potential issue exists regarding the plot or information regarding harvesting the plot;

receive a selection of the alert;
provide, for display, information identifying the recommended course of action based on receiving the selection of the alert;
receive a selection relating to the recommended course of action; and
cause, based on receiving the selection relating to the recommended course of action, a second device, associated with the geographic location, to perform the recommended course of action,
the second device

Assignees

• Accenture Global Services Ltd

Inventors

• Mathur Ankur
• Barsamian Paul M
• Garrison Daniel P
• Mullan Pramila
• Mendez Juan C

Classifications

• G06N7/01

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

• G01C11/025

  by scanning the object · CPC title

• G06F30/00

  Computer-aided design [CAD] · CPC title

• A01G25/00

  Watering gardens, fields, sports grounds or the like (special apparatus or adaptations for fertilising-liquids A01C23/00; nozzles or outlets, spraying apparatus B05B) · CPC title

• G06Q50/02 Primary

  Agriculture; Fishing; Forestry; Mining · CPC title