A high-precision tunnel boring machine operating posture recognition method

Abstract

This invention relates to the field of image processing technology, specifically to a high-precision method for recognizing the operating posture of a tunnel boring machine (TBM). The method includes: acquiring point cloud data; obtaining the structural confidence score of each TBM image point in each monitoring frame based on the overall movement of the same TBM image point in different monitoring frames; determining the consistency of positional changes of each TBM image point in each monitoring frame based on the structural confidence score; determining the necessity of adjusting the distance error between any two data points based on the structural differences in the consistency of positional changes; obtaining an adaptive distance metric between any two data points based on the necessity of adjusting the distance error; and performing matching detection based on the adaptive distance metric between any two data points. This invention reduces the probability of data point matching errors, improves the reliability of matching results, and increases the efficiency of posture detection.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and more specifically to a high-precision method for recognizing the operating posture of a tunnel boring machine. Background Technology

[0002] During tunnel excavation in soft soil layers, tunnel boring machines (TBMs) are used to ensure the smooth progress of road reconstruction and other engineering projects. Among these, the shield tunneling machine (TBM), suitable for excavating in soft soil layers, is frequently used. Then, using technologies such as micro-motion radar detection, point cloud data is dynamically extracted from the environmental conditions of the TBM's operating area and the TBM's own operating posture to achieve precise identification and control of the TBM's operating status.

[0003] Existing technologies typically use the PL-ICP algorithm to match point cloud data for real-time detection of tunnel boring machine (TBM) attitude. However, the point cloud data of some actual defect areas on the TBM can interfere with the distance relationships between different data points in the point cloud data, causing some data points to be mismatched, reducing the reliability of the matching results, and decreasing the efficiency of attitude detection. Summary of the Invention

[0004] This invention provides a high-precision method for recognizing the operating posture of a tunnel boring machine (TBM) to solve the existing problem: In the process of the PL-ICP algorithm matching point cloud data to detect the posture of the TBM in real time, the point cloud data of some actual defect areas on the TBM will interfere with the distance relationship between different data points in the point cloud data, causing some data points to be mismatched.

[0005] The high-precision tunnel boring machine operation posture recognition method of the present invention adopts the following technical solution:

[0006] Includes the following steps:

[0007] Point cloud data of the cutterhead area and the working face area of ​​the tunnel boring machine are collected in several monitoring frames. The point cloud data contains multiple shield image points with three-dimensional spatial location information. Each shield image point corresponds to a data point in each monitoring frame.

[0008] The cutterhead area and the working face area of ​​the tunnel boring machine are both recorded as a tunnel boring machine monitoring area. Based on the overall movement of the same tunnel boring machine image point in different monitoring frames, the structural confidence of each tunnel boring machine image point in each monitoring frame is obtained.

[0009] Based on the magnitude of positional changes of different shield image points in the same monitoring frame and the structural confidence level, the consistency of positional changes of each shield image point in the cutterhead area of ​​the tunnel boring machine in each monitoring frame is obtained. Based on the structured differences in the consistency of positional changes between different data points of each shield image point in each monitoring frame, the necessity of adjusting the distance error between any two data points of each shield image point in each monitoring frame is obtained. Based on the necessity of adjusting the distance error between different data points, the distance metric between different data points is adjusted to obtain an adaptive distance metric between any two data points.

[0010] Matching detection is performed based on an adaptive distance metric between any two data points.

[0011] Preferably, the method for obtaining the structural confidence level of each shield tunneling image point in each monitoring frame based on the overall movement of the same shield tunneling image point in different monitoring frames includes:

[0012] Obtain the face region in the first... The weighting coefficient of the tunnel face in the frame monitoring frame and the weighting coefficient of each shield image point in the frame. Frame monitoring: the structural confidence coefficient within the frame;

[0013] For any shield image point in the cutterhead area of ​​the tunnel boring machine, the shield image point is set to the 1st... The structural confidence coefficient in the frame monitoring frame is related to the tunnel face region in the 1st frame. The product of the face weight coefficients of the monitoring frames is denoted as the shield image point at the [number]th frame. Frame monitoring: structural confidence within frames.

[0014] Preferably, the acquisition of the face region is in the first... The weighting coefficient of the tunnel face in the frame monitoring frame and the weighting coefficient of each shield image point in the frame. The specific methods for monitoring the structural confidence coefficient in a frame include:

[0015] For any shield image point in the cutterhead area of ​​the tunnel boring machine, the shield image point is set to the 1st... The data points in the frame monitoring frame and the shield tunneling image points at the 1st frame... The Euclidean distance between data points in the frame monitoring frame is denoted as the first distance; the shield image points are then used to determine the distance in the second frame. The data points in the frame monitoring frame and the shield tunneling image points at the 1st frame... The Euclidean distance between data points in the frame monitoring frame is denoted as the second distance; the absolute value of the difference between the first distance and the second distance is denoted as the shield image point at the first [missing value]. The structural confidence factor in the frame monitoring is calculated; the structural confidence factor of the shield tunneling image points in all monitoring frames is linearly normalized, and the normalized structural confidence factor is recorded as the structural confidence coefficient.

