Intelligent inspection method and system for server rooms based on AR and artificial intelligence

Through the intelligent inspection method of server computer room based on AR and artificial intelligence, the inspection path is determined using three-dimensional models and neural network analysis, and the intelligent robot is automatically controlled to conduct inspections, solving the problem of low efficiency of traditional manual inspections and achieving efficient failures Detection and data analysis.

CN118314640BActive Publication Date: 2024-11-19THREE GORGES HI TECH INFORMATION TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410447576.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-04-15
Publication Date
2024-11-19
Estimated Expiration
2044-04-15

AI Technical Summary

Technical Problem

The fault warning, inspection, inspection and data analysis of existing server rooms rely on traditional manual or semi-manual operations, and are inefficient and cannot meet the needs of the era of 5G networks, Internet of Things, AR, big data and artificial intelligence.

Method used

Using the intelligent inspection method of server computer room based on AR and artificial intelligence, by obtaining three-dimensional model diagrams and image data, using neural network model analysis to determine the inspection points and paths, automatically control the intelligent robot for inspection, and synchronize the images to AR in real time Terminal, realizes patrol without on-site control.

Benefits of technology

It improves the efficiency of inspection, realizes automated fault detection and data analysis, reduces manual intervention, and improves the accuracy and efficiency of inspection.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118314640B_ABST
    Figure CN118314640B_ABST
Patent Text Reader

Abstract

The present invention provides a method and system for intelligent inspection of a server room based on AR and artificial intelligence, wherein the method comprises: obtaining a three-dimensional model diagram of the interior of the server room; analyzing the three-dimensional model diagram to determine a first inspection point; obtaining image data of the interior of the server room; analyzing the image data based on a preset neural network model to determine a second inspection point; determining an inspection path based on the first inspection point and the second inspection point; controlling an intelligent robot to inspect according to the inspection path, and obtaining images taken by the intelligent robot in real time and synchronizing them to an AR terminal. The present invention comprehensively analyzes the actual situation in the server room and the image data of the monitoring in the server room, and then determines the inspection path, automatically controls the intelligent robot to inspect, and synchronizes it to the AR terminal, so that the staff does not need to perform on-site control and on-site inspection, thereby improving the inspection efficiency.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of intelligent inspection technology, and in particular to an intelligent inspection method and system for a server room based on AR and artificial intelligence. Background Art

[0002] With the advent of the 5G network, Internet of Things, AR, big data, and artificial intelligence era, each technology is based on the server room as hardware support; the existing server room's related fault warning, safety production hazard inspection, investigation, data analysis, etc., traditional manual or semi-manual operations can no longer meet the requirements. Summary of the invention

[0003] One of the purposes of the present invention is to provide an intelligent inspection method and system for a server room based on AR and artificial intelligence, which comprehensively analyzes the actual situation in the server room and the image data of the monitoring in the server room, and then determines the inspection path, automatically controls the intelligent robot to perform inspections, and synchronizes it to the AR terminal, so that staff do not need on-site control and on-site inspections, thereby improving the inspection efficiency.

[0004] An embodiment of the present invention provides an intelligent inspection method for a server room based on AR and artificial intelligence, comprising:

[0005] Obtain a 3D model of the server room interior;

[0006] Analyze the three-dimensional model diagram to determine the first inspection point;

[0007] Acquire image data of the interior of the server room;

[0008] Analyzing the image data based on a preset neural network model to determine a second inspection point;

[0009] Determine an inspection path based on the first inspection point and the second inspection point;

[0010] The intelligent robot is controlled to perform inspection according to the inspection path, and the images taken by the intelligent robot are acquired in real time and synchronized to the AR terminal.

[0011] Preferably, analyzing the three-dimensional model diagram to determine the first inspection point includes:

[0012] Determine the reference points corresponding to the entrances and exits of the server room in the 3D model diagram;

[0013] When the reference point is unique, a first path is constructed based on the traversable area of ​​the intelligent robot in the three-dimensional model diagram and the reference point; the first path is adopted at intervals to obtain the first inspection point;

[0014] When the reference point is not unique, the entry point and the exit point are determined from the reference point; a second path is constructed within the walkable area with the entry point as the starting point and the exit point as the end point; the second path is adopted at intervals to obtain the first inspection point.

