Repositioning method, device, medium and equipment for patrol robot

By constructing and utilizing the image information collected by the gimbal camera of the patrol robot, combined with the neural network model, the autonomous relocation and autonomous initialization of the patrol robot are achieved, solving the problems of low intelligence and high equipment costs in the existing technology, and improving the Position accuracy and robustness.

CN115185278BActive Publication Date: 2024-11-08GUANGZHOU GOSUNCN ROBOTICS CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210830810.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-07-15
Publication Date
2024-11-08
Estimated Expiration
2042-07-15

AI Technical Summary

Technical Problem

The existing patrol robots have low intelligence, high equipment costs and poor positioning accuracy during the autonomous relocation process.

Method used

By constructing image information and position information of repositioning area points, using the gimbal camera to collect image information, and calculating similarity through neural network models, the autonomous repositioning and autonomous initialization of patrol robots can be realized, equipment costs are reduced, and positioning accuracy is improved. .

Benefits of technology

It improves the intelligence of patrol robots, reduces equipment costs, enhances positioning accuracy and robustness, and avoids the high equipment costs and signal interference problems of GPS positioning and UWB methods.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115185278B_ABST
    Figure CN115185278B_ABST
Patent Text Reader

Abstract

The present invention discloses a repositioning method for a patrol robot, comprising: constructing positioning information corresponding to a repositioning area point, the positioning information comprising image information and posture information of the repositioning area point; when the patrol robot loses positioning, acquiring the posture information of the area point to be positioned of the patrol robot according to the image information of the area point to be positioned of the patrol robot and the image information corresponding to the repositioning area point, thereby realizing the functions of autonomous repositioning and autonomous initialization of the patrol robot during positioning timing, improving the intelligence of the patrol robot, having the advantage of low equipment cost compared to GPS positioning or UWB positioning methods, and improving positioning accuracy and robustness.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of robot technology, and in particular to a repositioning method, device, medium and equipment for a patrol robot. Background Art

[0002] In recent years, the positioning technology of autonomous mobile robots has become one of the hot research technologies. Autonomous mobile robots refer to robots that have the ability to find their own paths and walk. To achieve this ability, the robot must know its location. The existing positioning methods widely used in autonomous mobile robots include laser positioning, GPS positioning, and visual positioning.

[0003] Patrol robots are a specific application of automatic mobile robots, mainly used in industrial parks, parks, buildings and other places to perform patrol, photo taking, video recording and other tasks. The existing patrol robot positioning technology is mainly based on laser positioning, which has the advantages of high positioning accuracy, good positioning stability and mature algorithms. However, it is necessary to manually set the initial position value before the patrol robot is started, which has high requirements for users and low intelligence. If the patrol robot loses its position during operation, it will be difficult to reposition itself, causing the patrol work to stagnate, greatly affecting the user experience. Existing technologies also assist patrol robots in achieving autonomous repositioning by carrying high-precision GPS positioning equipment or arranging UWB positioning systems in the scene. However, the equipment cost of these two methods is high, and the signal is easily affected by external objects, such as trees and building walls, and the accuracy of repositioning is poor. Summary of the invention

[0004] The embodiments of the present invention provide a patrol robot repositioning method, device, medium and equipment to solve the problems of low intelligence, high equipment cost and poor positioning accuracy in the autonomous repositioning process of the existing patrol robot.

[0005] A patrol robot repositioning method, the method comprising:

[0006] Constructing positioning information corresponding to the repositioning area point, the positioning information including image information and posture information of the repositioning area point;

[0007] When the patrol robot loses its positioning, the position and posture information of the point in the area to be positioned of the patrol robot is obtained according to the image information of the point in the area to be positioned of the patrol robot and the image information corresponding to the point in the repositioning area.

[0008] Optionally, the positioning information corresponding to the constructed relocation area point includes:

[0009] Select the relocation area points of the patrol robot according to the preset interval distance;

[0010] Start the patrol robot and initialize the patrol robot's position information in the map, and control the patrol robot to walk along the patrol route;

[0011] Whenever the patrol robot walks to a point in the relocation area, the pan-tilt camera is controlled to capture image information of the point in the relocation area at every preset rotation angle;

[0012] The image information and the posture information of the relocation area point are associated to obtain the positioning information corresponding to the relocation area point.