[0016] Record any shield tunneling image point in the tunnel face area as the target tunnel face image point; the reference shield tunneling image point is at the... The method for obtaining the structure confidence factor in the frame monitoring frame is to obtain the target face image point in the [frame number missing]. The structural confidence factor in the frame monitoring frame is used to analyze all shield tunneling image points in the tunnel face region in the 1st frame. The sum of the structural confidence factors in the frame monitoring frame is denoted as the sum of the structural confidence factors in the face region at the 1st frame. The face weight of the monitoring frame is calculated; the face weight of the face region in all monitoring frames is linearly normalized, and each normalized face weight is recorded as the face weight coefficient.

[0017] Preferably, the method for obtaining the consistency of positional change of each shield image point in the shield machine cutterhead area in each monitoring frame based on the magnitude of positional changes of different shield image points in the same monitoring frame and the structural confidence level includes the following specific methods:

[0018] The offset vector of each monitoring frame in the cutterhead area of ​​the tunnel boring machine is obtained based on the structural confidence level;

[0019] According to the first [section / area] of the tunnel boring machine cutterhead area The shield tunneling image point at the ... The difference in structural confidence between the monitoring frame and other monitoring frames, as well as the difference in offset vectors between adjacent monitoring frames, are used to obtain the first [frame name] of the shield machine cutterhead region. The shield tunneling image point at the ... The specific methods for monitoring the consistency of frame position changes are as follows:

[0020]

[0021] In the formula, The first area representing the cutterhead region of the tunnel boring machine The shield tunneling image point at the ... Frame monitoring ensures consistency in frame position changes; This indicates the number of all clusters in the cutterhead region of the tunnel boring machine; The first area representing the cutterhead region of the tunnel boring machine The offset vectors of each cluster; The first area representing the cutterhead region of the tunnel boring machine The offset vectors of each cluster; This represents the preset denominator hyperparameter; The first area representing the cutterhead region of the tunnel boring machine The shield tunneling image point at the ... Frame monitoring: Frame structural confidence. The first area representing the cutterhead region of the tunnel boring machine The mean of the structural confidence scores of each shield tunneling image point across all monitoring frames; Represents an exponential function with the natural constant as its base; This indicates taking the absolute value.

[0022] Preferably, the specific method for obtaining the offset vector of each monitoring frame in the shield machine cutterhead region based on structural confidence is as follows:

[0023] For any monitoring frame in the cutterhead region of the tunnel boring machine (TBM), the mean of the structural confidence scores of all shield image points in the TBM cutterhead region within the monitoring frame is recorded as the structural confidence coefficient of the monitoring frame. The structural confidence coefficients of all monitoring frames are obtained. The absolute value of the difference between the structural confidence coefficients of different monitoring frames is used as a distance metric. Based on the distance metric, hierarchical clustering is performed on all monitoring frames to obtain several clusters. For any cluster, the least squares method is used to fit all shield image points in the TBM cutterhead region to the data points of all monitoring frames in the cluster, resulting in a fitted straight line for the TBM cutterhead region within the cluster. The vector formed by the fitted straight line is recorded as the offset vector.

[0024] Preferably, the specific method for determining the necessity of adjusting the distance error between any two data points in each monitoring frame based on the structured differences in the consistency of positional changes between different data points of each shield image point in each monitoring frame includes:

[0025] All shield image points in the cutterhead area of ​​the tunnel boring machine at the 1st Frame monitoring frame and the first Data points in the monitoring frames are density-clustered to obtain several clusters; each cluster is recorded as a monitoring cluster; any shield image point in the cutterhead area of ​​the tunnel boring machine is recorded as a target data point in any monitoring frame, and a reference range is preset. Centered on the data point, with a radius of... The area is denoted as the neighborhood monitoring area of ​​the target data point;

[0026] Any data point in the neighborhood monitoring area of ​​the target data point is recorded as the monitoring reference data point; based on the difference in the consistency of positional change between the target data point and the monitoring reference data point as a whole, the necessary factor for adjusting the distance error between the target data point and the monitoring reference data point is obtained;

[0027] Obtain the necessary distance error adjustment factors between all data points in the neighborhood monitoring area of ​​the target data point and the target data point. Perform linear normalization on all necessary distance error adjustment factors and record each normalized necessary distance error adjustment factor as the distance error adjustment necessity.

