JP7829889B2 - Tunnel displacement measurement method and tunnel displacement measurement system - Google Patents

Description

This invention relates to a technique for measuring tunnel crest settlement and displacement of the internal shape during tunnel excavation work.

In tunnel excavation using the NATM (New Austrian Tunneling Method), explosives are used to blast the tunnel face, and after removing the resulting spoil, shoring, primary lining, and rock bolts are installed. This process is repeated at intervals of approximately 1.5 meters to advance the excavation.
In tunnel excavation work, displacement measurements (A-measurements) of the ground and support members are carried out periodically to ensure the safety and quality of the construction. In A-measurements, internal displacement measurements are performed to understand the behavior of the surrounding ground and the deformation mode of the support, to confirm the safety of the construction and the appropriateness of the support, and to consider the timing of lining placement. Conventionally, A-measurements were performed by setting control cross-sections at predetermined intervals and measuring the coordinates of targets set on the top and side walls of the tunnel in the control cross-sections using a distance measuring instrument such as a total station (see, for example, Patent Document 1).

A measurement technique is known for measuring the displacement of tunnel walls using a 3D laser scanner, with the aim of predicting the displacement of the surrounding wall surface based on the displacement at the target installation location. For example, Patent Document 2 discloses a tunnel internal displacement measurement method in which multiple targets are installed on the tunnel wall, the tunnel wall shape including the targets is measured using a 3D laser scanner, and the tunnel displacement is tracked by comparing tunnel cross-sectional shape data at the same location based on the target coordinates (see Patent Document 2). However, installing multiple targets on the tunnel wall presents problems such as complicated installation work and high target costs.

Therefore, a technique is known that allows for the measurement of tunnel internal displacement without using a target by utilizing scan data from a 3D laser scanner to determine internal displacement at countless points (see Patent Document 3). According to this method, since it does not require the installation of a target on the measurement cross-section, the measurement of internal displacement can be performed simply and efficiently. However, the tunnel internal displacement measurement method described in Patent Document 3 has the problem that high-speed processing is difficult because it determines internal displacement at countless points measured by a 3D laser scanner.

Japanese Patent Publication No. 2004-170164 Japanese Patent Publication No. 2014-002027 Japanese Patent Publication No. 2012-058167

In view of these circumstances, the present invention aims to provide a method and system that can measure the displacement inside a tunnel at high speed.

As a result of diligent research, the inventors have discovered that, rather than using a 3D laser scanner to perform surface measurement, applying point cloud data obtained from a 3D laser scanner to a point measurement technique enables simple and high-speed displacement measurement.

In other words, the tunnel displacement measurement method of the present invention is a method for measuring the displacement of the tunnel interior over time by setting a target using point cloud data of the tunnel interior acquired by a three-dimensional laser scanner, and consists of 1) an initial measurement step and 2) a displacement measurement step, wherein 1) the initial measurement step includes acquiring surface feature data which is the surface feature of the target, acquiring point cloud data of the tunnel interior including the target as first point cloud data after acquiring the surface feature data, and extracting surface feature data from the first point cloud data to acquire first position information of the target, and 2) the displacement measurement step performed after a predetermined time has elapsed includes setting a search area with a larger area than the target that includes a position identified based on the first position information, acquiring point cloud data of the search area, extracting the target from the point cloud data of the search area based on surface feature data to acquire second position information, and measuring the displacement of the tunnel interior based on the difference between the first position information and the second position information , and the target is a physical target member installed on the wall surface of the tunnel interior .
Alternatively , the following steps may be taken: 1) The initial measurement step includes acquiring point cloud data of the tunnel's inner surface as first point cloud data, extracting a predetermined area of the tunnel's inner surface from the first point cloud data as a target, extracting surface feature data which is the surface feature of the target, and registering the first position information and surface feature data of the target, and is performed after a predetermined time has elapsed. 2) The displacement measurement step may include setting a search area larger in area than the target, including the position identified based on the first position information, acquiring point cloud data of the search area, extracting the target from the point cloud data of the search area based on the surface feature data to acquire second position information, and measuring the displacement of the tunnel's interior based on the difference between the first position information and the second position information.

Thus, by limiting the comparison to the target range rather than using all of the acquired point cloud data for displacement measurement, rapid processing can be achieved.

1) The initial measurement step and 2) the displacement measurement step each comprise the following steps:

1) Initial measurement step 1-A) Scanner coordinate acquisition step to obtain the position coordinates of the 3D laser scanner.
1-B) A first point cloud data acquisition step in which point cloud data of the sprayed concrete surface inside the tunnel is acquired as the first point cloud data.
1-C) A pseudo-target setting step in which at least three regions are extracted from the first point cloud data and set as pseudo-targets.
1-D) A first surface feature data extraction step in which surface features based on the plane of the pseudo-target are extracted as first surface feature data.
1-E) A first registration step in which the position coordinates of a pseudo-target, which is first position information, and first surface feature data are registered in association with the position coordinates of a three-dimensional laser scanner.

2) Displacement measurement step 2-A) Second point cloud data acquisition step, in which point cloud data of the sprayed concrete surface is acquired as second point cloud data.
2-B) A search area setting step in which the search area is set from the second point cloud data based on the position coordinates of the pseudo-target.
2-C) A second surface feature data extraction step, in which surface features based on the plane of the pseudo-target are extracted from the search area as second surface feature data.
2-D) A pseudo-target identification step in which the position coordinates of a pseudo-target, which is second position information, are identified based on the similarity between the surface features of the first surface feature data and the partial surface features of the second surface feature data.
2-E) A second registration step in which the position coordinates of the identified pseudo-target are registered in association with the position coordinates of the 3D laser scanner.
2-F) A displacement measurement step that measures the displacement of the space inside the tunnel based on the changes in the position coordinates of the pseudo-target set in the initial measurement step and the pseudo-target identified in the displacement measurement step.

