IP Library Granted Patent US 12,536,639
Granted Patent B1
US 12,536,639 · App. 19/043,962 · Granted Jan 27, 2026

Computer-implemented system and method for automating inspections

Inventors: Kevin Kalajan (Arroyo Grande, CA); Dane Demicell (Watsonville, CA); Sam Elhag (San Diego, CA); Marcus Karr (San Diego, CA)
Assignee: VuSpex Inc.
G06T7/0002G06T2200/24G06T2207/10016
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 12,536,639
App. No.
19/043,962
Granted
Jan 27, 2026
Kind
B1
Abstract

A computer-implemented system and method for automating building inspections are disclosed herein. This system and method expedites interpretation and inspection against documented requirements, improves accuracy in measurement and other calculations, as well as securely stores documentation against said requirements. The system includes a device with a user interface, an extraction engine running a server, a classifier engine running on the server, a database, a large language model running on the server, and a tool library on the device.

Claims (52)

1 . A computer-implemented method comprising:

importing, at a server, a file comprising a plurality of requirements for a building inspection at an inspection site, the plurality of requirements comprising requirements from disparate systems in disparate formats;

extracting, at an extraction engine of the server, at least one requirement from the file;

comparing, at a classifier engine of the server, the plurality of requirements, against a database;

storing the plurality of requirements and metadata in a database;

determining, at the server, a procedure for generating a task list to complete an inspection for the plurality of requirements;

transmitting the procedure and the plurality of requirements from the server to a mobile communication device located at the inspection site;

calling a plurality of tool functions from a tool library on the mobile communication device, wherein the plurality of tool functions comprises at least one of a 3D LiDAR Tool, a video tool, or an image tool;

displaying on a user-interface of the mobile communication device, the plurality of tool functions;

performing, on the mobile communication device, a task of the plurality of tasks on the task list using a tool function of the plurality of tool functions;

marking, on the device, the task as finished on the task list;

accepting, at the server, an image or video of the finished task;

using the camera function of the device to obtain the image or video of the finished task;

confirming, at the server, that the task was completed using the image or video of the finished task; and

outputting a result, at the server, about at least one requirement.

2 . The method according to claim 1 wherein a format of the file is Word Doc, PDF, web URL, JPG, PNG, GIF.

3 . The method of claim 1 in which the extraction engine is a large language model.

4 . The method of claim 1 in which the database is a vector database.

5 . The method of claim 1 in which the result output is a pass, fail, or partial pass.

6 . The method of claim 1 in which the result output is an error or request to complete another task.

7 . The method of claim 1 in which the result output is a description of a requirement and the manner in which the requirement was not met.

8 . The method of claim 1 in which the result output is a description of a requirement and the manner in which the requirement was met.

9 . The method to claim 1 wherein the task list is aligned to a local or regional standard.

10 . The method according to claim 1 wherein the task list is aligned to an International Building Code 1011.5 from the International Code Council.

11 . A computer implemented system comprising:

a mobile communication device with a user interface;

an extraction engine running a server;

a classifier engine running on the server;

a database;

a large language model running on the server; and

a tool library on the mobile communication device;

wherein the server is configured to import a file comprising a plurality of requirements for a building inspection at an inspection site, the plurality of requirements comprising requirements from disparate systems in disparate formats;

wherein the extraction engine is configured to extract the plurality of requirements from the file;

wherein the server is configured to transmit the plurality of requirements to the classifier engine;

wherein the classifier engine is configured to compare the plurality of requirements against data in a database;

wherein the server is configured to store the plurality of requirements and metadata in the database;

wherein the server is configured to determine a procedure for generating a task list to complete an inspection for the plurality of requirements;

wherein the server is configured to transmit the procedure and the plurality of requirements to the mobile communication device located at the inspection site;

wherein the mobile communication device is configured to display the task list and the plurality of requirements on the user interface;

wherein the mobile communication device is configured to call a plurality of tool functions from the tool library, wherein the plurality of tool function comprises at least one of a 3D LiDAR Tool, a video tool, or an image tool;

wherein the mobile communication device is configured to perform a task of the plurality of tasks on the task list using a tool function of the plurality of tool functions;

wherein the mobile communication device is configured to mark the task as finished on the task list;

wherein the server is configured to output a result for at least one requirement.

12 . The system according to claim 11 wherein a format of the file is Word Doc, PDF, web URL, JPG, PNG, GIF.

13 . The system of claim 11 in which the extraction engine is a large language model.

14 . The system of claim 11 in which the database is a vector database.

15 . The system of claim 11 in which the result output is a pass, fail, or partial pass.

16 . The system of claim 11 in which the result output is an error or request to complete another task.

17 . The system of claim 11 in which the result output is a description of a requirement and the manner in which the requirement was not met.

18 . The system of claim 11 in which the result output is a description of a requirement and the manner in which the requirement was met.

19 . The system to claim 11 wherein the task list is aligned to a local or regional standard.

20 . The system according to claim 11 wherein the task list is aligned to an International Building Code 1011.5 from the International Code Council.

References Cited (25)
US 7639943B1 · Kalajan · 2009 [cited by applicant]
US 8385950B1 · Wagner · 2013 [cited by applicant]
US 8418231B2 · Denny · 2013 [cited by applicant]
US 8665337B2 · Mikawa · 2014 [cited by applicant]
US 8818018B2 · Simske · 2014 [cited by applicant]
US 10402064B1 · Al-Sallami · 2019 [cited by applicant]
US 10796374B1 · Cuttell · 2020 [cited by applicant]
US 10878388B2 · Wang et al. · 2020 [cited by applicant]
US 11163941B1 · Al Majid · 2021 [cited by applicant]
US 11190679B2 · Ebsen · 2021 [cited by applicant]
US 11228551B1 · Baron · 2022 [cited by applicant]
US 11361423B2 · Toth et al. · 2022 [cited by applicant]
US 11963072B2 · Kalajan · 2024 [cited by applicant]
US 20160267270A1 · Lee · 2016 [cited by applicant]
US 20170332421A1 · Sternberg · 2017 [cited by applicant]
US 20220197620A1 · Vihar · 2022 [cited by applicant]
US 20230247393A1 · Kalajan · 2023 [cited by applicant]
US 20230269105A1 · McGregor et al. · 2023 [cited by applicant]
US 20230308499A1 · Kalajan · 2023 [cited by applicant]
US 20240193754A1 · Karaaslan et al. · 2024 [cited by applicant]
CN 114723077B · 2024 [cited by examiner]
Chen, Hongrui, et al. “InspectionGPT: A Large Language Model-Based System for Inspection Task Planning.” 2024 IEEE 13th Data Driven Control and Learning Systems Conference (DDCLS). IEEE, 2024. (Year: 2024). [cited by examiner]
Svennberg, Kaisa, and Jan Ekman. “Structuring Semi-structured Data from Building Inspection Reports Using a Large Language Model.” International Association of Building Physics. Singapore: Springer Nature Singapore, 202… [cited by examiner]
Zhu, Xueying, et al. “LIm-Based Regulation Query System for Textual and Tabular Information in Building Codes.” Available at SSRN 5033360 (2024). (Year: 2024). [cited by examiner]
Ying, Huaquan, and Rafael Sacks. “From Automatic to Autonomous: A Large Language Model-driven Approach for Generic Building Compliance Checking.” Proceedings of the CIB W78 Conference. 2024. (Year: 2024). [cited by examiner]