[0013] Optionally, acquiring the position and posture information of the point in the area to be located of the patrol robot according to the image information of the point in the area to be located of the patrol robot and the image information corresponding to the point in the relocation area comprises:

[0014] Obtain the image information of the current area point collected by the pan-tilt camera on the patrol robot as the image information of the area point to be located;

[0015] Using a preset neural network model, the similarity between the image information corresponding to the relocated area point and the image information of the area point to be located is obtained;

[0016] When there is at least one piece of image information in the image information corresponding to the relocated area point and the similarity thereof is greater than or equal to a preset similarity threshold, the position and posture information of the relocated area point is used as the position and posture information of the area point to be located.

[0017] Optionally, the preset neural network model is a neural network implementation method of local aggregation descriptor vectors.

[0018] Optionally, before acquiring the position and posture information of the point in the area to be located of the patrol robot according to the image information of the point in the area to be located of the patrol robot and the image information corresponding to the point in the relocation area, the method further comprises:

[0019] Detect the charging status of the patrol robot;

[0020] If the patrol robot is in charging, the position and posture information of the charging location is obtained as the position and posture information of the area point to be positioned of the patrol robot.

[0021] Optionally, acquiring the position and posture information of the point in the area to be located of the patrol robot according to the image information of the point in the area to be located of the patrol robot and the image information corresponding to the point in the relocation area further comprises:

[0022] When the similarities of the image information corresponding to the relocation area points are all less than the preset similarity threshold, identifying the lane line according to the image information captured by the gimbal camera;

[0023] Controlling the patrol robot to move forward along the lane line, and comparing the travel distance of the patrol robot with a preset distance threshold;

[0024] If the travel distance is less than or equal to the preset distance threshold, acquiring the next area point to be located at a preset frequency;

[0025] According to the image information of the point in the area to be positioned and the image information corresponding to the point in the repositioning area, the position and posture information of the point in the area to be positioned is obtained.

[0026] Optionally, acquiring the position and posture information of the point in the area to be located of the patrol robot according to the image information of the point in the area to be located of the patrol robot and the image information corresponding to the point in the relocation area further comprises:

[0027] If the travel distance is greater than the preset distance threshold, a message indicating that the location is lost is reported to the server.

[0028] A repositioning device for a patrol robot, the device comprising:

[0029] A construction module, used to construct positioning information corresponding to the relocation area point, wherein the positioning information includes image information and posture information of the relocation area point;

[0030] The repositioning module is used to obtain the position and posture information of the patrol robot's area point to be positioned according to the image information of the patrol robot's area point to be positioned and the image information corresponding to the repositioning area point when the patrol robot loses its positioning.

[0031] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the patrol robot repositioning method described above is implemented.

[0032] A terminal device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned patrol robot repositioning method when executing the computer program.

[0033] The present invention introduces image information to realize the autonomous repositioning and autonomous initialization of the patrol robot, and constructs the positioning information corresponding to the repositioning area point, wherein the positioning information includes the image information and the posture information of the repositioning area point; when the patrol robot loses its positioning, the posture information of the area point to be positioned of the patrol robot is obtained according to the image information of the area point to be positioned of the patrol robot and the image information corresponding to the repositioning area point, thereby realizing the functions of autonomous repositioning and autonomous initialization of the patrol robot when positioning is lost, improving the intelligence of the patrol robot, and having the advantage of low equipment cost compared with GPS positioning or UWB positioning, and improving the positioning accuracy and robustness. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0035] Figure 1 is a flow chart of a patrol robot relocation method according to an embodiment of the present invention;

[0036] Figure 2 is a schematic diagram of a repositioning device for a patrol robot provided by an embodiment of the present invention;

[0037] Figure 3 is a schematic diagram of a computer device in one embodiment of the present invention. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] In an embodiment of the present invention, the patrol robot is a specific application of an automatic mobile robot, and is mainly used for patrolling, taking photos, and recording videos in industrial parks, parks, buildings, and other places. Compared with ordinary autonomous mobile robots, the patrol robot is equipped with an integrated pan-tilt camera device on the fuselage. The so-called integrated pan-tilt camera device, as the name implies, refers to the integration of a pan-tilt, a camera, a protective cover, and a decoder, wherein the pan-tilt can drive the camera to rotate left and right, up and down, and realize functions such as high-speed capture and recording, and is an essential equipment for patrol robots. The repositioning method of the patrol robot provided in an embodiment of the present invention combines the integrated pan-tilt camera device carried by the patrol robot itself, and introduces images to realize the autonomous repositioning and autonomous initialization functions of the patrol robot. The image information is rich, closer to the way humans find their own position, and improves the intelligence of the patrol robot. Compared with GPS positioning or UWB positioning methods, it has the advantages of low equipment cost and high global repositioning robustness.