By using a simulated target, the displacement of the tunnel's internal shape can be measured inexpensively, simply, and with high accuracy, without the need for existing targets (prisms).
Generally, for top-surface settlement and internal displacement measurements, targets are set every 10 to 30 meters, and measurements are taken at a frequency of twice a day in some cases, and at least once a week in others. When setting up new dummy targets, 1) an initial measurement step is performed at the locations where the new dummy targets are set up, and 2) a displacement measurement step is performed at locations where the initial measurement has already been completed. By repeatedly performing measurements in this manner, the shape displacement of the tunnel interior can be accurately measured.

1-A) The scanner coordinate acquisition step can utilize a wide range of methods for acquiring the position coordinates of a 3D laser scanner within the tunnel. For example, the position coordinates of the 3D laser scanner can be acquired using a laser scanner target attached to a point whose position coordinates are known. Alternatively, if the 3D laser scanner has surveying capabilities, its surveying functions can be used to acquire absolute coordinates or tunnel coordinates. The acquired position coordinates may be absolute coordinates (global coordinates) or tunnel coordinates (local coordinates).

1-B) Instead of treating all of the point cloud data acquired in the first point cloud data acquisition step as countless targets, 1-C) by extracting three or more regions from the point cloud data in the pseudo-target setting step and setting them as pseudo-targets, the locations where displacement is measured can be limited, and the processing speed of displacement measurement can be improved. In order to improve processing speed while maintaining measurement accuracy, it is preferable that the extracted regions be 3 to 5 locations per measurement cross section.
Because the surface of the sprayed concrete inside the tunnel has fine irregularities, it is possible to extract unique regions from the point cloud data. Here, "the sprayed concrete surface inside the tunnel" refers to the areas where concrete was sprayed onto the tunnel walls and support structures after blasting. The term "sprayed concrete surface" is not strictly defined and is intended to include areas where the concrete spraying was insufficient. Therefore, rock bolts fixed after the concrete spraying are also included.
A LiDAR (Light Detection And Ranging) scanner is preferably used as the 3D laser scanner. Each region that serves as the pseudo-target is a region cut out in a predetermined shape, such as a polygon such as a square, or a circle or ellipse, for example, a roughly square shape with dimensions of 150 mm on each side.

1-E) The position coordinates of the pseudo-target registered in the first registration step are preferably the first point cloud data included in the pseudo-target, but may also be the centroid, normal, or plane data as described later.

1 -C) The pseudo-target setting step preferably comprises a first extraction step of extracting a region, a first centroid calculation step of calculating the centroid of the pseudo-target based on the point cloud data included in the extracted region, a first normal calculation step of calculating the normal of the pseudo-target based on the point cloud data and the centroid, and a first planarization step of converting the pseudo-target into planar data based on the centroid and the normal. The normal can be calculated using known methods for processing point cloud data, for example, as a line segment passing through the centroid coordinates from the eigenvectors of the covariance matrix of the point cloud data. By calculating the centroid and normal and performing planarization processing instead of directly comparing the point cloud data, feature data extraction becomes easier, enabling high-speed and high-precision identification of pseudo-targets and displacement measurement.

1 -D) The first step of extracting surface feature data preferably involves calculating cross-sectional data from the relative positions of each point in the normal direction for at least one cross-section in the planar data. Calculating cross-sectional data facilitates comparison of surface feature data. Preferably, there are multiple cross-sectional data to be calculated, and more preferably, there are multiple cross-sectional data with different orientations.

2 -B) The search area setting step preferably includes: a second extraction step of extracting a search area from second point cloud data based on the position coordinates of a pseudo-target set in the initial measurement step; a second centroid calculation step of calculating the centroid of the search area based on the second point cloud data included in the extracted search area; a second normal calculation step of calculating the normal of the search area based on the second point cloud data and the centroid; and a second planarization step of converting the search area into planar data based on the centroid and the normal. Including such steps makes it easier to identify the pseudo-target.

2 -C) The second surface feature data extraction step preferably involves calculating cross-sectional data from the relative positions of each point in the normal direction for at least one cross-section in the planar data. This makes it easier to identify pseudo-targets by comparing the first surface feature data with the second surface feature data.

2 -D) The pseudo-target identification step preferably involves comparing the cross-sectional data calculated in 1-D) the first surface feature data extraction step with the cross-sectional data calculated in 2-C) the second surface feature data extraction step, and identifying a pseudo-target based on the similarity of the relative positions of each point in the normal direction.

The second cutting step may also involve 1) cutting out an area with an area approximately four times or less the size of the pseudo-target, centered on the position coordinates of the pseudo-target set in the initial measurement step.
In the second extraction step, the search area is extracted from the second point cloud data based on the position coordinates of the pseudo-target registered in the first registration step. However, since the pseudo-target may move or change shape after the initial measurement, it is necessary to extract the area with a predetermined margin. Therefore, by setting the length and width of the pseudo-target to approximately twice its original size and extracting an area with an area of approximately four times or less its original size, it is possible to prevent the pseudo-target from becoming unidentifiable due to its movement or change in shape.

2 -F) In the displacement measurement step, the change in the position coordinates of the pseudo-target is preferably the change in the position coordinates of the first point cloud data and the second point cloud data included in the pseudo-target. The change in position coordinates refers to the change in the three-dimensional position coordinates of each point. By comparing only the point cloud data included in the pseudo-target, rather than all of the acquired point cloud data, high-speed and high-precision displacement measurement can be achieved. Alternatively, the displacement of the tunnel's internal shape may be measured based on changes in the center of gravity, normals, and planar data, or the displacement may be measured based on changes in the first and second surface feature data.

The coordinates of the simulated target can be absolute coordinates or tunnel coordinates.

