IP Library Granted Patent US 12,696,890
Granted Patent B1
US 12,696,890 · App. 18/948,319 · Granted Aug 4, 2026

Termite detection and mitigation system

Inventors: Bradly Jay Billman (Celina, TX); Brian Tougas (Spring Branch, TX); Arthur Quentin Smith (Fredericksburg, TX); Mark Paxman Warnick (San Antonio, TX); Oscar Roberto Tijerina (San Antonio, TX)
Assignee: United Services Automobile Association (USAA)
A01M1/026A01M1/2011G06N3/08G08B21/182
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,696,890
App. No.
18/948,319
Granted
Aug 4, 2026
Kind
B1
Abstract

A termite detection and mitigation system can gather data describing conditions at a location, including the location's elevation, soil type, topography, weather conditions, and season. The termite detection and mitigation system can additionally obtain sensor data describing environmental conditions within a structure (e.g., a home or other building) at the location. Sensor data may be obtained using a termite sensing device placed within the structure. The termite detection and mitigation system can develop models that characterize how likely it is that termites are present within a structure based on its sensor data and associated location data. Using the models, the termite detection and mitigation system can alert users when there is a likelihood that termites are present. The termite detection and mitigation system can also deploy mitigation responses, such as poisons or sealants, to mitigate the termite intrusion.

Claims (64)

1 . A method comprising:

generating a termite detection score model by:

generating one or more training items based on location-related data and sensor data, at least one of the one or more training items including

a collection of words or statistics representing environmental conditions of a respective structure at a geographic area, and

a label specifying a respective termite detection score for that respective structure;

applying the at least one of the one or more training items to a neural network model;

comparing output from the neural network model for each respective structure to the label specifying the respective termite detection score for that respective structure; and

based on the comparison, applying a loss function to modify aspects of the neural network model;

generating, based on the termite detection score model, a termite detection score corresponding to a structure associated with a user; and

generating a termite presence alert for the user based on the termite detection score.

2 . The method of claim 1 , further comprising:

receiving the location-related data, wherein the location-related data comprises elevation information, soil type, topographical information, current weather information, seasonal conditions information, or a combination thereof, and

receiving the sensor data, wherein the sensor data comprises sound information, motion information, heat information, moisture information, chemical information, or a combination thereof.

3 . The method of claim 1 , further comprising:

generating the termite detection score corresponding to the structure associated with the user based on the location-related data and the sensor data associated with the structure associated with the user, wherein the sensor data associated with the structure associated with the user is obtained from a termite sensing device located in the structure.

4 . The method of claim 1 , further comprising:

determining a mitigation response comprises at least one of a poison or a sealant; and

transmitting a mitigation instruction, based on the mitigation response, to a termite sensing device located in the structure.

5 . The method of claim 1 , wherein the termite detection score model is a statistical model generated by applying a regression analysis on the location-related data and the sensor data.

6 . The method of claim 1 , wherein the termite detection score model is a machine learning model generated by (A) transforming the location-related data and the sensor data associated with the structure associated with the user into one or more statistical models and (B) training the machine learning model based on the location-related data and the sensor data.

7 . The method of claim 1 , wherein the termite detection score characterizes a likelihood of a presence of termites at the structure.

8 . A system comprising:

one or more processors; and

one or more memories storing instructions that, when executed by the one or more processors, cause the system to perform a process comprising:

generating a termite detection score model by:

generating one or more training items based on location-related data and sensor data, at least one of the one or more training items including

a collection of words or statistics representing environmental conditions of a respective structure at a geographic area, and

a label specifying a respective termite detection score for that respective structure;

applying the at least one of the one or more training items to a neural network model;

comparing output from the neural network model for each respective structure to the label specifying the respective termite detection score for that respective structure; and

based on the comparison, applying a loss function to modify aspects of the neural network model;

generating, based on the termite detection score model, a termite detection score corresponding to a structure associated with a user; and

generating a termite presence alert for the user based on the termite detection score.

9 . The system of claim 8 , wherein the process further comprises:

receiving the location-related data, wherein the location-related data comprises elevation information, soil type, topographical information, current weather information, seasonal conditions information, or a combination thereof, and

receiving the sensor data, wherein the sensor data comprises sound information, motion information, heat information, moisture information, chemical information, or a combination thereof.

10 . The system of claim 8 , wherein the process further comprises:

generating the termite detection score corresponding to the structure associated with the user based on the location-related data and the sensor data associated with the structure associated with the user, wherein the sensor data associated with the structure associated with the user is obtained from a termite sensing device located in the structure.

11 . The system of claim 8 , wherein the process further comprises:

determining a mitigation response comprises at least one of a poison or a sealant; and

transmitting a mitigation instruction, based on the mitigation response, to a termite sensing device located in the structure.

12 . The system of claim 8 , wherein the termite detection score model is a statistical model generated by applying a regression analysis on the location-related data and the sensor data.

13 . The system of claim 8 , wherein the termite detection score model is a machine learning model generated by (A) transforming the location-related data and the sensor data associated with the structure associated with the user into one or more statistical models and (B) training the machine learning model based on the location-related data and the sensor data.

14 . The system of claim 8 , wherein the termite detection score characterizes a likelihood of a presence of termites at the structure.

15 . A non-transitory computer-readable medium storing instructions that, when executed by a computing system, cause the computing system to perform operations comprising:

generating a termite detection score model by:

generating one or more training items based on location-related data and sensor data, at least one of the one or more training items including

a collection of words or statistics representing environmental conditions of a respective structure at a geographic area, and

a label specifying a respective termite detection score for that respective structure;

applying the at least one of the one or more training items to a neural network model;

comparing output from the neural network model for each respective structure to the label specifying the respective termite detection score for that respective structure; and

based on the comparison, applying a loss function to modify aspects of the neural network model;

generating, based on the termite detection score model, a termite detection score corresponding to a structure associated with a user; and

generating a termite presence alert for the user based on the termite detection score.

16 . The non-transitory computer-readable medium of claim 15 , wherein the operations further comprise:

receiving the location-related data, wherein the location-related data comprises elevation information, soil type, topographical information, current weather information, seasonal conditions information, or a combination thereof; and

receiving the sensor data, wherein the sensor data comprises sound information, motion information, heat information, moisture information, chemical information, or a combination thereof.

17 . The non-transitory computer-readable medium of claim 15 , wherein the operations further comprise:

generating the termite detection score corresponding to the structure associated with the user based on the location-related data and the sensor data associated with the structure associated with the user, wherein the sensor data associated with the structure associated with the user is obtained from a termite sensing device located in the structure.

18 . The non-transitory computer-readable medium of claim 15 , wherein the operations further comprise:

determining a mitigation response comprises at least one of a poison or a sealant; and

transmitting a mitigation instruction, based on the mitigation response, to a termite sensing device located in the structure.

19 . The non-transitory computer-readable medium of claim 15 , wherein the termite detection score model is a statistical model generated by applying a regression analysis on the location-related data and the sensor data.

20 . The non-transitory computer-readable medium of claim 15 , wherein the termite detection score model is a machine learning model generated by (A) transforming the location-related data and the sensor data associated with the structure associated with the user into one or more statistical models and (B) training the machine learning model based on the location-related data and the sensor data.