[0040] The following is a detailed description of the patrol robot repositioning method provided by an embodiment of the present invention. Figure 1 The patrol robot repositioning method provided by the embodiment of the present invention. Figure 1 As shown, the repositioning method of the patrol robot includes:

[0041] In step S101, positioning information corresponding to the repositioning area point is constructed, and the positioning information includes image information and posture information of the repositioning area point.

[0042] Here, the present invention constructs a relocation area point in advance according to the patrol path of the patrol robot, and obtains the positioning information corresponding to the relocation area point. The relocation area point is a reference position for the patrol robot to perform autonomous positioning and autonomous initialization. Optionally, as a preferred embodiment of the present invention, the step S101 also includes:

[0043] In step S201, a relocation area point of the patrol robot is selected according to a preset interval.

[0044] Here, the embodiment of the present invention pre-sets the interval distance according to the demand, for example, greater than or equal to 100 meters, and then selects the relocation area point of the patrol robot according to the preset interval distance. In practical applications, the relocation area point should have rich features and be clearly distinguished from the non-relocation area point.

[0045] In step S202, the patrol robot is started and the position information of the patrol robot in the map is initialized, and the patrol robot is controlled to walk along the patrol route.

[0046] The relocation area point is entered into the patrol robot, the patrol robot is started and the position information of the patrol robot in the map is manually initialized, and the patrol robot is controlled to walk along the patrol path.

[0047] In step S203, every time the patrol robot walks to a point in the relocation area, the pan-tilt camera is controlled to capture image information of the point in the relocation area at every preset rotation angle.

[0048] Here, the preset rotation angle is 20°. Whenever the patrol robot walks to the relocation area, the pan / tilt camera on the patrol robot is controlled to take a surround shot of the relocation area, and an image is obtained every preset rotation angle, thereby obtaining multiple images corresponding to the relocation area. By taking multiple images, the image information used for subsequent relocation is enriched to meet the needs of the patrol robot to stop and relocate at different angles.

[0049] In step S204, the image information and the posture information of the relocation area point are associated to obtain the positioning information corresponding to the relocation area point.

[0050] The plurality of image information obtained by the patrol robot shooting around the relocation area point and the posture information obtained by the patrol robot at the relocation area point are associated and stored as the positioning information of the relocation area point. Optionally, the posture information includes the displacement information, heading angle, and horizontal tilt angle of the patrol robot.

[0051] In step S102, when the patrol robot loses its positioning, the position and posture information of the point in the area to be positioned of the patrol robot is obtained according to the image information of the point in the area to be positioned of the patrol robot and the image information corresponding to the point in the repositioning area.

[0052] When the patrol robot loses its position, the embodiment of the present invention collects image information of the patrol robot's current environment as image information of the area point to be positioned. The image information of the area point to be positioned is compared with a plurality of image information corresponding to each repositioning area point obtained in step S101, and the repositioning area point corresponding to the image information with the greatest image similarity is obtained, and the posture information of the repositioning area point obtained is used as the posture information of the area point to be positioned of the patrol robot. Optionally, step S102 also includes:

[0053] In step S301, image information of the current area point collected by the pan-tilt camera on the patrol robot is obtained as image information of the area point to be located.

[0054] Here, when the relocation process is started, the embodiment of the present invention uses the current environment as the area point to be located, and collects image information of the area point to be located through the pan-tilt camera on the patrol robot.

[0055] In step S302, a preset neural network model is used to obtain the similarity between the image information corresponding to the relocated area point and the image information of the area point to be located.

[0056] Through the preset neural network comparison, the similarity between each image information in the image information corresponding to the relocated area point and the image information of the area point to be located is calculated. Here, the similarity calculation of the image information is achieved by feature encoding a single image, and then comparing the feature encoding vectors between the two images to determine the similarity between the two images and complete the image similarity calculation. Optionally, in an embodiment of the present invention, the preset neural network model is a neural network implementation method NetVlad for local aggregation descriptor vectors.