The simulated targets are preferably set in at least one of the following areas in tunnel excavation work using the NATM (New Austrian Tunneling Method): areas where A, B, and C lines can be measured, and areas where D line can be measured, based on the location or shape of the tunnel interior, support structures, or rock bolts. By setting the simulated targets in such areas, A measurement can be easily performed. The setting may be done automatically based on the location or shape of the tunnel interior, support structures, or rock bolts, or it may be set manually. In addition, even outside of these areas, for example, areas with significant irregularities on the tunnel interior surface may be detected and set automatically or by visual inspection by a worker.

The target member has protrusions and possesses the uneven shape characteristic of the tunnel's interior. The target is preferably a fixing plate for a rock bolt installed on the tunnel's inner surface. The target is also preferably a steel support structure installed on the tunnel's inner surface. This eliminates the need for a dedicated target for measurement. Furthermore, by having protrusions at multiple locations in the vertical direction, the characteristic uneven shape of the target's surface allows for accurate determination of horizontal and vertical displacements as changes in position coordinates. This enables inexpensive, simple, and highly accurate measurement of the tunnel's interior shape displacement. In other words, the target member can be any metal object installed on the tunnel's interior wall with uneven characteristics on the portion exposed to the inner surface .

Three-dimensional laser scanners are installed along each of the walls on both sides in the tunnel width direction. Each three-dimensional laser scanner acquires point cloud data from the wall on which it is installed and the wall on the opposite side. Topographic feature data, which represents the surface characteristics of the target, can be obtained from the imaging results obtained by imaging the target with an imaging device.

The tunnel displacement measurement system of the present invention is a system comprising: an apparatus that measures the displacement of the tunnel interior over time by setting a target using point cloud data of the tunnel interior acquired by a three-dimensional laser scanner ; and a physical target member that is installed as a target on the wall surface of the tunnel interior. The apparatus consists of an initial measurement unit and a displacement measurement unit. The initial measurement unit acquires surface feature data, which is the surface feature of the target ; after acquiring the surface feature data, acquires point cloud data of the tunnel interior including the target as first point cloud data; extracts surface feature data from the first point cloud data to acquire first position information of the target. The displacement measurement unit sets a search area that includes a position identified based on the first position information and has a larger area than the target; acquires point cloud data of the search area; extracts the target from the point cloud data of the search area based on surface feature data to acquire second position information; and measures the displacement of the tunnel interior based on the difference between the first position information and the second position information .

Alternatively, the device may be characterized in that the initial measurement unit acquires point cloud data of the tunnel's inner surface as first point cloud data, extracts a predetermined area of the tunnel's inner surface from the first point cloud data as a target, extracts surface feature data which is the surface feature of the target, and registers the first position information and surface feature data of the target; the displacement measurement unit sets a search area that includes a position identified based on the first position information and has a larger area than the target, acquires point cloud data of the search area, extracts a target from the point cloud data of the search area based on the surface feature data to acquire second position information, and measures the displacement of the tunnel's interior based on the difference between the first position information and the second position information.
The initial measurement unit and the displacement measurement unit each include the following means.

1) Initial measurement unit 1-a) Scanner coordinate acquisition means for acquiring the position coordinates of a 3D laser scanner.
1-b) A first point cloud data acquisition means for acquiring point cloud data of the sprayed concrete surface inside the tunnel as first point cloud data.
1-c) A pseudo-target setting means for extracting at least three regions from the first point cloud data and setting them as pseudo-targets.
1-d) A first surface feature data extraction means for extracting surface features based on the plane of a pseudo-target as first surface feature data.
1-e) A first registration means for registering the position coordinates of a pseudo-target, which is first position information, and first surface feature data, in association with the position coordinates of a three-dimensional laser scanner.
2) Displacement measurement unit 2-a) Second point cloud data acquisition means for acquiring point cloud data of the sprayed concrete surface as second point cloud data.
2-b) Search area setting means for setting a search area from second point cloud data based on the position coordinates of a pseudo-target.
2-c) A second surface feature data extraction means for extracting surface features based on the plane of a pseudo-target from the search area as second surface feature data.
2-d) A pseudo-target identification means that identifies the position coordinates of a pseudo-target, which is second position information, based on the similarity between the surface features of the first surface feature data and the partial surface features in the second surface feature data.
2-e) A second registration means for registering the position coordinates of an identified pseudo-target in association with the position coordinates of a three-dimensional laser scanner.
2-f) Displacement measuring means for measuring the displacement of the space inside a tunnel based on the change in the position coordinates of a pseudo-target set in the initial measurement means and the pseudo-target identified in the displacement measuring means.

The tunnel displacement measurement system of the present invention comprises a tunnel displacement measurement device and a physical target member installed as a target on the wall surface inside the tunnel . The target member has protrusions and has uneven characteristics in the shape of the inside of the tunnel. Alternatively, the target member is a metal fitting installed on the wall surface inside the tunnel and has uneven characteristics in the portion exposed to the inner surface. If the tunnel is constructed using the NATM method, the target is a rock bolt anchoring plate or a steel support. If the tunnel is constructed using the shield tunneling method, the target is a gripping fitting or joint fitting provided on the inner surface of the tunnel segment piece.

The tunnel displacement measurement method and system of the present invention have the effect of being able to measure the displacement inside a tunnel at high speed.

Functional block diagram of the tunnel displacement measuring device Schematic diagram of the tunnel displacement measurement device. Schematic flowchart of tunnel displacement measurement method Diagram (1) illustrating the initial measurement step and displacement measurement step. Diagram illustrating the initial measurement step and displacement measurement step (2) Initial measurement flow chart Image diagram of the placement of simulated targets Pseudo-target setting flowchart Diagram illustrating the steps for setting a simulated target. Image diagram of simulated target and feature data Feature data extraction diagram Displacement Measurement Flowchart Search Area Setting Flowchart Image diagram of the search area setting Diagram illustrating the steps for identifying a pseudo-target (1) Diagram illustrating the steps for identifying a pseudo-target (2) Schematic diagram of the tunnel displacement measurement system. Image of a rock bolt plate Target image Schematic diagram showing the relationship between the number of protrusions and measurement accuracy. Image of a protrusion Schematic diagram of the tunnel displacement measurement system. Conceptual diagram of a protrusion on a steel support structure.