[0015] Preferably, analyzing the image data based on a preset neural network model to determine the second inspection point includes:

[0016] Screening and grouping the image data to obtain multiple groups of images to be analyzed;

[0017] Inputting images in a group of images to be analyzed into a preset first model in chronological order to obtain personnel portrait data;

[0018] Determine if there are discrepancies in the profile data;

[0019] When there is a discrepancy, the standing position of the person who caused the discrepancy shall be used as the second inspection point;

[0020] When there is no difference, determine the standing position points of the persons corresponding to each image in the group; determine the standing time corresponding to each standing position point, and use the standing position point whose standing time is greater than or equal to the preset time threshold as the second inspection point.

[0021] Preferably, when determining the inspection path based on the first inspection point and the second inspection point, a set of sight lines of interest corresponding to each inspection point on the inspection path is also determined;

[0022] Among them, determining the sight lines corresponding to each inspection point on the inspection path includes:

[0023] When the inspection point corresponds to the first inspection point, the three-dimensional model corresponding to the preset intelligent robot is placed in the three-dimensional model map and based on the preset first area range threshold, the line of sight area corresponding to the inspection point and the target points corresponding to each line of sight direction in the line of sight area are determined; according to the preset point and hidden degree correspondence table, the hidden degree corresponding to each point is determined; the line of sight directions with a hidden degree greater than the preset hidden threshold are extracted as objects to be screened; according to the preset extraction rules, the line of sight directions are extracted from the objects to be screened to construct the line of sight set.

[0024] Preferably, determining the sight lines of interest corresponding to each inspection point on the inspection path further includes:

[0025] When the inspection point corresponds to the second inspection point, the three-dimensional model corresponding to the preset intelligent robot is placed in the three-dimensional model map and based on the preset second area range threshold, the line of sight area corresponding to the inspection point and the target points corresponding to each line of sight direction in the line of sight area are determined; the concealment degree corresponding to each point is determined according to the preset point and concealment degree correspondence table; according to the personnel behavior, the correlation comparison table is extracted; according to the correlation comparison table, the correlation degree of each point is determined; based on the correlation degree and the concealment degree, the priority corresponding to each point is determined; the concerned lines of sight corresponding to a preset number of points with the largest priority are extracted to construct the concerned line of sight set.

[0026] Preferably, the intelligent inspection method for server rooms based on AR and artificial intelligence also includes:

[0027] Obtain monitoring data from temperature and humidity monitoring devices installed at various preset locations in the server room;

[0028] Based on the monitoring data and the three-dimensional model diagram, a temperature and humidity AR monitoring diagram is generated and output to the AR terminal.

[0029] The present invention also provides an intelligent inspection system for server rooms based on AR and artificial intelligence, comprising:

[0030] A first acquisition module is used to acquire a three-dimensional model image of the interior of the server room;

[0031] A first determination module is used to analyze the three-dimensional model diagram to determine a first inspection point;

[0032] A second acquisition module is used to acquire image data of the interior of the server room;

[0033] A second determination module, used to analyze the image data based on a preset neural network model to determine a second inspection point;

[0034] A path determination module, used to determine an inspection path based on the first inspection point and the second inspection point;

[0035] The inspection and synchronization module is used to control the intelligent robot to perform inspections according to the inspection path, and to obtain images taken by the intelligent robot in real time and synchronize them to the AR terminal.

[0036] Preferably, the first determination module analyzes the three-dimensional model diagram, determines the first inspection point, and performs the following operations:

[0037] Determine the reference points corresponding to the entrances and exits of the server room in the 3D model diagram;

[0038] When the reference point is unique, a first path is constructed based on the traversable area of ​​the intelligent robot in the three-dimensional model diagram and the reference point; the first path is adopted at intervals to obtain the first inspection point;

[0039] When the reference point is not unique, the entry point and the exit point are determined from the reference point; a second path is constructed within the walkable area with the entry point as the starting point and the exit point as the end point; the second path is adopted at intervals to obtain the first inspection point.