[0028] Preferably, the method for obtaining the necessary distance error adjustment factor between the target data point and the monitoring reference data point based on the difference in the consistency of positional changes between the target data point and the overall monitoring cluster to which the monitoring reference data point belongs includes:

[0029]

[0030] In the formula, This represents the necessary factor for adjusting the distance error between the target data point and the monitoring reference data point. This indicates the consistency of the positional changes of the target data points; This indicates the consistency of the location changes of the monitoring reference data points; This represents the mean of the consistency in positional changes among all data points in the monitoring cluster to which the monitoring reference data point belongs; This represents the Euclidean distance between the target data point and the monitoring reference data point; This represents the preset denominator reference hyperparameter.

[0031] Preferably, the specific method for adjusting the distance metric between different data points based on the necessity of adjusting the distance error between different data points to obtain an adaptive distance metric between any two data points includes:

[0032] Based on the necessity of distance error adjustment, obtain all data point pairs to be adjusted and fixed data point pairs;

[0033] For any pair of data points to be adjusted, the product of the Euclidean distance between the two data points and the necessity of adjusting the distance error is denoted as the adaptive distance metric between the two data points; the adaptive distance metric between the two data points in each pair of data points to be adjusted is obtained; the Euclidean distance between the two data points in each fixed pair of data points is used as the adaptive distance metric between the two data points in each fixed pair of data points.

[0034] Preferably, the specific method for obtaining all pairs of data points to be adjusted and fixed data points based on the necessity of distance error adjustment includes:

[0035] Preset a distance error adjustment necessity threshold For any two shield image points in the cutterhead area of ​​the tunnel boring machine, if the distance error between these two data points is more than [a certain value], then [the following condition applies]. The data point pair formed by these two data points is denoted as the data point pair to be adjusted; if the distance error between these two data points is less than or equal to the required adjustment... The data point pair formed by these two data points is denoted as a fixed data point pair.

[0036] Preferably, the specific method for performing matching detection based on the adaptive distance metric between any two data points is as follows:

[0037] The adaptive distance metric between any two different data points is used as the distance metric, and the point cloud data matching result is obtained by matching based on the distance metric using the PL-ICP algorithm.

[0038] The beneficial effects of the technical solution of this invention are as follows: Based on the overall movement of the same shield tunneling image point in different monitoring frames, the structural confidence of each shield tunneling image point in each monitoring frame is obtained; based on the magnitude of the change in position information of different shield tunneling image points in the same monitoring frame, the consistency of position change of each shield tunneling image point in each monitoring frame is obtained; based on the structural differences in the consistency of position change between different data points in each monitoring frame, the necessity of adjusting the distance error between any two data points in each monitoring frame is obtained; based on the necessity of adjusting the distance error between different data points, the distance measurement between different data points is adjusted to obtain the consistency of position change between any two data points in each monitoring frame. An adaptive distance metric between two data points is used for matching detection based on this adaptive distance metric. Position change consistency describes the degree of change in the tunnel boring machine's (TBM) operating posture, indicating a slow settlement or offset trend, making the association between the TBM's operating posture and point cloud data clearer. Distance error adjustment necessity describes the structural changes between different data points caused by the TBM's wear area or other environmental factors, making the influence of actual defect areas on the distance metric during point cloud data matching more explicit. This reduces the probability of data point matching errors, improves the reliability of matching results, and increases the efficiency of posture detection. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 This is a flowchart illustrating the steps of a high-precision tunnel boring machine operation posture recognition method according to the present invention. Detailed Implementation

[0041] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a high-precision tunnel boring machine operation posture recognition method proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0042] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0043] The following description, in conjunction with the accompanying drawings, details the specific scheme of the high-precision tunnel boring machine operation posture recognition method provided by the present invention.

[0044] Please see Figure 1 The diagram illustrates a flowchart of a high-precision tunnel boring machine operation posture recognition method according to an embodiment of the present invention. The method includes the following steps:

[0045] Step S001: Collect point cloud data of the cutterhead area and the tunnel face area of ​​the tunnel boring machine in several monitoring frames.

[0046] It should be noted that existing technologies typically use the PL-ICP algorithm to match point cloud data and perform real-time detection of the tunnel boring machine's attitude. In the process, the point cloud data of some actual defect areas on the tunnel boring machine can interfere with the distance relationship between different data points in the point cloud data, causing some data points to be matched incorrectly, reducing the reliability of the matching results and reducing the efficiency of attitude detection.

