Tunnel path positioning method and system

Abstract

The invention relates to the technical field of tunnel engineering, in particular to a tunnel path positioning method and system. The method comprises the following steps that at least one observation point is arranged on a path of a target tunnel, and the probability that a target observed object appears at the observation point is higher than the probability that the target observed object appears near the observation point; obtaining an engineering drawing of the target tunnel, and obtaining initial positioning data of the observation point in the target tunnel based on the engineering drawing; with the initial positioning data as the core and the path as the global range, a Gaussian probability model is constructed, and any point represents the occurrence probability of the target observation object at the corresponding position; in the target tunnel, obtaining a plurality of possible positioning data of the observation point based on the sensing positioning of the robot; calculating the credibility of each piece of possible positioning data based on a Gaussian probability model, and setting the possible positioning data with the highest credibility as final positioning data of the observation point; and summarizing final positioning data corresponding to all observation points in the path to obtain a positioning track of the target tunnel.

Description

Technical Field

[0001] The present invention relates to the technical field of tunnel engineering, and in particular to a tunnel path positioning method and system. Background Art

[0002] In underground tunnels and subway systems, accurate positioning technology plays a crucial role in applications such as autonomous navigation, safety inspection, equipment maintenance, and emergency rescue. However, due to challenges commonly present in the tunnel environment, such as insufficient lighting, sparse texture, and repetitive structure, traditional vision-based or laser-based positioning methods often struggle to obtain sufficient stable features, resulting in insufficient positioning accuracy and error accumulation. This, in turn, affects the path planning and navigation performance of inspection robots or automated equipment, reducing the overall efficiency and safety of operations. Summary of the Invention

[0003] The present invention provides a tunnel path positioning method and system, aiming to generate an automated high-precision positioning trajectory and provide comprehensive and reliable technical support for tunnel detection, maintenance, and three-dimensional reconstruction.

[0004] In one embodiment provided by the present invention, the tunnel path positioning method includes the following steps: S01. Set at least one observation point on the path of the target tunnel, where the probability of the target observation object appearing at the observation point is higher than the probability of appearing near the observation point; S02. Obtain the engineering drawings of the target tunnel, and based on the engineering drawings, obtain the initial positioning data of the observation point in the target tunnel; S03. With the initial positioning data as the core and the path as the global scope, construct a Gaussian probability model, where any point in the Gaussian probability model represents the probability of the target observation object appearing at the corresponding position; S04. In the target tunnel, based on the sensing positioning of the robot, obtain multiple possible positioning data of the observation point, and any possible positioning data is the actual measurement data of the robot for the observation point; S05. Based on the Gaussian probability model, calculate the credibility of each possible positioning data, and set the possible positioning data with the highest credibility as the final positioning data of the observation point; S06. Summarize the final positioning data corresponding to all observation points in the path to obtain the positioning trajectory of the target tunnel. By comprehensively utilizing the prior information of the tunnel engineering drawings and the real-time sensing data of the robot, the present invention realizes precise positioning at key observation points, summarizes and generates a complete high-precision tunnel positioning trajectory, effectively improves the positioning accuracy and robustness in harsh environments, and provides reliable data support for tunnel detection and three-dimensional reconstruction.

[0005] In this embodiment or some other embodiments, when only one observation point is set in the path and the output positioning coordinate data of the observation point is represented by two-dimensional plane coordinates, the first Gaussian probability model constructed in step S03 is: where represents the probability of the target object appearing at the position in the target tunnel under the first Gaussian probability model, represents the influence range coefficient of the target object in the x direction, represents the influence range coefficient of the target object in the y direction, and represents the initial positioning data of the observation point. By adopting the first Gaussian probability model, the present invention precisely defines the influence range of the target object in the x and y directions with the initial positioning data of a single two-dimensional observation point as the center, enabling the candidate positioning data to highly match the prior information, thereby achieving high-precision positioning and reducing measurement errors.