[0040] Preferably, the second determination module analyzes the image data based on a preset neural network model, determines the second inspection point, and performs the following operations:

[0041] Screening and grouping the image data to obtain multiple groups of images to be analyzed;

[0042] Inputting images in a group of images to be analyzed into a preset first model in chronological order to obtain personnel portrait data;

[0043] Determine if there are discrepancies in the profile data;

[0044] When there is a discrepancy, the standing position of the person who caused the discrepancy shall be used as the second inspection point;

[0045] When there is no difference, determine the standing position points of the persons corresponding to each image in the group; determine the standing time corresponding to each standing position point, and use the standing position point whose standing time is greater than or equal to the preset time threshold as the second inspection point.

[0046] Preferably, the path determination module determines the inspection path based on the first inspection point and the second inspection point, and also determines the sight line set corresponding to each inspection point on the inspection path;

[0047] Among them, the path determination module determines the focus sight set corresponding to each inspection point on the inspection path, and performs the following operations:

[0048] When the inspection point corresponds to the first inspection point, the three-dimensional model corresponding to the preset intelligent robot is placed in the three-dimensional model map and based on the preset first area range threshold, the line of sight area corresponding to the inspection point and the target points corresponding to each line of sight direction in the line of sight area are determined; according to the preset point and hidden degree correspondence table, the hidden degree corresponding to each point is determined; the line of sight directions with a hidden degree greater than the preset hidden threshold are extracted as objects to be screened; according to the preset extraction rules, the line of sight directions are extracted from the objects to be screened to construct the line of sight set.

[0049] Preferably, the path determination module determines the set of sight lines of interest corresponding to each inspection point on the inspection path, and further performs the following operations:

[0050] When the inspection point corresponds to the second inspection point, the three-dimensional model corresponding to the preset intelligent robot is placed in the three-dimensional model map and based on the preset second area range threshold, the line of sight area corresponding to the inspection point and the target points corresponding to each line of sight direction in the line of sight area are determined; the concealment degree corresponding to each point is determined according to the preset point and concealment degree correspondence table; according to the personnel behavior, the correlation comparison table is extracted; according to the correlation comparison table, the correlation degree of each point is determined; based on the correlation degree and the concealment degree, the priority corresponding to each point is determined; the concerned lines of sight corresponding to a preset number of points with the largest priority are extracted to construct the concerned line of sight set.

[0051] Preferably, the server room intelligent inspection system based on AR and artificial intelligence also includes: a temperature and humidity monitoring module; the temperature and humidity monitoring module performs the following operations:

[0052] Obtain monitoring data from temperature and humidity monitoring devices installed at various preset locations in the server room;

[0053] Based on the monitoring data and the three-dimensional model diagram, a temperature and humidity AR monitoring diagram is generated and output to the AR terminal.

[0054] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0055] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0057] Figure 1 Schematic diagram of an intelligent inspection method for a server room based on AR and artificial intelligence in an embodiment of the present invention;

[0058] Figure 2 Schematic diagram of an intelligent inspection system for a server room based on AR and artificial intelligence in an embodiment of the present invention. DETAILED DESCRIPTION

[0059] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0060] The embodiment of the present invention provides an intelligent inspection method for a server room based on AR and artificial intelligence, such as Figure 1 As shown, including:

[0061] Step S1: Obtain a three-dimensional model of the interior of the server room;

[0062] Step S2: Analyze the three-dimensional model diagram to determine the first inspection point;

[0063] Step S3: Acquire image data of the interior of the server room;

[0064] Step S4: Analyze the image data based on a preset neural network model to determine the second inspection point;

[0065] Step S5: determining an inspection path based on the first inspection point and the second inspection point;

[0066] Step S6: Control the intelligent robot to perform inspection according to the inspection path, and obtain the images taken by the intelligent robot in real time and synchronize them to the AR terminal.

[0067] The three-dimensional model diagram is analyzed to determine the first inspection point, including:

[0068] Determine the reference points corresponding to the entrances and exits of the server room in the 3D model diagram;

[0069] When the reference point is unique, a first path is constructed based on the traversable area of ​​the intelligent robot in the three-dimensional model diagram and the reference point; the first path is adopted at intervals to obtain the first inspection point;

[0070] When the reference point is not unique, the entry point and the exit point are determined from the reference point; a second path is constructed within the walkable area with the entry point as the starting point and the exit point as the end point; the second path is adopted at intervals to obtain the first inspection point.