[0047] Specifically, the first step is to collect point cloud data and several monitoring frames for each shield tunneling monitoring area. The process is as follows: The shield machine's cutterhead area and face area are both designated as the shield machine monitoring area; point cloud data for each monitoring area is acquired using 3D laser scanning, and each data point in the point cloud data is recorded as a shield image point; taking any given shield machine monitoring area as an example, several monitoring frames for that area over the past day are obtained from the shield machine's historical monitoring frame database, with each frame taken every 5 seconds; several monitoring frames for each shield machine monitoring area are then acquired. The time interval and total duration of recording monitoring frames can be determined based on the specific implementation requirements.

[0048] It should be noted that the cutterhead area of ​​the tunnel boring machine mainly consists of the main beam, auxiliary beam, roller cutter, scraper, side scraper, and over-excavation cutter; the face area mainly consists of the working face area during the tunnel boring machine's advancement process; the point cloud data of each tunnel boring machine monitoring area contains multiple data points, and each data point corresponds to a data point with three-dimensional spatial location information in each monitoring frame, and the three-dimensional spatial location corresponding to each data point is not always the same in different monitoring frames.

[0049] Thus, point cloud data of the tunnel boring machine cutterhead area and the tunnel face area in several monitoring frames were obtained using the above method.

[0050] Step S002: Based on the overall movement of the same shield tunnel image point in different monitoring frames, obtain the structural confidence of each shield tunnel image point in each monitoring frame.

[0051] It should be noted that the image data related to the tunnel boring machine's operating environment mainly includes point cloud data of the cutterhead assembly structure in the cutterhead area and point cloud data of the corresponding soil structure in the tunnel face area. The positions of different data points in the point cloud data are affected by environmental factors such as vibration and the soil structure of the tunnel face in different operating segments, making it difficult to determine the degree of wear on different data points in each shield monitoring area. When the positions of corresponding data points change within multiple monitoring frames, the positions of the data points caused by wear will slowly change over time. Therefore, by combining the change characteristics of structural points in the shield monitoring area, the structural confidence of each data point can be obtained, so as to achieve accurate cleaning of data points and complete the accurate visualization processing of the tunnel boring machine cutter and operating environment.

[0052] Preferably, in one embodiment of the present invention, the structural confidence of each shield tunneling image point in each monitoring frame is obtained based on the overall movement of the same shield tunneling image point in different monitoring frames. The specific method includes:

[0053] Taking any shield image point in the cutterhead area of ​​the tunnel boring machine as an example, the shield image point is placed at the 1st... The data points in the frame monitoring frame and the shield tunneling image points at the 1st frame... The Euclidean distance between data points in the frame monitoring frame is denoted as the first distance; the shield image point is then used in the second frame... The data points in the frame monitoring frame and the shield tunneling image points at the 1st frame... The Euclidean distance between data points in the frame monitoring frame is denoted as the second distance; the absolute value of the difference between the first distance and the second distance is denoted as the distance between the shield tunneling image points in the first and second frames. The structural confidence factor in the frame monitoring frame; the structural confidence factor of the shield image point in the shield machine cutterhead area in all monitoring frames is linearly normalized, and the normalized structural confidence factor is recorded as the structural confidence coefficient.

[0054] Furthermore, any shield tunneling image point in the tunnel face area is designated as the target tunnel face image point; this shield tunneling image point is referenced in the [missing information - likely a specific location or event]. The method for obtaining the structure confidence factor in the frame monitoring frame is to obtain the target face image point in the [frame number missing]. The structural confidence factor in the frame monitoring frame is used to analyze all shield tunneling image points in the tunnel face region in the 1st frame. The sum of the structural confidence factors in the frame monitoring frame is denoted as the sum of the structural confidence factors in the face region at the 1st frame. The face weight of the monitoring frame is calculated; the face weight of the face region in all monitoring frames is linearly normalized, and each normalized face weight is recorded as the face weight coefficient.

[0055] Furthermore, taking any shield image point in the cutterhead area of ​​the tunnel boring machine as an example, the shield image point is set at the 1st... The structural confidence coefficient in the frame monitoring frame is related to the tunnel face region in the 1st frame. The product of the face weight coefficients of the monitoring frames is denoted as the number of shield tunneling image points at the [number]th frame. Frame monitoring: structural confidence within frames.

[0056] It should be noted that, by default, this embodiment uses the Euclidean distance between the data points of the shield tunneling image point in the first monitoring frame and the data points of the shield tunneling image point in the second monitoring frame as the structural confidence factor of the shield tunneling image point in the first monitoring frame; and uses the Euclidean distance between the data points of the shield tunneling image point in the last monitoring frame and the data points of the shield tunneling image point in the penultimate monitoring frame as the structural confidence factor of the shield tunneling image point in the last monitoring frame.