The following describes in detail an example of an embodiment of the present invention with reference to the drawings. However, the scope of the present invention is not limited to the following embodiments and illustrated examples, and numerous modifications and variations are possible.

Figure 1 shows a functional block diagram of the tunnel displacement measuring device. Figure 2 shows an image diagram of the configuration of the tunnel displacement measuring device. As shown in Figure 1, the tunnel displacement measuring device 1 consists of an initial measurement unit 2 and a displacement measuring unit 3. The initial measurement unit 2 includes a scanner coordinate acquisition means 21, a first point cloud data acquisition means 22, a pseudo-target setting means 23, a first unevenness feature data extraction means 24, and a first registration means 26. The displacement measuring unit 3 includes a second point cloud data acquisition means 32, a search area setting means 33, a second unevenness feature data extraction means 34, a pseudo-target identification means 35, a second registration means 36, and a displacement measuring means 37.
Furthermore, as shown in Figure 2, the tunnel displacement measuring device 1 consists of a LiDAR scanner 4, a computer 5, and scanner targets (11a to 11c), with the LiDAR scanner 4 and the computer 5 connected by wire or wireless. The scanner targets (11a to 11c) are attached to points in the interior of the tunnel 7 whose position coordinates are known. The scanner targets (11a to 11c) can be installed, for example, on the wall surface 72 of the tunnel 7.

The scanner coordinate acquisition means 21 acquires the position coordinates of the 3D laser scanner. A LiDAR scanner 4 is used as the 3D laser scanner. In this embodiment, the scanner coordinate acquisition means 21 acquires the absolute coordinates of the LiDAR scanner 4 using scanner targets (11a to 11c). The first point cloud data acquisition means 22 acquires point cloud data of the sprayed concrete surface (inner surface of the tunnel) inside the tunnel using the 3D laser scanner. The pseudo-target setting means 23 extracts at least three predetermined regions from the first point cloud data and sets them as pseudo-targets (targets). The first surface feature data extraction means 24 extracts first surface feature data from the pseudo-targets. The first registration means 26 registers the position coordinates (first position information) of the pseudo-targets and the first surface feature data in association with the position coordinates of the 3D laser scanner.

The second point cloud data acquisition means 32 acquires point cloud data of the sprayed concrete surface inside the tunnel using a three-dimensional laser scanner. The search area setting means 33 sets the search area of a pseudo-target (target) from the second point cloud data. The search area includes a location identified based on the position coordinates of the pseudo-target (first position information) and is a larger area than the pseudo-target. The second surface feature data extraction means 34 extracts second surface feature data from the point cloud data of the search area. The pseudo-target identification means 35 extracts a pseudo-target based on the similarity between the first surface feature data and the second surface feature data, identifies its position coordinates (second position information), and acquires it. The second registration means 36 registers the identified position coordinates of the pseudo-target in association with the position coordinates of the three-dimensional laser scanner. The displacement measuring means 37 measures the displacement of the tunnel's internal shape based on the change in the position coordinates (the difference between the first and second position information) of the pseudo-target set in the initial measurement unit 2 and the pseudo-target identified in the displacement measuring unit 3.

In this embodiment, the LiDAR scanner 4 comprises a first point cloud data acquisition means 22 and a second point cloud data acquisition means 32, and the computer 5 comprises a pseudo-target setting means 23, a first surface feature data extraction means 24, a first registration means 26, a second point cloud data acquisition means 32, a search area setting means 33, a second surface feature data extraction means 34, a pseudo-target identification means 35, a second registration means 36, and a displacement measurement means 37.

Figure 3 shows a schematic flowchart of the tunnel displacement measurement method. As shown in Figure 3, the tunnel displacement measurement method in this embodiment first performs an initial measurement (Step S01: Initial Measurement Step), and then performs a displacement measurement (Step S02: Displacement Measurement Step). Displacement measurements are performed at a frequency of once a week to twice a day. Specifically, this is determined based on the distance between the measurement position and the tunnel face, and the displacement velocity of the internal displacement. Displacement measurements are performed repeatedly, and once all displacement measurements are completed (Step S03), the displacement measurement is terminated.
Figures 4 and 5 illustrate the initial measurement step and the displacement measurement step. As shown in Figure 4 or Figure 5, when the initial measurement step (step S01) is performed for range _R2 , if the initial measurement step (step S01) has already been completed for range _R1 as shown in Figure 5, the initial measurement step (step S01) is performed for range _R2 , and the displacement measurement step (step S02) is performed for range _R1 . After a predetermined time has elapsed, as the tunnel excavation work progresses, if a pseudo-target is to be set up for a new range _R3 , the initial measurement step (step S01) is performed for range _R3 , and the displacement measurement step (step S02) is performed for the ranges ( _R1 , _R2 ) where the initial measurement has already been completed. In this way, displacement measurements can be repeatedly performed while increasing the number of pseudo-targets.

(Regarding the initial measurement step)
Figure 6 shows the initial measurement flow diagram. As shown in Figure 6, first, the position coordinates of the 3D laser scanner are obtained (step S11: scanner coordinate acquisition step). As mentioned above, in this embodiment, the absolute coordinates of the LiDAR scanner 4 are obtained using scanner targets (11a to 11c).
A 3D laser scanner is used to acquire point cloud data of the sprayed concrete surface inside the tunnel (Step S12: First point cloud data acquisition step). In this embodiment, a LiDAR scanner 4 is used to acquire point cloud data relating to the tunnel face 71 and wall surface 72 in front of the LiDAR scanner 4. Concrete is sprayed onto the wall surface 72. The range of point cloud data acquisition here is not limited to the range set as a pseudo-target; for example, as shown in Figure 5, if it is possible to acquire point cloud data for range R, point cloud data is acquired for the entire range R.