[0071] The method of analyzing the image data based on a preset neural network model to determine the second inspection point includes:

[0072] Screening and grouping the image data to obtain multiple groups of images to be analyzed;

[0073] Inputting images in a group of images to be analyzed into a preset first model in chronological order to obtain personnel portrait data;

[0074] Determine if there are discrepancies in the profile data;

[0075] When there is a discrepancy, the standing position of the person who caused the discrepancy shall be used as the second inspection point;

[0076] When there is no difference, determine the standing position points of the persons corresponding to each image in the group; determine the standing time corresponding to each standing position point, and use the standing position point whose standing time is greater than or equal to the preset time threshold as the second inspection point.

[0077] The working principle and beneficial effects of the above technical solution are as follows:

[0078] The intelligent inspection method of the server room based on AR and artificial intelligence of the present invention comprehensively analyzes the actual situation in the server room and the image data of the monitoring in the server room, and then determines the inspection path, automatically controls the intelligent robot to conduct inspection, and synchronizes it to the AR terminal, so that the staff does not need on-site control and on-site inspection, thereby improving the inspection efficiency. Among them, the analysis of the image data focuses on the analysis of the behavior and path of the personnel entering the room, and a neural network is used in the analysis to realize the intelligent analysis of artificial intelligence. The AR terminal can be configured as AR glasses; the present invention adopts an intelligent robot and an intelligent recognition technology monitoring device. Two control modes of the robot are provided: manual control and automatic control, and the two control modes can be switched arbitrarily. The manual control function can realize the control operation of the robot. During automatic control, the system can realize the detection function of the equipment in a fully autonomous mode according to the pre-set tasks or the tasks temporarily specified by the user through the cooperation of the various functional units of the robot.

[0079] In order to ensure the progress of the inspection, in one embodiment, when determining the inspection path based on the first inspection point and the second inspection point, a set of sight lines of interest corresponding to each inspection point on the inspection path is also determined;

[0080] Among them, determining the sight lines corresponding to each inspection point on the inspection path includes:

[0081] When the inspection point corresponds to the first inspection point, the three-dimensional model corresponding to the preset intelligent robot is placed in the three-dimensional model map and based on the preset first area range threshold, the line of sight area corresponding to the inspection point and the target points corresponding to each line of sight direction in the line of sight area are determined; according to the preset point and hidden degree correspondence table, the hidden degree corresponding to each point is determined; the line of sight directions with a hidden degree greater than the preset hidden threshold are extracted as objects to be screened; according to the preset extraction rules, the line of sight directions are extracted from the objects to be screened to construct the line of sight set.

[0082] In addition, the focus sight set corresponding to each inspection point on the inspection path is determined, and also includes:

[0083] When the inspection point corresponds to the second inspection point, the three-dimensional model corresponding to the preset intelligent robot is placed in the three-dimensional model map and based on the preset second area range threshold, the line of sight area corresponding to the inspection point and the target points corresponding to each line of sight direction in the line of sight area are determined; the concealment degree corresponding to each point is determined according to the preset point and concealment degree correspondence table; according to the personnel behavior, the correlation comparison table is extracted; according to the correlation comparison table, the correlation degree of each point is determined; based on the correlation degree and the concealment degree, the priority corresponding to each point is determined; the concerned lines of sight corresponding to a preset number of points with the largest priority are extracted to construct the concerned line of sight set.

[0084] In order to realize the temperature and humidity monitoring of the computer room, in one embodiment, the intelligent inspection method of the server room based on AR and artificial intelligence also includes:

[0085] Obtain monitoring data from temperature and humidity monitoring devices installed at various preset locations in the server room;

[0086] Based on the monitoring data and the three-dimensional model diagram, a temperature and humidity AR monitoring diagram is generated and output to the AR terminal.