[0006] In some other embodiments, when multiple observation points are set in the path and the output positioning coordinate data of the observation points is represented by two-dimensional plane coordinates, the second Gaussian probability model constructed in step S03 is: where represents the probability of the target object appearing at the position in the target tunnel under the second Gaussian probability model, is the total number of observation points, represents the influence range coefficient of the target object in the x direction at the i-th observation point, represents the influence range coefficient of the target object in the y direction at the i-th observation point, and represents the initial positioning data of the i-th observation point. By adopting the second Gaussian probability model, the present invention weights and superimposes the initial positioning data of multiple two-dimensional observation points to construct a comprehensive probability distribution, effectively integrating multi-source prior information, enhancing the stability and accuracy of the entire tunnel path positioning, and laying a solid foundation for subsequent global optimization.

[0007] In some other embodiments, when one observation point is set in the path and the output positioning coordinate data of the observation point is represented by three-dimensional space coordinates, the third Gaussian probability model constructed in step S03 is: where represents the probability of the target object appearing at the position in the target tunnel under the third Gaussian probability model, represents the influence range coefficient of the target object in the x direction, represents the influence range coefficient of the target object in the y direction, represents the influence range coefficient of the target object in the z direction, and represents the initial positioning data of the observation point. By adopting the third Gaussian probability model, based on a single three-dimensional observation point, the present invention comprehensively considers the uncertainties in the x, y, and z directions and precisely defines the appearance probability of the target object in three-dimensional space, thereby achieving high-precision three-dimensional positioning of the observation point in a complex tunnel environment.

[0008] In some other embodiments, when a plurality of observation points are provided in the path and the output positioning coordinate data of the observation points are represented by three-dimensional space coordinates, the fourth Gaussian probability model constructed in step S03 is: where represents the probability of the target object appearing at the position in the target tunnel under the fourth Gaussian probability model, is the total number of observation points, represents the weight of the i-th observation point, represents the influence range coefficient of the target object appearing in the x direction at the i-th observation point, represents the influence range coefficient of the target object appearing in the y direction at the i-th observation point, represents the influence range coefficient of the target object appearing in the z direction at the i-th observation point, and represents the initial positioning data of the i-th observation point. By adopting the fourth Gaussian probability model, the present invention performs weighted superposition on the initial positioning data of multiple three-dimensional observation points to construct a global three-dimensional probability distribution, accurately describes the probability of the target object appearing in each dimension, and significantly improves the overall positioning accuracy and the tunnel three-dimensional reconstruction effect.

[0009] In this embodiment or some other embodiments, when there are no duplicate items among the multiple possible positioning data corresponding to the same observation point, step S05 includes the following sub-steps: substituting the initial positioning data of the observation point into the Gaussian probability model, and respectively substituting the multiple possible positioning data into the updated Gaussian probability model to obtain the credibility of each possible positioning data, and setting the possible positioning data with the highest credibility as the final positioning data of the observation point. By substituting each candidate positioning data into the Gaussian probability model to calculate the credibility and directly selecting the data with the highest credibility as the final positioning when there are no duplicate candidate data, the present invention realizes fast and accurate data screening, simplifies the processing flow, and improves the response speed.

[0010] In some other embodiments, when there is a set of duplicate items among the multiple possible positioning data corresponding to the same observation point, step S05 includes the following sub-steps: setting the possible positioning data corresponding to the duplicate items as the final positioning data of the observation point. By directly using the set of duplicate data as the final positioning data when there is a set of duplicate items among the candidate data of the same observation point, the present invention utilizes the data centrality to reflect the on-site measurement stability, thereby ensuring the accuracy of the positioning result and the simplicity of the processing flow, and reducing the calculation redundancy.

[0011] In some other embodiments, when there are multiple sets of duplicate items among the multiple possible positioning data corresponding to the same observation point, step S05 includes the following sub-steps: Substitute the initial positioning data of the observation point into the Gaussian probability model, and substitute the possible positioning data corresponding to multiple sets of duplicate items into the updated Gaussian probability model respectively to obtain the credibility of the possible positioning data corresponding to each set of duplicate items, and set the possible positioning data with the highest credibility as the final positioning data of the observation point. By calculating the intra-group credibility of each set of data by substituting them into the Gaussian probability model respectively when there are multiple sets of duplicate items in the candidate data of the same observation point, and then selecting the representative data of the group with the highest credibility as the final positioning result, the present invention effectively merges the scattered data, improves the positioning consistency and accuracy, and adapts to complex measurement environments.