[0057] Thus, the structural confidence level of each shield tunneling image point in each monitoring frame is obtained using the above method.

[0058] Step S003: Based on the magnitude of positional changes of different shield image points in the same monitoring frame and the structural confidence level, obtain the consistency of positional changes of each shield image point in the cutterhead area of ​​the tunnel boring machine in each monitoring frame; based on the structural differences in the consistency of positional changes between different data points of each shield image point in each monitoring frame, obtain the necessity of adjusting the distance error between any two data points of each shield image point in each monitoring frame; based on the necessity of adjusting the distance error between different data points, adjust the distance metric between different data points to obtain an adaptive distance metric between any two data points.

[0059] It should be noted that the structural confidence score of each shield tunneling image point in each monitoring frame can be used to determine the shield machine's operating attitude. However, the actual operating attitude of the shield machine is greatly affected by the working environment. When environmental factors such as rainwater deposition occur, the soil structure changes, and the shield machine's operating attitude shows a slow settlement or displacement trend. Therefore, based on the structural change characteristics of each shield tunneling image point in each monitoring frame, the accurate determination of the shield machine's operating attitude can be achieved. Furthermore, since the data cleaning process uses matching based on the corresponding location's neighboring points, it can only remove relatively obvious abnormal point cloud data caused by environmental factors affecting the shield machine's location. Data points in actual defect areas will also affect the determination of the shield machine's operating attitude. Therefore, it is necessary to analyze the operating attitude from the structural offset trend of all point cloud data in the shield machine area, and reduce the influence of point cloud data at defect locations through fine registration using the PL-ICP algorithm to achieve accurate determination of the shield machine's operating attitude.

[0060] Preferably, in one embodiment of the present invention, the consistency of positional change of each shield image point in the cutterhead area of ​​the tunnel boring machine in each monitoring frame is obtained based on the magnitude of the change in positional information of different shield image points in the same monitoring frame. The specific method includes:

[0061] Taking any monitoring frame of the tunnel boring machine (TBM) cutterhead region as an example, the mean of the structural confidence scores of all shield image points in the TBM cutterhead region within that monitoring frame is recorded as the structural confidence coefficient of that monitoring frame. The structural confidence coefficients of all monitoring frames are then obtained. The absolute value of the difference between the structural confidence coefficients of different monitoring frames is used as a distance metric. Based on this distance metric, hierarchical clustering is performed on all monitoring frames to obtain several clusters. Taking any cluster as an example, the least squares method is used to fit all shield image points in the TBM cutterhead region to the data points of all monitoring frames within that cluster, resulting in a fitted straight line for the TBM cutterhead region within that cluster. The vector formed by this fitted straight line is recorded as the offset vector. The process of obtaining the fitted straight line from the data points is a well-known aspect of the least squares method and will not be elaborated upon in this embodiment.

[0062] Furthermore, regarding the first [section / area] of the tunnel boring machine cutterhead area The shield tunneling image point at the ... The difference in structural confidence between the monitoring frame and other monitoring frames, as well as the difference in offset vectors between adjacent monitoring frames, are used to obtain the first [frame name] of the shield machine cutterhead region. The shield tunneling image point at the ... Frame monitoring ensures consistent positional changes across frames. As an example, the positional changes of the shield tunneling machine cutterhead region can be calculated using the following formula. The shield tunneling image point at the ... Frame monitoring: Consistency of frame position changes

[0063]

[0064] In the formula, The first area representing the cutterhead region of the tunnel boring machine The shield tunneling image point at the ... Frame monitoring ensures consistency in frame position changes; This indicates the number of all clusters in the cutterhead region of the tunnel boring machine; The first area representing the cutterhead region of the tunnel boring machine The offset vectors of each cluster; The first area representing the cutterhead region of the tunnel boring machine The offset vectors of each cluster; This represents the preset denominator hyperparameter, which is preset in this embodiment. This is used to prevent the denominator from being 0; The first area representing the cutterhead region of the tunnel boring machine The shield tunneling image point at the ... Frame monitoring: Frame structural confidence. The first area representing the cutterhead region of the tunnel boring machine The mean of the structural confidence scores of each shield tunneling image point across all monitoring frames; This represents an exponential function with the natural constant as its base. The example uses... The model is used to represent the inverse proportional relationship and for normalization processing. As input to the model, implementers can choose the inverse proportional function and the normalization function according to the actual situation; This indicates taking the absolute value.

[0065] It should be noted that, This indicates the differences in the position changes of all shield tunneling image points across all monitoring frames; where the first position of the shield tunneling machine cutterhead area... The shield tunneling image point at the ... The greater the consistency of the positional changes of the frame monitoring frames, the more obvious the trend of slow settlement or displacement of the tunnel boring machine's operating posture.