[0087] The present invention also provides an intelligent inspection system for server rooms based on AR and artificial intelligence, such as Figure 2 As shown, including:

[0088] The first acquisition module 1 is used to acquire a three-dimensional model of the interior of the server room;

[0089] A first determination module 2 is used to analyze the three-dimensional model diagram to determine a first inspection point;

[0090] The second acquisition module 3 is used to acquire image data of the interior of the server room;

[0091] A second determination module 4 is used to analyze the image data based on a preset neural network model to determine a second inspection point;

[0092] A path determination module 5, configured to determine an inspection path based on the first inspection point and the second inspection point;

[0093] The inspection and synchronization module 6 is used to control the intelligent robot to perform inspection according to the inspection path, and to obtain the images taken by the intelligent robot in real time and synchronize them to the AR terminal.

[0094] The first determination module 2 analyzes the three-dimensional model diagram, determines the first inspection point, and performs the following operations:

[0095] Determine the reference points corresponding to the entrances and exits of the server room in the 3D model diagram;

[0096] When the reference point is unique, a first path is constructed based on the traversable area of ​​the intelligent robot in the three-dimensional model diagram and the reference point; the first path is adopted at intervals to obtain the first inspection point;

[0097] When the reference point is not unique, the entry point and the exit point are determined from the reference point; a second path is constructed within the walkable area with the entry point as the starting point and the exit point as the end point; the second path is adopted at intervals to obtain the first inspection point.

[0098] The second determination module 4 analyzes the image data based on a preset neural network model, determines the second inspection point, and performs the following operations:

[0099] Screening and grouping the image data to obtain multiple groups of images to be analyzed;

[0100] Inputting images in a group of images to be analyzed into a preset first model in chronological order to obtain personnel portrait data;

[0101] Determine if there are discrepancies in the profile data;

[0102] When there is a discrepancy, the standing position of the person who caused the discrepancy shall be used as the second inspection point;

[0103] When there is no difference, determine the standing position points of the persons corresponding to each image in the group; determine the standing time corresponding to each standing position point, and use the standing position point whose standing time is greater than or equal to the preset time threshold as the second inspection point.

[0104] The working principle and intended effects of the above technical solution are:

[0105] In addition, the system also includes: power distribution monitoring subsystem, network monitoring subsystem, video monitoring subsystem and safety detection subsystem.

[0106] Among them, the power distribution monitoring subsystem: real-time acquisition and monitoring of the switch quantity, analog quantity, protection action signal, etc. of various devices such as the power distribution room and the power distribution line. Form a circuit network topology diagram for intuitive viewing, and use AR glasses visual interaction technology to control robots for inspection.

[0107] Network monitoring subsystem: Automatically discover network terminal devices, all-round IP management, access permission management, and comprehensive monitoring of key network devices such as routers, switches, and firewalls to obtain the in-depth visibility required to manage complex networks.

[0108] Video surveillance subsystem: It has the function of detecting object movement, collecting and storing visible light and infrared videos through the visual interaction system based on AR glasses, and supports functions such as video playback, stopping, capturing pictures, recording, and full-screen display; abnormal situation data is sent to the data center in real time.

[0109] Security detection subsystem: Automatically or manually scan network or terminal application systems, various third-party applications, database vulnerabilities, and storage file compliance, generate reports, and automatically repair vulnerabilities.

[0110] In one embodiment, the path determination module determines the inspection path based on the first inspection point and the second inspection point, and also determines the set of sight lines of interest corresponding to each inspection point on the inspection path;

[0111] Among them, the path determination module determines the focus sight set corresponding to each inspection point on the inspection path, and performs the following operations:

[0112] When the inspection point corresponds to the first inspection point, the three-dimensional model corresponding to the preset intelligent robot is placed in the three-dimensional model map and based on the preset first area range threshold, the line of sight area corresponding to the inspection point and the target points corresponding to each line of sight direction in the line of sight area are determined; according to the preset point and hidden degree correspondence table, the hidden degree corresponding to each point is determined; the line of sight directions with a hidden degree greater than the preset hidden threshold are extracted as objects to be screened; according to the preset extraction rules, the line of sight directions are extracted from the objects to be screened to construct the line of sight set.