At least three regions are extracted from the point cloud data and set as pseudo-targets (Step S13: Pseudo-target setting step). In this embodiment, as shown in Figure 2, five regions are extracted to include the measurement cross-section 70 and set as pseudo-targets (6a to 6e). Figure 7 shows an image diagram of the pseudo-target arrangement. Regarding the placement of the pseudo-targets, as shown in Figure 7, pseudo-targets 6a and 6e are placed at positions where the D measurement line can be measured. Furthermore, pseudo-targets 6b and 6c are placed at positions where the A measurement line can be measured, pseudo-targets 6c and 6d are placed at positions where the B measurement line can be measured, and pseudo-targets 6b and 6d are placed at positions where the C measurement line can be measured.

Here, we will explain the method for setting a pseudo-target. Figure 8 shows a flowchart for setting a pseudo-target. Figure 9 is an explanatory diagram of the pseudo-target setting steps, where (1) shows region extraction, (2) shows centroid calculation, (3) shows normal vector calculation, and (4) shows the image of planarization. As shown in Figure 8, in the pseudo-target setting step (step S13), first, a region is extracted from the first point cloud data (step S131: first extraction step). Figure 9(1) shows the point cloud 8 in the extracted region. Next, as shown in Figure 9(2), the centroid G of the pseudo-target is calculated based on the point cloud data included in the extracted region (step S132: first centroid calculation step). Based on the point cloud data and centroid, the normal vector of the pseudo-target is calculated (step S133: first normal vector calculation step). The normal vector can be calculated using known methods for processing point cloud data, for example, as a line segment passing through the centroid coordinates from the eigenvectors of the covariance matrix of the point cloud data. Figure 9(3) illustrates the normal direction N. Finally, based on the centroid and normal, the pseudo-target is converted into planar data (Step S134: First planarization step). Figure 9(4) illustrates the planarized plane P.

Next, as shown in Figure 6, first surface feature data is extracted from the pseudo-target (Step S14: First surface feature data extraction step). The first surface feature data extraction step calculates cross-sectional data from the relative positions of each point with respect to the normal direction N for at least one cross-section in the planar data.
Figure 10 is an illustrative diagram of a pseudo-target and surface feature data, where (1) shows the state in which a pseudo-target is set and (2) shows the state in which surface feature data is extracted. Figure 11 is an illustrative diagram of the extraction of surface feature data. As shown in Figure 10(1), after a pseudo-target 6 is set for the point cloud 80 acquired with respect to the wall surface 72, multiple cross-sectional data are calculated as shown in Figure 11. Specifically, in this embodiment, cross-sectional data relating to the vertical cross-section ( _C1 to _C3 ) and the horizontal cross-section ( _C4 to _C6 ) of the pseudo-target 6 are calculated. The vertical direction corresponds to, for example, the circumferential direction of the tunnel cross-section perpendicular to the tunnel axis direction, and the horizontal direction corresponds to, for example, the tunnel axis direction. In addition, unlike this embodiment, more cross-sectional data may be calculated, or, for example, cross-sectional data in the diagonal direction may be calculated. The diagonal direction is a direction inclined with respect to the vertical and horizontal directions.

The position coordinates of the pseudo-target and the first surface feature data are registered in association with the position coordinates of the 3D laser scanner (Step S15: First registration step). In this embodiment, the absolute coordinates of the point cloud 8 within the pseudo-targets (6a-6e), along with the centroid, normal, and plane data, are registered based on the absolute coordinates of the LiDAR scanner 4. However, it is also acceptable to register only the absolute coordinates of the point cloud 8 within the pseudo-targets (6a-6e).

(Regarding the displacement measurement step)
Figure 12 shows a displacement measurement flowchart. As shown in Figure 12, a 3D laser scanner is used to acquire point cloud data of the sprayed concrete surface inside the tunnel (Step S21: Second point cloud data acquisition step). The method for acquiring point cloud data is the same as in the first point cloud data acquisition step (Step S12).

Next, the search area for the pseudo-target is set from the second point cloud data (Step S22: Search Area Setting Step). Figure 13 shows a flowchart for setting the search area. Figure 14 is an image diagram of the setting of the search area 60, and Figures 15 and 16 are explanatory diagrams of the pseudo-target identification step.
As shown in Figures 13 and 14(1), first, based on the position coordinates 60a of the pseudo-target 6 set in the initial measurement step (step S01), the search area 60 is extracted from the second point cloud data 81 (step S221: second extraction step).
Next, second point cloud data included in the extracted search area 60 is acquired, and the centroid of the search area 60 is calculated based on this point cloud data (Step S222: Second centroid calculation step). Based on the second point cloud data and the centroid, the normal of the search area 60 is calculated (Step S223: Second normal calculation step). Based on the centroid and the normal, the search area 60 is converted into planar data (Step S224: Second planarization step). This process is the same as the pseudo-target setting step shown in Figure 9. Figure 14(1) shows the state in which the search area 60 has been converted into planar data.

Second surface feature data is extracted from the search area 60 (Step S23: Second surface feature data extraction step). Figure 14(2) shows an image of the second surface feature data, which includes cross-sectional data relating to the vertical cross-section ( _C10 to _C30 ) and the horizontal cross-section ( _C40 to _C60 ) of the search area 60.
As shown in Figure 15(1), the pseudo-target 6 is extracted and its position coordinates are identified based on the similarity between the first surface feature data of the pseudo-target 6 and the second surface feature data of the search area 60 (Step S24: Pseudo-target identification step). In this embodiment, as shown in Figure 15(2), the similarity is determined to be high between sections _C20 and _C2 , and between sections _C50 and _C5 . Then, as shown in Figure 16(1), the pseudo-target 6 is extracted from the position coordinates of the areas with high similarity ( _A1 , _A2 ), and its position coordinates are identified and obtained as shown in Figure 16(2).