[0066] Preferably, in one embodiment of the present invention, the necessity of adjusting the distance error between any two data points in each monitoring frame for each shield tunneling image point is determined based on the difference in the consistency of positional changes between different data points in each monitoring frame. The specific method includes:

[0067] All shield image points in the cutterhead area of ​​the tunnel boring machine at the 1st Frame monitoring frame and the first Data points in the monitoring frames are density-clustered to obtain several clusters; each cluster is recorded as a monitoring cluster; any shield image point in the cutterhead area of ​​the tunnel boring machine is recorded as a target data point in any monitoring frame, and a reference range is preset. In this embodiment, This example is used for illustration; no specific limitations are set in this embodiment. This can be determined based on the specific implementation situation; the target data point is the center, and the radius is... The region is denoted as the neighborhood monitoring area of ​​the target data point; different monitoring clusters contain different monitoring frames; in addition, density clustering algorithm is a well-known technology, and will not be described in detail in this embodiment.

[0068] Furthermore, any data point in the neighborhood monitoring area of ​​the target data point is designated as the monitoring reference data point. Based on the difference in the consistency of positional changes between the target data point and the monitoring reference data point as a whole, the necessary factor for adjusting the distance error between the target data point and the monitoring reference data point is obtained. As an example, the necessary factor for adjusting the distance error between the target data point and the monitoring reference data point can be calculated using the following formula:

[0069]

[0070] In the formula, This represents the necessary factor for adjusting the distance error between the target data point and the monitoring reference data point. This indicates the consistency of the positional changes of the target data points; This indicates the consistency of the location changes of the monitoring reference data points; This represents the mean of the consistency in positional changes among all data points in the monitoring cluster to which the monitoring reference data point belongs; This represents the Euclidean distance between the target data point and the monitoring reference data point; This represents the preset denominator reference hyperparameter, which is preset in this embodiment. This is used to prevent the denominator from being 0.

[0071] It should be noted that the larger the necessary adjustment factor for the distance error between the target data point and the monitoring reference data point, the more significant the structural change between the target data point and the monitoring reference data point, reflecting that the structural change between the target data point and the monitoring reference data point is more likely to be caused by the wear area of ​​the tunnel boring machine or other environmental factors.

[0072] Furthermore, the necessary factors for adjusting the distance error between the target data point and all data points within the neighborhood monitoring area are obtained. All necessary factors for adjusting the distance error are linearly normalized, and each normalized necessary factor for adjusting the distance error is recorded as the distance error adjustment necessity.

[0073] Preferably, in one embodiment of the present invention, the distance metric between different data points is adjusted according to the necessity of adjusting the distance error between different data points to obtain an adaptive distance metric between any two data points. The specific method includes:

[0074] Preset a distance error adjustment necessity threshold In this embodiment, This example is used for illustration; no specific limitations are set in this embodiment. It depends on the specific implementation situation; taking any two shield image points in the cutterhead area of ​​the tunnel boring machine as an example, if the distance error between these two data points is more than necessary to adjust... The data point pair formed by these two data points is denoted as the data point pair to be adjusted; if the distance error between these two data points is less than or equal to the required adjustment... The data point pair formed by these two data points is denoted as the fixed data point pair; all data point pairs to be adjusted and the fixed data point pairs are obtained.

[0075] Furthermore, taking any pair of data points in a data point pair to be adjusted as an example, the product of the Euclidean distance between these two data points and the necessity of adjusting the distance error is denoted as the adaptive distance metric between these two data points; the adaptive distance metric between the two data points in each pair of data points to be adjusted is obtained. The Euclidean distance between the two data points in each fixed data point pair is used as the adaptive distance metric between the two data points in each fixed data point pair.

[0076] Thus, the adaptive distance metric between any two data points is obtained using the above method.

[0077] Step S004: Perform matching detection based on the adaptive distance metric between any two data points.

[0078] Preferably, in one embodiment of the present invention, the matching detection based on an adaptive distance metric between any two data points includes the following specific method:

[0079] An adaptive distance metric between any two different data points is used as the distance metric, and the point cloud data matching result is obtained based on the distance metric. Attitude detection is then performed based on the point cloud data matching result. The process of obtaining the point cloud data matching result based on the distance metric is a well-known part of the PL-ICP (Point-to-line Iterative Closest Point) algorithm, which will not be elaborated in this embodiment. Attitude detection based on the point cloud data matching result is the content disclosed in the paper "Research on Autonomous Correction and Parameters of Shield Tunneling Based on Machine Learning" published by Zhang Jun and Li Maopeng in the journal "Foreign Highway" in 2024.