[0113] In addition, the path determination module determines the sight lines corresponding to each inspection point on the inspection path, and also performs the following operations:

[0114] When the inspection point corresponds to the second inspection point, the three-dimensional model corresponding to the preset intelligent robot is placed in the three-dimensional model map and based on the preset second area range threshold, the line of sight area corresponding to the inspection point and the target points corresponding to each line of sight direction in the line of sight area are determined; the concealment degree corresponding to each point is determined according to the preset point and concealment degree correspondence table; according to the personnel behavior, the correlation comparison table is extracted; according to the correlation comparison table, the correlation degree of each point is determined; based on the correlation degree and the concealment degree, the priority corresponding to each point is determined; the concerned lines of sight corresponding to a preset number of points with the largest priority are extracted to construct the concerned line of sight set.

[0115] In one embodiment, the server room intelligent inspection system based on AR and artificial intelligence further includes: a temperature and humidity monitoring module; the temperature and humidity monitoring module performs the following operations:

[0116] Obtain monitoring data from temperature and humidity monitoring devices installed at various preset locations in the server room;

[0117] Based on the monitoring data and the three-dimensional model diagram, a temperature and humidity AR monitoring diagram is generated and output to the AR terminal

[0118] The working principle and beneficial effects of the above technical solution are:

[0119] The temperature and humidity monitoring module can generate thermal imaging distribution maps and humidity distribution maps, focus on monitoring and inspection of high temperature and high humidity areas; promptly initiate prevention and control measures for abnormal situations, and send data to the system data center in real time. The specific standards are: Class A computer room: summer room temperature standard is 22±2℃; winter room temperature standard is 20±2℃; air humidity is within the range of 40~55%. Class B computer room: room temperature requirements in all seasons are 22±2℃, and air humidity is 40~55%. Class C computer room: room temperature requirements are 10~28℃, and relative humidity requirements are 30~75%.

[0120] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An intelligent inspection method for server rooms based on AR and artificial intelligence. It is characterized in that include: Obtain a 3D model of the interior of the server room; Analyze the three-dimensional model diagram to determine the first inspection point; Acquire image data of the interior of the server room; Analyzing the image data based on a preset neural network model to determine a second inspection point; Determine an inspection path based on the first inspection point and the second inspection point; Controlling the intelligent robot to perform inspection according to the inspection path, and acquiring images taken by the intelligent robot in real time and synchronizing them to the AR terminal; When determining the inspection path based on the first inspection point and the second inspection point, a set of sight lines of interest corresponding to each inspection point on the inspection path is also determined; Among them, determining the sight lines corresponding to each inspection point on the inspection path includes: When the inspection point corresponds to the first inspection point, the three-dimensional model corresponding to the preset intelligent robot is placed in the three-dimensional model map and based on the preset first area range threshold, the line of sight area corresponding to the inspection point and the target points corresponding to each line of sight direction in the line of sight area are determined; according to the preset point and hidden degree correspondence table, the hidden degree corresponding to each point is determined; the line of sight directions with a hidden degree greater than the preset hidden threshold are extracted as objects to be screened; according to the preset extraction rules, the line of sight directions are extracted from the objects to be screened to construct the line of sight set.

2. The server room intelligent inspection method based on AR and artificial intelligence as claimed in claim 1, It is characterized in that Analyzing the three-dimensional model diagram to determine the first inspection point includes: Determine the reference points corresponding to the entrances and exits of the server room in the 3D model diagram; When the reference point is unique, a first path is constructed based on the traversable area of ​​the intelligent robot in the three-dimensional model diagram and the reference point; the first path is adopted at intervals to obtain the first inspection point; When the reference point is not unique, the entry point and the exit point are determined from the reference point; a second path is constructed within the walkable area with the entry point as the starting point and the exit point as the end point; the second path is adopted at intervals to obtain the first inspection point.

3. The server room intelligent inspection method based on AR and artificial intelligence as claimed in claim 1, It is characterized in that Analyzing the image data based on a preset neural network model to determine a second inspection point includes: Screening and grouping the image data to obtain multiple groups of images to be analyzed; Inputting images in a group of images to be analyzed into a preset first model in chronological order to obtain personnel portrait data; Determine if there are discrepancies in the profile data; When there is a discrepancy, the standing position of the person who caused the discrepancy shall be used as the second inspection point; When there is no difference, determine the standing position points of the persons corresponding to each image in the group; determine the standing time corresponding to each standing position point, and use the standing position point whose standing time is greater than or equal to the preset time threshold as the second inspection point.