The position coordinates of the identified pseudo-target are registered in association with the position coordinates of the 3D laser scanner (Step S25: Second registration step). The method for registering the coordinates is the same as in the first registration step (Step S15).
Based on the changes in the position coordinates of the pseudo-target set in the initial measurement step and the pseudo-target identified in the displacement measurement step, the displacement of the tunnel's internal shape is measured (Step S26: Displacement Measurement Step). In this embodiment, the point cloud data 8 of the pseudo-target 6 shown in Figure 10(2) and the point cloud data 8a of the pseudo-target 6 shown in Figure 16(2) are compared to measure the displacement of the tunnel's internal shape.

In Example 1, point cloud data (a simulated target) is used as the target, but the target may be a physical target element installed in the tunnel. However, to eliminate the effort of installing the target, it is preferable to use a target that is not dedicated solely to measurement, but is shared with other elements normally installed in the tunnel.

One example is the tunnel displacement measurement system 10 shown in Figure 17, which includes a tunnel displacement measurement device 1a and targets 100a to 100e installed on the wall surface 72 of the tunnel 7. The targets 100a to 100e are each positioned in the same locations as the pseudo-targets 6a to 6e in each range _R1 , _R2 , _R3 , ... (see Figure 5) in the tunnel axial direction.

Furthermore, the LiDAR scanners 4 are installed along each of the wall surfaces 72 on both sides in the tunnel width direction, and each LiDAR scanner 4 is configured to acquire point cloud data of the wall surface 72 opposite to the wall surface 72 on which the LiDAR scanner 4 is installed. The tunnel width direction is perpendicular to the tunnel axis direction in the plane. In Figure 17, the tunnel axis direction is indicated by symbol A, and the tunnel width direction is indicated by symbol B.

Figure 18 shows the target 100a. In this example, a plate for anchoring the rock bolt 200 is used as the target 100a. The rock bolt 200 is driven from the wall surface 72 of the tunnel 7 into the ground behind it, and the plate of the rock bolt 200, i.e., the target 100a, is installed on the sprayed concrete of the wall surface 72. The end of the rock bolt 200 protrudes from the outer surface of the target 100a, and a nut 210 is tightened onto this protruding portion.

The target 100a is a plate-shaped metal object with projections 110 on its outer surface, giving it an uneven surface. The projections 110 are continuous horizontal ridges, arranged in multiple parallel rows vertically within the target 100a. More specifically, two rows of projections 110 are arranged on the upper part of the target 100a, and two rows on the lower part. A through-hole (not shown) for passing the end of the lock bolt 200 is provided between the two upper and two lower projections 110. The cross-section of the projections 110 is rectangular.

The other targets 100b to 100e have similar shapes; only their orientation during installation differs depending on the position of the targets 100b to 100e. For example, the target 100c at the top of the tunnel 7 shown in Figure 17 is positioned with its plate surface horizontal, and the protrusions 110 are arranged in multiple rows with spacing in the tunnel width direction B.

The function of the tunnel displacement measuring device 1a is substantially the same as that of the tunnel displacement measuring device 1 described in Figure 1. However, the initial measurement unit 2 replaces the pseudo-target setting means 23 and the first surface feature data extraction means 24 with a surface feature data acquisition means that receives input of imaging results obtained in advance from a separate imaging device (not shown) such as a 3D laser scanner, and acquires surface feature data, which is the surface feature of targets 100a to 100e, from the imaging results. It also includes a position information acquisition means that extracts surface feature data from point cloud data (first point cloud data) of the tunnel interior including targets 100a to 100e and acquires the position coordinates (first position information) of targets 100a to 100e. The first registration means 26 registers the position information and surface feature data of targets 100a to 100e.

The displacement measurement unit 3 replaces the pseudo-target identification means 35 with a target identification means that extracts targets 100a to 100e from the point cloud data of the search area 60 and identifies and acquires their position coordinates (second position information). The search area 60 includes the positions identified based on the position coordinates (first position information) of targets 100a to 100e acquired by the position information acquisition means, and is a larger area than targets 100a to 100e. The second registration means 36 and the displacement measurement means 37 perform similar processing on the area of targets 100a to 100e, instead of the pseudo-targets 6a to 6e.

The initial measurement flow in this embodiment is basically the same as that described in Figure 6. However, in this embodiment, the computer 5 receives input of imaging results obtained in advance from a separate imaging device (not shown) that has imaged targets 100a to 100e, and acquires surface feature data of targets 100a to 100e from these imaging results. Then, instead of performing the pseudo-target setting step (step S13) and the first surface feature data extraction step (step S14) in Figure 6, surface feature data is extracted from the point cloud data of the tunnel interior including targets 100a to 100e acquired in the first point cloud data acquisition step (step S12) to obtain the position coordinates of targets 100a to 100e. Subsequently, in the first registration step (step S15), the position coordinates and surface feature data of targets 100a to 100e are registered.

Since targets 100a to 100e have protrusions 110, in the surface feature data, the surface features due to the protrusions 110 appear in the vertical cross-sectional data (see reference numerals _C1 to _C3 in Figure 11), and the height of the protrusions and the depth of the recesses in the surface features appear in the horizontal cross-sectional data (see reference numerals _C4 to _C6 in Figure 11). It is desirable that the cross-section used to calculate the cross-sectional data be a position that avoids the ends of the lock bolts 200 and the nuts 210.