[0080] This concludes the embodiment.

[0081] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for recognizing the operating posture of a high-precision tunnel boring machine, characterized in that, The method includes the following steps: Point cloud data of the cutterhead area and the working face area of ​​the tunnel boring machine are collected in several monitoring frames. The point cloud data contains multiple shield image points with three-dimensional spatial location information. Each shield image point corresponds to a data point in each monitoring frame. The cutterhead area and the working face area of ​​the tunnel boring machine are both recorded as a tunnel boring machine monitoring area. Based on the overall movement of the same tunnel boring machine image point in different monitoring frames, the structural confidence of each tunnel boring machine image point in each monitoring frame is obtained. Based on the magnitude of positional changes of different shield image points in the same monitoring frame and the structural confidence level, the consistency of positional changes of each shield image point in the cutterhead area of ​​the tunnel boring machine in each monitoring frame is obtained. Based on the structured differences in the consistency of positional changes between different data points of each shield image point in each monitoring frame, the necessity of adjusting the distance error between any two data points of each shield image point in each monitoring frame is obtained. Based on the necessity of adjusting the distance error between different data points, the distance metric between different data points is adjusted to obtain an adaptive distance metric between any two data points. Matching detection is performed based on an adaptive distance metric between any two data points; The method for obtaining the structural confidence of each shield image point in each monitoring frame based on the overall movement of the same shield image point in different monitoring frames includes the following specific methods: Obtain the face region in the first... The weighting coefficient of the tunnel face in the frame monitoring frame and the weighting coefficient of each shield image point in the frame. Frame monitoring: the structural confidence coefficient within the frame; For any shield image point in the cutterhead area of ​​the tunnel boring machine, the shield image point is set to the 1st... The structural confidence coefficient in the frame monitoring frame is related to the tunnel face region in the 1st frame. The product of the face weight coefficients of the monitoring frames is denoted as the shield image point at the [number]th frame. Frame monitoring: structural confidence within frames; The acquisition of the face region in the first The weighting coefficient of the tunnel face in the frame monitoring frame and the weighting coefficient of each shield image point in the frame. The specific methods for monitoring the structural confidence coefficient in a frame include: For any shield image point in the cutterhead area of ​​the tunnel boring machine, the shield image point is set to the 1st... The data points in the frame monitoring frame and the shield tunneling image points at the 1st frame... The Euclidean distance between data points in the frame monitoring frame is denoted as the first distance; the shield image points are then used to determine the distance in the second frame. The data points in the frame monitoring frame and the shield tunneling image points at the 1st frame... The Euclidean distance between data points in the frame monitoring frame is denoted as the second distance; the absolute value of the difference between the first distance and the second distance is denoted as the shield image point at the first [missing value]. The structural confidence factor in the frame monitoring is calculated; the structural confidence factor of the shield tunneling image points in all monitoring frames is linearly normalized, and the normalized structural confidence factor is recorded as the structural confidence coefficient. Record any shield tunneling image point in the tunnel face area as the target tunnel face image point; the reference shield tunneling image point is at the... The method for obtaining the structure confidence factor in the frame monitoring frame is to obtain the target face image point in the [frame number missing]. The structural confidence factor in the frame monitoring frame is used to analyze all shield tunneling image points in the tunnel face region in the 1st frame. The sum of the structural confidence factors in the frame monitoring frame is denoted as the sum of the structural confidence factors in the face region at the 1st frame. The face weight of the monitoring frame; the face weight of the face region in all monitoring frames is linearly normalized, and each normalized face weight is recorded as the face weight coefficient. The method for obtaining the consistency of positional change of each shield image point in the cutterhead region of the tunnel boring machine in each monitoring frame based on the magnitude of positional changes of different shield image points in the same monitoring frame and the structural confidence level includes the following specific methods: The offset vector of each monitoring frame in the cutterhead area of ​​the tunnel boring machine is obtained based on the structural confidence level; According to the first [section / area] of the tunnel boring machine cutterhead area The shield tunneling image point at the first The difference in structural confidence between the monitoring frame and other monitoring frames, as well as the difference in offset vectors between adjacent monitoring frames, are used to obtain the first [frame name] of the shield machine cutterhead region. The shield tunneling image point at the first The specific methods for monitoring the consistency of frame position changes are as follows: In the formula, The first area representing the cutterhead region of the tunnel boring machine The shield tunneling image point at the first Frame monitoring ensures consistency in frame position changes; This indicates the number of all clusters in the cutterhead region of the tunnel boring machine; The first area representing the cutterhead region of the tunnel boring machine The offset vectors of each cluster; The first area representing the cutterhead region of the tunnel boring machine The offset vectors of each cluster; This represents the preset denominator hyperparameter; The