[0057] In step S303, when there is at least one piece of image information in the image information corresponding to the relocated area point and the similarity is greater than or equal to a preset similarity threshold, the position and posture information of the relocated area point is used as the position and posture information of the area point to be located.

[0058] Here, the embodiment of the present invention pre-sets a similarity threshold, compares the similarity of each image information in the image information corresponding to the relocation area point with the similarity threshold, and determines whether the similarity is greater than or equal to the preset similarity threshold. If a relocation area point that meets the preset similarity threshold is found, the posture information corresponding to the relocation area point is used as the posture information of the patrol robot at the point in the area to be located, as the initial value for positioning initialization. The laser alignment and positioning module with a given initial value is called to complete the relocation, thereby effectively speeding up the alignment speed and alignment robustness.

[0059] Optionally, as a preferred example of the present invention, step S102 may further include:

[0060] In step S304, when the similarities of the image information corresponding to the relocation area points are all less than the preset similarity threshold, the lane line is identified according to the image information captured by the gimbal camera.

[0061] Here, when the similarities of the image information corresponding to the relocation area points are all less than the preset similarity threshold, that is, no relocation area points satisfying the preset similarity threshold are found, the image information captured by the gimbal camera is further subjected to lane line recognition to obtain the lane line.

[0062] In step S305, the patrol robot is controlled to move forward along the lane line, and the travel distance of the patrol robot is compared with a preset distance threshold.

[0063] In step S306, if the travel distance is less than or equal to the preset distance threshold, the next area point to be located is acquired at a preset frequency.

[0064] In step S307, the position and posture information of the point in the area to be positioned is obtained according to the image information of the point in the area to be positioned and the image information corresponding to the point in the repositioning area.

[0065] According to the lane line obtained in step S304, the patrol robot is controlled to move forward along the lane line, and steps S301 to S303 are repeated during the moving process. If there is a similarity greater than or equal to a preset similarity threshold, the posture information of the repositioning area point corresponding to the similarity is used as the posture information of the area point to be positioned, and the repositioning and position initialization are completed.

[0066] It should be understood that the patrol robot will repeat steps S301 to S303 to seek relocation and position initialization only when the travel distance of the patrol robot is within the preset distance threshold range. Among them, lane line recognition is achieved by image recognition, using the image recognition algorithm provided by the open source library opencv.

[0067] When the travel distance of the patrol robot exceeds the distance threshold and repositioning has not been achieved, step S102 may further include:

[0068] In step S308, if the travel distance is greater than the preset distance threshold, a message indicating that the location is lost is reported to the server.

[0069] Optionally, the preset distance threshold is preferably 120 meters. When the patrol robot travels a distance of more than 120 meters and has not yet achieved relocation, it actively reports a message of position loss to the background server to notify the staff, thereby greatly reducing the technical requirements of the staff and having strong practicality. After the staff obtains the position loss message, they can manually push the patrol robot to the nearest relocation area point to complete the relocation.

[0070] Optionally, since patrol robots are usually rechargeable robots, the embodiment of the present invention can also perform relocation and position initialization at the charging location. Before obtaining the position and posture information of the patrol robot's area to be located point according to the image information of the patrol robot's area to be located point and the image information corresponding to the relocation area point in step S102, the method further includes:

[0071] Detect the charging status of the patrol robot;

[0072] If the patrol robot is in charging, the position and posture information of the charging location is obtained as the position and posture information of the area point to be positioned of the patrol robot.

[0073] Here, the embodiment of the present invention first interacts with the battery management system to obtain the charging status sent by the battery management system. According to the charging status, if the patrol robot is being charged, the posture information of the charging location is directly given to the patrol robot to complete the relocation and position initialization. If the patrol robot is not being charged, step S102 is executed to relocate through image information.

[0074] In summary, the patrol robot repositioning method provided by the embodiment of the present invention directly uses the PTZ camera on the patrol robot, does not require any additional equipment cost, and is more economical and efficient than deploying UWB or using GPS positioning equipment. By introducing image processing, the patrol robot's autonomous repositioning and autonomous initialization functions are realized, the image information is rich, and it is closer to the way humans find their own position, which effectively improves the intelligence of the patrol robot, and the global repositioning has high robustness.