[0012] The present invention also provides a tunnel path positioning system, which includes an input device, a processor, a memory, and an output device; wherein, the input device, the processor, the memory, and the output device are interconnected, the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute the tunnel path positioning method provided by the above embodiments. By integrating the input device, the processor, the memory, and the output device into the tunnel path positioning system, and using the computer program to process the engineering drawing and sensing data in real time, the present invention realizes the generation of an automated high-precision positioning trajectory, providing comprehensive and reliable technical support for tunnel detection, maintenance, and three-dimensional reconstruction. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 is a flowchart of the tunnel path positioning method provided by an embodiment of the present invention;

[0014] Figure 2 is a structural diagram of the tunnel path positioning system provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0015] In the following description, specific details such as specific system structures and technologies are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present application; those skilled in the art should understand that the present application can also be implemented in other embodiments without these specific details.

[0016] In the description of the present application, the detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present application; in addition, terms such as "first" and "second" are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0017] In one embodiment, the tunnel path positioning method provided by the present invention includes the following Figure 1 flow steps as shown:

[0018] S01. Set at least one observation point on the path of the target tunnel, where the probability of the target observation object appearing at the observation point is higher than that near the observation point.

[0019] In this embodiment, the target tunnel refers to the actual tunnel structure that needs to be detected, located, or three-dimensionally reconstructed, such as a subway tunnel, an underground tunnel, etc.

[0020] Furthermore, the path on the target tunnel refers to the predetermined route for the robot to move in the target tunnel, which usually appears as the central axis of the tunnel or a curve along the track direction.

[0021] Even further, the observation points set on the path refer to the key positions preset in the target tunnel, where the probability of the target observation object appearing is significantly higher than that in the surrounding area.

[0022] Still further, the target observation object in the target tunnel refers to the object that needs to be detected and inspected during the tunnel inspection process, which usually has obvious geometric structure or material characteristics, including but not limited to key structures of the track, fasteners, support members, linings, fixing devices, etc.

[0023] S02. Obtain the engineering drawings of the target tunnel, and based on the engineering drawings, obtain the initial positioning data of the observation points in the target tunnel.

[0024] In this embodiment, the engineering drawings refer to the design blueprints for tunnel construction, which detail information such as the overall orientation of the tunnel, structural steps, installation positions and dimensions of key components, etc. It can be further understood that there are differences between the parameters in the engineering drawings and the actual situation in the tunnel.

[0025] Even further, the initial positioning data of the observation points extracted from the engineering drawings is usually represented by two-dimensional plane coordinates or three-dimensional space coordinates.

[0026] It should be noted that in subsequent steps, the representation of the possible positioning data obtained by the robot is consistent with the initial positioning data, and the positioning data is expressed based on the same coordinate system.

[0027] S03. With the initial positioning data as the core and the path as the global scope, construct a Gaussian probability model, where any point in the Gaussian probability model represents the probability of the target observation object appearing at the corresponding position.

[0028] In some embodiments, when only one observation point is set in the path and the output positioning coordinate data of the observation point is represented by two-dimensional plane coordinates, the first Gaussian probability model constructed in step S03 is: where represents the probability of the target object appearing at the position in the target tunnel under the first Gaussian probability model, represents the influence range coefficient of the target object in the x direction, represents the influence range coefficient of the target object in the y direction, and represents the initial positioning data of the observation point.

[0029] In some embodiments, when multiple observation points are set in the path and the output positioning coordinate data of the observation point is represented by two-dimensional plane coordinates, the second Gaussian probability model constructed in step S03 is: where represents the probability of the target object appearing at the position in the target tunnel under the second Gaussian probability model, is the total number of observation points, represents the influence range coefficient of the target object in the x direction at the i-th observation point, represents the influence range coefficient of the target object in the y direction at the i-th observation point, and represents the initial positioning data of the i-th observation point.

[0030] In some embodiments, when only one observation point is set in the path and the output positioning coordinate data of the observation point is represented by three-dimensional space coordinates, the third Gaussian probability model constructed in step S03 is: where represents the probability of the target object appearing at the position in the target tunnel under the third Gaussian probability model, represents the influence range coefficient of the target object in the x direction, represents the influence range coefficient of the target object in the y direction, represents the influence range coefficient of the target object in the z direction, and represents the initial positioning data of the observation point.