first area representing the cutterhead region of the tunnel boring machine The shield tunneling image point at the first Frame monitoring: Frame structural confidence. The first area representing the cutterhead region of the tunnel boring machine The mean of the structural confidence scores of each shield tunneling image point across all monitoring frames; Represents an exponential function with the natural constant as its base; Indicates taking the absolute value; The specific method for obtaining the offset vector of each monitoring frame in the shield machine cutterhead region based on structural confidence is as follows: For any monitoring frame in the cutterhead region of the tunnel boring machine (TBM), the mean of the structural confidence scores of all shield image points in the TBM cutterhead region within the monitoring frame is recorded as the structural confidence coefficient of the monitoring frame. The structural confidence coefficients of all monitoring frames are then obtained. The absolute value of the difference between the structural confidence coefficients of different monitoring frames is used as a distance metric. Based on this distance metric, hierarchical clustering is performed on all monitoring frames to obtain several clusters. For any cluster, the least squares method is used to fit all shield image points in the TBM cutterhead region to the data points of all monitoring frames within the cluster, resulting in a fitted straight line for the TBM cutterhead region within the cluster. The vector formed by the fitted straight line is recorded as the offset vector. The method for determining the necessity of adjusting the distance error between any two data points in each monitoring frame based on the structured differences in the consistency of positional changes of each shield image point among different data points in each monitoring frame includes the following specific methods: All shield image points in the cutterhead area of ​​the tunnel boring machine at the 1st Frame monitoring frame and the first Data points in the monitoring frames are density-clustered to obtain several clusters; each cluster is recorded as a monitoring cluster; any shield image point in the cutterhead area of ​​the tunnel boring machine is recorded as a target data point in any monitoring frame, and a reference range is preset. Centered on the data point, with a radius of... The area is denoted as the neighborhood monitoring area of ​​the target data point; Any data point in the neighborhood monitoring area of ​​the target data point is recorded as the monitoring reference data point; based on the difference in the consistency of positional change between the target data point and the monitoring reference data point as a whole, the necessary factor for adjusting the distance error between the target data point and the monitoring reference data point is obtained; Obtain the necessary distance error adjustment factors between all data points in the neighborhood monitoring area of ​​the target data point and the target data point. Perform linear normalization on all the necessary distance error adjustment factors and record each normalized necessary distance error adjustment factor as the distance error adjustment necessity. The method for obtaining the necessary distance error adjustment factor between the target data point and the monitoring reference data point based on the difference in the consistency of positional changes between the target data point and the overall monitoring cluster to which the monitoring reference data point belongs includes the following specific methods: In the formula, This represents the necessary factor for adjusting the distance error between the target data point and the monitoring reference data point. This indicates the consistency of the positional changes of the target data points; This indicates the consistency of the location changes of the monitoring reference data points; This represents the mean of the consistency in positional changes among all data points in the monitoring cluster to which the monitoring reference data point belongs; This represents the Euclidean distance between the target data point and the monitoring reference data point; This indicates the preset denominator reference hyperparameter; The method for adjusting the distance metric between different data points based on the necessity of adjusting the distance error between different data points to obtain an adaptive distance metric between any two data points includes the following specific methods: Based on the necessity of distance error adjustment, obtain all data point pairs to be adjusted and fixed data point pairs; For any pair of data points to be adjusted, the product of the Euclidean distance between the two data points and the necessity of adjusting the distance error is denoted as the adaptive distance metric between the two data points; the adaptive distance metric between the two data points in each pair of data points to be adjusted is obtained; the Euclidean distance between the two data points in each fixed pair of data points is used as the adaptive distance metric between the two data points in each fixed pair of data points.

2. The method for recognizing the operating posture of a high-precision tunnel boring machine according to claim 1, characterized in that, The specific method for obtaining all pairs of data points to be adjusted and fixed data points based on the necessity of distance error adjustment is as follows: Preset a distance error adjustment necessity threshold For any two shield image points in the cutterhead area of ​​the tunnel boring machine, if the distance error between these two data points is more than [a certain value], then [the following condition applies]. The data point pair formed by these two data points is denoted as the data point pair to be adjusted; if the distance error between these two data points is less than or equal to the required adjustment... The data point pair formed by these two data points is denoted as a fixed data point pair.

3. The method for recognizing the operating posture of a high-precision tunnel boring machine according to claim 1, characterized in that, The specific method for matching and detecting based on the adaptive distance metric between any two data points is as follows: The adaptive distance metric between any two different data points is used as the distance metric, and the point cloud data matching result is obtained by matching based on the distance metric using the PL-ICP algorithm.

Images