[0075] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0076] In one embodiment, the present invention further provides a patrol robot repositioning device, which corresponds one-to-one to the patrol robot repositioning method in the above embodiment. Figure 2 As shown, the repositioning device of the patrol robot includes a construction module 21 and a repositioning module 22. The functional modules are described in detail as follows:

[0077] A construction module 21 is used to construct positioning information corresponding to the relocation area point, wherein the positioning information includes image information and posture information of the relocation area point;

[0078] The repositioning module 22 is used to obtain the position and posture information of the point in the area to be positioned of the patrol robot according to the image information of the point in the area to be positioned of the patrol robot and the image information corresponding to the point in the repositioning area when the patrol robot loses its positioning.

[0079] Optionally, the building module 21 includes:

[0080] A selection unit, used for selecting a relocation area point of the patrol robot according to a preset interval distance;

[0081] The first control unit is used to start the patrol robot and initialize the position information of the patrol robot in the map, and control the patrol robot to walk along the patrol route;

[0082] The second control unit is used to control the pan-tilt camera to capture image information of the relocation area point at every preset rotation angle whenever the patrol robot walks to the relocation area point;

[0083] The associating unit is used to associate the image information and the posture information of the relocation area point to obtain the positioning information corresponding to the relocation area point.

[0084] Optionally, the relocation module 22 includes:

[0085] A first acquisition unit is used to acquire image information of a current area point collected by a pan / tilt camera on the patrol robot as image information of the area point to be located;

[0086] A second acquisition unit, used to use a preset neural network model to acquire the similarity between the image information corresponding to the relocated area point and the image information of the area point to be located;

[0087] The repositioning unit is used to use the posture information of the repositioning area point as the posture information of the area point to be positioned when there is at least one image information in the image information corresponding to the repositioning area point whose similarity is greater than or equal to a preset similarity threshold.

[0088] Optionally, the preset neural network model is a neural network implementation method of local aggregation descriptor vectors.

[0089] Optionally, the device further comprises:

[0090] Charging detection module, used to detect the charging status of the patrol robot;

[0091] If the patrol robot is in charging, the position and posture information of the charging location is obtained as the position and posture information of the area point to be positioned of the patrol robot.

[0092] The relocation module 22 also includes:

[0093] an identification unit, configured to identify lane lines according to image information captured by the PTZ camera when the similarities of the image information corresponding to the relocation area points are all less than the preset similarity threshold;

[0094] A comparison unit, used for controlling the patrol robot to move forward along the lane line, and comparing the travel distance of the patrol robot with a preset distance threshold;

[0095] A regional point acquisition unit, if the travel distance is less than or equal to the preset distance threshold, acquires the next regional point to be located at a preset frequency;

[0096] The repositioning unit is used to obtain the position and posture information of the point in the area to be positioned according to the image information of the point in the area to be positioned and the image information corresponding to the point in the repositioning area.

[0097] The relocation module 22 also includes:

[0098] The reporting unit is used to report a message of position loss to the server if the travel distance is greater than the preset distance threshold.

[0099] The specific definition of the repositioning device of the patrol robot can be found in the definition of the repositioning method of the patrol robot above, which will not be repeated here. Each module in the above-mentioned repositioning device of the patrol robot can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0100] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a repositioning method of a patrol robot is implemented.

[0101] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0102] Constructing positioning information corresponding to the repositioning area point, the positioning information including image information and posture information of the repositioning area point;

[0103] When the patrol robot loses its positioning, the position and posture information of the point in the area to be positioned of the patrol robot is obtained according to the image information of the point in the area to be positioned of the patrol robot and the image information corresponding to the point in the repositioning area.