The displacement measurement flow in this embodiment is basically the same as the flow in Embodiment 1 described in Figure 12. However, in this embodiment, instead of identifying the position coordinates of pseudo-targets 6a to 6e in the pseudo-target identification step (step S24) in Figure 12, the position coordinates (second position information) of the target area 100a to 100e are identified. The target identification step extracts areas with similar surface features to targets 100a to 100e from the search area 60 based on the similarity between the surface feature data of targets 100a to 100e and the surface feature data of the search area 60, and identifies and obtains their position coordinates. The target identification method may be the same template matching method as in Embodiment 1, but is not limited to this; any method that searches for surface features due to the protrusions 110 of targets 100a to 100e within the search area 60 is acceptable, and various known matching methods can be applied.

In this embodiment, the placement of protrusions 110 on the outer surfaces of targets 100a to 100e allows for accurate positioning of these targets. Specifically, as shown in Figure 19(a), if the surface of target 100a were smooth, there would be no characteristic features in the shape of the tunnel cavity including target 100a. However, as in this embodiment, when protrusions 110 are formed on target 100a, characteristic irregularities appear in the shape of the tunnel cavity including target 100a, as shown by the solid line portion in Figure 19(b).

In this embodiment, depending on the location of these uneven features, even when horizontal or vertical displacement occurs in the tunnel width direction B of the target 100a, the horizontal and vertical positions of the target 100a along the tunnel width direction B after displacement can be accurately determined. In particular, if the surface of the target 100a is smooth, as shown by arrow a, even if vertical displacement occurs in the target 100a (the target 100a after displacement is shown by a dashed line), the shape of the tunnel interior hardly changes, making it difficult to determine the vertical displacement of the target 100a. In contrast, in this embodiment, as shown by Figure 19(b), when vertical displacement occurs in the target 100a, the position of the protrusion 110 clearly changes, thereby easily determining the vertical displacement of the target 100a.

In the second registration step (step S25) and the displacement measurement step (step S26), processing is performed on the region of targets 100a to 100e instead of the pseudo-targets 6a to 6e. In the displacement measurement step (step S26), the vertical displacement of targets 100a to 100e, i.e., the vertical displacement of the tunnel interior, and the horizontal displacement of targets 100a to 100e in the tunnel width direction B, i.e., the horizontal displacement of the tunnel interior in the tunnel width direction B, can be determined by the change in the position coordinates of targets 100a to 100e (the difference between the first position information and the second position information).

Thus, in this embodiment, by using the plates of the rock bolts 200 as targets 100a to 100e, there is no need to install a dedicated measurement target in the tunnel 7, simplifying the measurement process. Furthermore, while some dedicated measurement targets use prisms or mirrors, which may be damaged during blasting of the tunnel cutting edge, targets 100a to 100e (plates of the rock bolts 200) in this embodiment do not pose such a risk. Therefore, it is possible to measure the displacement immediately after blasting.

Furthermore, since target 100a has protrusions 110 on its outer surface, the horizontal and vertical displacements of the area of target 100a can be measured accurately due to the surface irregularities caused by the protrusions 110. This is also true for the other targets 100b to 100e.

Furthermore, since the projections 110 of target 100a are arranged in multiple vertical rows, the vertical displacement of the area of target 100a can be measured with greater accuracy. This is also true for targets 100b, 100d, and 100e. For target 100c at the top of tunnel 7, the projections 110, arranged in multiple rows at intervals along the tunnel width B, allow for accurate measurement of the horizontal displacement in the tunnel width direction B.

The shapes of targets 100a to 100e are not limited to those described above. For example, in the example shown in Figure 18, targets 100a to 100e are provided with four protrusions 110, but the number of protrusions 110 is not particularly limited. However, in terms of accuracy, it is desirable to provide three or more protrusions 110. Figure 20 briefly shows the results of the inventor's study on the relationship between the number of protrusions 110 and measurement accuracy, showing that accuracy improves when there are three or more protrusions 110.

Furthermore, while the projections 110 of target 100a are continuous horizontal ridges, the shape of the projections 110 is not limited to this. For example, as shown in target 100a' in Figure 21, which is a modified example of target 100a in Figure 18, the projections 110' may be discretely arranged with intervals between them in the up, down, left, and right directions. However, having the projections 110 continuous in the horizontal direction makes it easier to set up a large number of vertical cross-sectional data (see reference numerals _C1 to _C3 in Figure 11) where the unevenness features caused by the projections 110 appear, thus contributing to improved measurement accuracy.

Furthermore, the protrusion height of the projection 110 can be freely set, taking into consideration measurement accuracy and handling. For example, increasing the protrusion height of the projection 110 improves measurement accuracy by clearly showing the convex portion in the cross-sectional data, but it also makes the projection 110 more susceptible to damage during transportation and installation, and increases the storage space required. It is also possible to change the pattern of the projection 110 for each individual target, allowing for identification of individual targets based on their patterns.

In addition, as a physical target member installed inside the tunnel, steel support structures 73 (metal fittings) installed inside the tunnel may be used as targets 100a to 100e, as shown in the tunnel displacement measurement system 10a in Figure 22. In this case, as described above, a plate with projections 110 may be attached to the outer surface of the steel support structure 73, and the area including the projections 110 of the steel support structure 73 may be designated as targets 100a to 100e. Alternatively, as shown in target 100a in Figure 23, the steel support structure 73 itself may have the projections 110 on its outer surface. In this case as well, target 100a is defined as the area including the projections 110 of the steel support structure 73.

In this embodiment, five targets 100a to 100e are provided for each measurement cross-section 70, but some may be omitted. For example, targets 100b to 100d can be omitted, and measurements can be performed using the aforementioned dummy targets at these locations.