4. The server room intelligent inspection method based on AR and artificial intelligence as claimed in claim 1, It is characterized in that Determine the sight lines corresponding to each inspection point on the inspection path, including: When the inspection point corresponds to the second inspection point, the three-dimensional model corresponding to the preset intelligent robot is placed in the three-dimensional model map and based on the preset second area range threshold, the line of sight area corresponding to the inspection point and the target points corresponding to each line of sight direction in the line of sight area are determined; the concealment degree corresponding to each point is determined according to the preset point and concealment degree correspondence table; according to the personnel behavior, the correlation degree comparison table is extracted; according to the correlation degree comparison table, the correlation degree of each point is determined; based on the correlation degree and the concealment degree, the priority corresponding to each point is determined; the concerned lines of sight corresponding to a preset number of points with the largest priority are extracted to construct the concerned line of sight set.

5. The server room intelligent inspection method based on AR and artificial intelligence as claimed in claim 1, It is characterized in that Also includes: Obtain monitoring data from temperature and humidity monitoring devices installed at various preset locations in the server room; Based on the monitoring data and the three-dimensional model diagram, a temperature and humidity AR monitoring diagram is generated and output to the AR terminal.

6. An intelligent inspection system for server rooms based on AR and artificial intelligence, It is characterized in that include: A first acquisition module is used to acquire a three-dimensional model image of the interior of the server room; A first determination module is used to analyze the three-dimensional model diagram to determine a first inspection point; A second acquisition module is used to acquire image data of the interior of the server room; A second determination module, used to analyze the image data based on a preset neural network model to determine a second inspection point; A path determination module, used to determine an inspection path based on the first inspection point and the second inspection point; The inspection and synchronization module is used to control the intelligent robot to perform inspections according to the inspection path, and to obtain images taken by the intelligent robot in real time and synchronize them to the AR terminal; Wherein, the path determination module determines the inspection path based on the first inspection point and the second inspection point, and also determines the sight line set corresponding to each inspection point on the inspection path; Among them, the path determination module determines the focus sight set corresponding to each inspection point on the inspection path, and performs the following operations: When the inspection point corresponds to the first inspection point, the three-dimensional model corresponding to the preset intelligent robot is placed in the three-dimensional model map and based on the preset first area range threshold, the line of sight area corresponding to the inspection point and the target points corresponding to each line of sight direction in the line of sight area are determined; according to the preset point and hidden degree correspondence table, the hidden degree corresponding to each point is determined; the line of sight directions with a hidden degree greater than the preset hidden threshold are extracted as objects to be screened; according to the preset extraction rules, the line of sight directions are extracted from the objects to be screened to construct the line of sight set.

7. The server room intelligent inspection system based on AR and artificial intelligence as claimed in claim 6, It is characterized in that The first determination module analyzes the three-dimensional model diagram, determines the first inspection point, and performs the following operations: Determine the reference points corresponding to the entrances and exits of the server room in the 3D model diagram; When the reference point is unique, constructing a first path based on the intelligent robot's walkable area and the reference point in the three-dimensional model map; The first inspection point is obtained by performing interval sampling on the first path; When the reference point is not unique, determining the entry point and the exit point from the reference point; constructing a second path within the walkable area with the entry point as the starting point and the exit point as the end point; The second path is sampled at intervals to obtain the first inspection points.

8. The server room intelligent inspection system based on AR and artificial intelligence as claimed in claim 6, It is characterized in that The second determination module analyzes the image data based on a preset neural network model, determines the second inspection point, and performs the following operations: Screening and grouping the image data to obtain multiple groups of images to be analyzed; Inputting images in a group of images to be analyzed into a preset first model in chronological order to obtain personnel portrait data; Determine if there are discrepancies in the profile data; When there is a discrepancy, the standing position of the person who caused the discrepancy shall be used as the second inspection point; When there is no difference, the standing position points of the persons corresponding to each image in the group are determined; The standing time corresponding to each standing position point is determined, and the standing position point whose standing time is greater than or equal to a preset time threshold is used as the second inspection point.