[0031] In some embodiments, when multiple observation points are set in the path and the output positioning coordinate data of the observation point is represented by three-dimensional space coordinates, the fourth Gaussian probability model constructed in step S03 is: where represents the probability of the target object appearing at the position in the target tunnel under the fourth Gaussian probability model, is the total number of observation points, represents the weight of the i-th observation point, represents the influence range coefficient of the target object in the x direction at the i-th observation point, represents the influence range coefficient of the target object in the y direction at the i-th observation point, represents the influence range coefficient of the target object in the z direction at the i-th observation point, and represents the initial positioning data of the i-th observation point.

[0032] In any of the above embodiments, the influence range coefficient is used to describe the attenuation speed of the probability of actually detecting the target object in the corresponding direction as the distance deviates from the initial coordinate; the weight is used to reflect the relative contribution of the corresponding observation point in the overall probability model.

[0033] S04. In the target tunnel, based on the sensing and positioning of the robot, obtain multiple possible positioning data of the observation point, and any possible positioning data is the actual measurement data of the robot for the observation point.

[0034] Further, the robot can collect the positioning data of the observation point by means of a variety of sensors such as lidar, camera, inertial measurement unit (IMU), and wheel speedometer; and through the processor connected thereto, calculate the potential positioning data in the same coordinate system as the initial positioning data.

[0035] Furthermore, among the multiple potential positioning data corresponding to the same observation point, there may be duplicate data or there may be no duplicate data.

[0036] S05. Based on the Gaussian probability model, calculate the credibility of each possible positioning data, and set the possible positioning data with the highest credibility as the final positioning data of the observation point.

[0037] In some embodiments, when there are no duplicate items among the multiple possible positioning data corresponding to the same observation point, step S05 specifically includes the following sub-steps: Substitute the initial positioning data of the observation point into the Gaussian probability model, and substitute the multiple possible positioning data into the updated Gaussian probability model respectively to obtain the credibility of each possible positioning data, and set the possible positioning data with the highest credibility as the final positioning data of the observation point.

[0038] In some embodiments, when there is a set of duplicate items among the multiple possible positioning data corresponding to the same observation point, step S05 specifically includes the following sub-steps: Set the possible positioning data corresponding to the duplicate items as the final positioning data of the observation point.

[0039] In some embodiments, when there are multiple sets of duplicate items among the multiple possible positioning data corresponding to the same observation point, step S05 specifically includes the following sub-steps: Substitute the initial positioning data of the observation point into the Gaussian probability model, and substitute the possible positioning data corresponding to the multiple sets of duplicate items into the updated Gaussian probability model respectively to obtain the credibility of the possible positioning data corresponding to each set of duplicate items, and set the possible positioning data with the highest credibility as the final positioning data of the observation point.

[0040] S06. Summarize the final positioning data corresponding to all the observation points in the path to obtain the positioning trajectory of the target tunnel.

[0041] It can be further understood that when continuously setting the observation points on the path, step S06 will generate a corresponding continuous positioning trajectory.

[0042] In one embodiment, please refer to Figure 2, the present invention also provides a tunnel path positioning system; further, as Figure 2 shown, the tunnel path positioning system includes an input device, a processor, a memory, and an output device; furthermore, the input device, the processor, the memory, and the output device are interconnected, the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute the tunnel path positioning method provided in the above embodiments.

[0043] In the above embodiments, the descriptions of the respective embodiments have their own focuses. For parts not described or recorded in detail in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

[0044] It should be noted that the above embodiments can be freely combined as needed. The above is only the preferred implementation mode of the present invention; it should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A tunnel path positioning method, characterized in that: The steps include: S01. On the path of the target tunnel, at least one observation point is set, and the probability of the target observation object appearing at the observation point is higher than the probability of appearing near the observation point; S02, obtaining an engineering drawing of the target tunnel, and obtaining initial positioning data of the observation point in the target tunnel based on the engineering drawing; S03, taking the initial positioning data as the core and the path as the entire domain, constructing a Gaussian probability model, wherein any point in the Gaussian probability model represents the probability of the target observation object appearing at the corresponding position; S04. In the target tunnel, based on the sensor positioning of the robot, a plurality of possible positioning data of the observation point are obtained, wherein any possible positioning data is actual measurement data of the observation point by the robot; S05. Calculate the credibility of each possible positioning data based on the Gaussian probability model, and set the possible positioning data with the highest credibility as the final positioning data of the observation point; S06. Summarize the final positioning data corresponding to all observation points in the path to obtain the positioning trajectory of the target tunnel.