While the tunnel displacement measurement systems 10 and 10a in Example 2 are primarily for mountain tunnels and tunnels constructed using the NATM method, this example will instead describe an example of a displacement measurement system primarily for urban tunnels and tunnels constructed using the shield tunneling method.

In shield tunneling, the displacement of the interior space of the lining constructed by the shield method is measured. In a shield tunnel, multiple segment pieces are connected in the circumferential direction to construct a ring-shaped lining called a segment ring in the excavated shaft created by the shield boring machine. Furthermore, multiple segment rings are connected in the axial direction of the tunnel to construct a lining inside the excavated shaft.

The segment piece has its outer surface (outside) facing the ground of the excavation tunnel and its inner surface (inside) facing the tunnel cavity. The tunnel displacement measurement system in this embodiment measures the displacement of the inner surface of the segment using the tunnel displacement measurement device 1a. A gripping device, which is held by the erector of the shield tunneling machine, is exposed on the inner surface (inside) of the segment piece. The gripping device has a recess into which the gripping tool of the erector is inserted, and the gripping device itself has an uneven surface. Furthermore, adjacent segment pieces in the circumferential direction of the tunnel are connected in the circumferential direction of the tunnel via jointing devices provided at the circumferential ends of each segment piece. The jointing devices also have an uneven surface, similar to the gripping device.

In this embodiment, instead of the rock bolt anchoring plate or steel support structure that serves as the target member on the inner surface of the tunnel in Embodiment 2, a metal fitting with uneven surfaces on the inner surface of the segment piece, which is exposed to the inner surface of the tunnel, is used as the target member. For example, gripping fittings and joint fittings are examples of metal fittings with uneven surfaces. In Embodiment 3, the displacement of the tunnel's interior can be measured using the same tunnel displacement measurement method as in Embodiment 2.

This invention is useful for measuring tunnel top settlement and internal cavity shape displacement during tunnel excavation work.

1, 1a Tunnel displacement measuring device 2 Initial measurement unit 3 Displacement measurement unit 4 LiDAR scanner 5 Computer 6, 6a-6e Simulated target 7 Tunnel 8, 8a, 80, 81 Point cloud (data)
10, 10a Tunnel displacement measurement system 11a-11c Scanner target 21 Scanner coordinate acquisition means 22 First point cloud data acquisition means 23 Pseudo target setting means 24 First unevenness feature data extraction means 26 First registration means 32 Second point cloud data acquisition means 33 Search area setting means 34 Second unevenness feature data extraction means 35 Pseudo target identification means 36 Second registration means 37 Displacement measurement means 60 Search area 70 Measurement cross section 71 Tunnel face 72 Wall surface 100a-100e, 100a' Target 110, 110' Protrusion

Claims (9)

A method for measuring the displacement of the tunnel interior over time by setting a target using point cloud data of the tunnel interior acquired by a 3D laser scanner,
It consists of an initial measurement step and a displacement measurement step.
1) The initial measurement step is:
To obtain surface feature data, which is the surface feature of the aforementioned target,
The process includes obtaining the aforementioned surface feature data, then obtaining point cloud data of the inner surface of the tunnel including the target as first point cloud data, and extracting the surface feature data from the first point cloud data to obtain first position information of the target,
Step 2), which is performed after a predetermined time has elapsed, is:
A search area is set that includes a location identified based on the first location information and has a larger area than the target; point cloud data of the search area is acquired; the target is extracted from the point cloud data of the search area based on the surface feature data to acquire second location information.
This includes measuring the displacement of the space inside the tunnel based on the difference between the first position information and the second position information,
The tunnel displacement measurement method is characterized in that the target is a physical target member installed on the wall surface inside the tunnel . The tunnel displacement measurement method according to claim 1, characterized in that the target member has protrusions and the shape of the tunnel cavity has the aforementioned uneven features . The tunnel displacement measurement method according to claim 1, characterized in that the target member is a metal fitting installed on the wall surface inside the tunnel and having the aforementioned uneven surface characteristics in the portion exposed to the inner surface. The 3D laser scanners are installed along each of the walls on both sides in the tunnel width direction.
The tunnel displacement measurement method according to any one of claims 1 to 3, characterized in that each three-dimensional laser scanner acquires point cloud data of the wall surface opposite to the wall surface on which the three-dimensional laser scanner is installed. The tunnel displacement measurement method according to any one of claims 1 to 3, characterized in that the surface feature data, which is the surface feature of the target, is obtained from the imaging results obtained by imaging the target with an imaging device. The tunnel displacement measurement method according to claim 2 , characterized in that the target has protrusions provided at multiple locations in the vertical direction. A device that uses point cloud data of the tunnel's inner surface acquired by a 3D laser scanner to set a target and measure the displacement of the tunnel's interior space over time,
A physical target member is installed as the target on the wall surface inside the tunnel,
A tunnel displacement measurement system having ,
The aforementioned device is
It consists of an initial measurement unit and a displacement measurement unit.
The initial measurement unit is,
To obtain surface feature data, which is the surface feature of the aforementioned target,
After acquiring the aforementioned surface feature data, point cloud data of the inner surface of the tunnel including the target is acquired as first point cloud data, and the surface feature data is extracted from the first point cloud data to obtain the first position information of the target.
The displacement measurement unit is,
A search area is set that includes a location identified based on the first location information and has a larger area than the target; point cloud data of the search area is acquired; the target is extracted from the point cloud data of the search area based on the surface feature data to acquire second location information.
A tunnel displacement measurement system characterized by measuring the displacement of the space inside a tunnel based on the difference between the first position information and the second position information. The tunnel displacement measurement system according to claim 7, characterized in that the target member has protrusions and the shape of the tunnel cavity has the aforementioned uneven features. The tunnel displacement measurement system according to claim 7, characterized in that the target member is a metal fitting installed on the wall surface inside the tunnel and having the aforementioned uneven surface characteristics in the portion exposed to the inner surface.