[0104] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0105] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0106] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A patrol robot repositioning method, characterized in that: The method comprises: Constructing positioning information corresponding to a repositioning area point, wherein the repositioning area point is a reference position for the patrol robot to perform autonomous positioning and autonomous initialization on a patrol route, and the positioning information includes multiple image information and posture information of the repositioning area point; When the patrol robot loses its positioning, the position and posture information of the patrol robot's area point to be positioned is obtained according to the image information of the patrol robot's area point to be positioned and the image information corresponding to the repositioning area point; The step of acquiring the position and posture information of the point in the area to be located of the patrol robot according to the image information of the point in the area to be located of the patrol robot and the image information corresponding to the point in the relocation area comprises: Obtain the image information of the current area point collected by the pan-tilt camera on the patrol robot as the image information of the area point to be located; Using a preset neural network model, the similarity between the image information corresponding to the relocated area point and the image information of the area point to be located is obtained; When the similarity of at least one piece of image information in the image information corresponding to the relocated area point is greater than or equal to a preset similarity threshold, the position and posture information of the relocated area point is used as the position and posture information of the area point to be located; The positioning information corresponding to the constructed relocation area point includes: Select the relocation area points of the patrol robot according to the preset interval distance; Start the patrol robot and initialize the patrol robot's position information in the map, and control the patrol robot to walk along the patrol route; Whenever the patrol robot walks to a point in the relocation area, the pan-tilt camera is controlled to capture image information of the point in the relocation area at every preset rotation angle; The image information and the posture information of the relocation area point are associated to obtain the positioning information corresponding to the relocation area point.

2. The patrol robot repositioning method according to claim 1, characterized in that: The preset neural network model is a neural network implementation method for locally clustering descriptor vectors.

3. The patrol robot repositioning method according to claim 1, characterized in that: Before acquiring the position and posture information of the point in the area to be located of the patrol robot according to the image information of the point in the area to be located of the patrol robot and the image information corresponding to the point in the relocation area, the method further includes: Detect the charging status of the patrol robot; If the patrol robot is in charging, the position and posture information of the charging location is obtained as the position and posture information of the area point to be positioned of the patrol robot.

4. The patrol robot repositioning method according to claim 1, characterized in that: The step of obtaining the position and posture information of the point in the area to be located of the patrol robot according to the image information of the point in the area to be located of the patrol robot and the image information corresponding to the point in the relocation area further comprises: When the similarities of the image information corresponding to the relocation area points are all less than the preset similarity threshold, identifying the lane line according to the image information captured by the gimbal camera; Controlling the patrol robot to move forward along the lane line, and comparing the travel distance of the patrol robot with a preset distance threshold; If the travel distance is less than or equal to the preset distance threshold, acquiring the next area point to be located at a preset frequency; According to the image information of the point in the area to be positioned and the image information corresponding to the point in the repositioning area, the position and posture information of the point in the area to be positioned is obtained.

5. The patrol robot repositioning method according to claim 4, characterized in that: The step of obtaining the position and posture information of the point in the area to be located of the patrol robot according to the image information of the point in the area to be located of the patrol robot and the image information corresponding to the point in the relocation area further comprises: If the travel distance is greater than the preset distance threshold, a message indicating that the location is lost is reported to the server.

6. A patrol robot repositioning device, characterized in that: The device comprises: A construction module is used to construct positioning information corresponding to a repositioning area point, wherein the repositioning area point is a reference position for the patrol robot to perform autonomous positioning and autonomous initialization on a patrol route, and the positioning information includes multiple image information and posture information of the repositioning area point; A repositioning module, used for obtaining the position and posture information of the point in the area to be positioned of the patrol robot according to the image information of the point in the area to be positioned of the patrol robot and the image information corresponding to the point in the repositioning area when the patrol robot loses its position; The relocation module comprises: A first acquisition unit is used to acquire image information of a current area point collected by a pan / tilt camera on the patrol robot as image information of the area point to be located; A second acquisition unit, used to use a preset neural network model to acquire the similarity between the image information corresponding to the relocated area point and the image information of the area point to be located; A repositioning unit, configured to use the position and posture information of the repositioning area point as the position and posture information of the area point to be positioned when the similarity of at least one image information in the image information corresponding to the repositioning area point is greater than or equal to a preset similarity threshold; The building blocks include: A selection unit, used for selecting a relocation area point of the patrol robot according to a preset interval distance; The first control unit is used to start the patrol robot and initialize the position information of the patrol robot in the map, and control the patrol robot to walk along the patrol route; The second control unit is used to control the pan-tilt camera to capture image information of the relocation area point at every preset rotation angle whenever the patrol robot walks to the relocation area point; The associating unit is used to associate the image information and the posture information of the relocation area point to obtain the positioning information corresponding to the relocation area point.

7. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the patrol robot repositioning method according to any one of claims 1 to 5 is implemented.

8. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the patrol robot repositioning method according to any one of claims 1 to 5 is implemented.