2. The tunnel path positioning method according to claim 1, characterized in that: When only one observation point is set in the path and the output positioning coordinate data of the observation point is represented by two-dimensional plane coordinates, the first Gaussian probability model constructed in step S03 is: , wherein represents the probability of the target observation object appearing at the position in the target tunnel under the first Gaussian probability model, represents the influence range coefficient of the target observation object appearing in the x direction, represents the influence range coefficient of the target observation object appearing in the y direction, and represents the initial positioning data of the observation point.

3. The tunnel path positioning method according to claim 1, characterized in that: When multiple observation points are set in the path and the output positioning coordinate data of the observation points are represented by two-dimensional plane coordinates, the second Gaussian probability model constructed in step S03 is: , wherein represents the probability of the target observation object appearing at the position in the target tunnel under the second Gaussian probability model, is the total number of observation points, represents the influence range coefficient of the target observation object appearing in the x direction at the i-th observation point, represents the influence range coefficient of the target observation object appearing in the y direction at the i-th observation point, and represents the initial positioning data of the i-th observation point.

4. The tunnel path positioning method according to claim 1, characterized in that: An observation point is set in the path, and when the output positioning coordinate data of the observation point is represented by three-dimensional space coordinates, the third Gaussian probability model constructed in step S03 is: , wherein represents the probability of the target observation object appearing at the position in the target tunnel under the third Gaussian probability model, represents the influence range coefficient of the target observation object appearing in the x direction, represents the influence range coefficient of the target observation object appearing in the y direction, represents the influence range coefficient of the target observation object appearing in the z direction, and represents the initial positioning data of the observation point.

5. The tunnel path positioning method according to claim 1, characterized in that: When multiple observation points are set in the path and the output positioning coordinate data of the observation points are represented by three-dimensional space coordinates, the fourth Gaussian probability model constructed in step S03 is: , wherein represents the probability of the target observation object appearing at the position in the target tunnel under the fourth Gaussian probability model, is the total number of observation points, represents the weight of the i-th observation point, represents the influence range coefficient of the target observation object appearing in the x direction at the i-th observation point, represents the influence range coefficient of the target observation object appearing in the y direction at the i-th observation point, represents the influence range coefficient of the target observation object appearing in the z direction at the i-th observation point, and represents the initial positioning data of the i-th observation point.

6. The tunnel path positioning method according to any one of claims 2 to 5, characterized in that: When there are no duplicates in the multiple possible positioning data corresponding to the same observation point, step S05 includes the following sub-steps: substituting the initial positioning data of the observation point into the Gaussian probability model, and substituting the multiple possible positioning data into the updated Gaussian probability model respectively, obtaining the credibility of each possible positioning data, and setting the possible positioning data with the highest credibility as the final positioning data of the observation point.

7. The tunnel path positioning method according to any one of claims 2 to 5, characterized in that: When there is a set of duplicate items in the multiple possible positioning data corresponding to the same observation point, step S05 includes the following sub-steps: setting the possible positioning data corresponding to the duplicate items as the final positioning data of the observation point.

8. The tunnel path positioning method according to any one of claims 2 to 5, characterized in that: When there are multiple sets of duplicate items in the multiple possible positioning data corresponding to the same observation point, step S05 includes the following sub-steps: substituting the initial positioning data of the observation point into the Gaussian probability model, and substituting the possible positioning data corresponding to the multiple sets of duplicate items into the updated Gaussian probability model respectively, obtaining the credibility of the possible positioning data corresponding to each set of duplicate items, and setting the possible positioning data with the highest credibility as the final positioning data of the observation point.

9. A tunnel path positioning system, characterized in that: The vehicle positioning system includes an input device, a processor, a memory and an output device; Wherein, the input device, the processor, the memory and the output device are interconnected, the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute the tunnel path positioning method as described in any one of claims 1 to 8.

Images