JPH07103799B2 - Automatic segment assembly equipment - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an automatic segment assembling apparatus for shield work, and in particular, in order to improve the segment assembling accuracy, a relative position between an assembling segment roughly positioned near the assembling position and an existing segment is provided. The present invention relates to a means for detecting a deviation amount of position / orientation.

[0002]

2. Description of the Related Art Conventionally, in the case of automatically assembling a segment by shield work, an optical cutting method as shown in FIG. 3 is used as a means for detecting the amount of deviation between the relative position and orientation of the existing segment and the assembled segment. Some have been adopted (Tetsuji Sonoda et al .; Development of automatic segment assembly system, 1988
Of Construction Machinery and Construction Method Symposium, p. 20 ~
23, Japan Construction Mechanization Association).

This is to irradiate the slit light images A, A ', B, B'of two slit lights from the light projectors 3a, 3b onto the boundary between the existing segment 1 and the assembled segment 2 in the tunnel longitudinal direction. Are imaged by the television cameras 4a and 4b, the image data from these television cameras are subjected to image processing to obtain the coordinates of the end points of each slit light image, and the relative positions of the existing segment 1 and the assembly segment 2 are calculated from the coordinate values. The deviation amount of the posture is calculated, and the positioning of the assembly segment 2 is corrected based on the calculated deviation amount.

[0004]

However, as the deviation amount to be corrected between the existing segment and the assembled segment at the time of assembling the segment, there are six items as shown in FIG. 2 (heaving ΔZ, turning amount ΔY, sliding amount ΔX). , Pitching amount Δ
φ, rolling amount Δθ, yawing amount ΔΨ). With the information obtained from the two-point slit light image shown in FIG. 3, it is not possible to calculate the deviation amount to be corrected for all of the six items (particularly lacking information regarding turning and pitching), Therefore, the assembling accuracy may be insufficient.

An object of the present invention is to provide a segment automatic assembling position capable of improving the segment assembling accuracy by complementing the above-mentioned point that the segment positioning correction cannot be sufficiently performed by the prior art. . Another object is to shorten the time required for segment positioning correction and improve the efficiency of segment assembly work.

[0006]

In order to achieve the above object, an invention according to claim 1 is an erector body installed in a shield machine, and an assembly segment roughly positioned by the erector body in the vicinity of an assembly position. How to cross the tunnel
The first existing segment assembled in a ring shape toward
And a deviation amount detecting means for detecting a deviation amount of relative position / orientation with respect to the second existing segment assembled in the first existing segment, and control according to the detected deviation amount.
A segment automatic assembling apparatus including a main body control device for outputting a signal to the erector main body to perform positioning correction of the assembling segment, wherein the deviation amount detecting means includes the assembling segment and the first existing segment.
Of the slit light on the boundary of the tunnel longitudinal direction.
Image capturing means for capturing a slit light image by the light means and the light projecting means
First visual means comprising an image means and the assembly segment
Of the tunnel and the second existing segment
Projecting means for irradiating the boundary with slit light and this projecting means
Second image pickup means for picking up a slit light image by
And the visual means of the
At least one of visual means and the second visual means
One of them is composed of a plurality of light projecting means and a plurality of image pickup means.
In addition, the image pickup means and the image pickup means that constitute the first visual means.
And an image input / output device and an image memory for inputting signals from the image pickup means constituting the second visual means and storing them as image data, and image processing of the image data stored in the image memory to perform slit light processing. The assembly of the statue
Coordinates on the side of the front end and the end points on the side of the first existing segment
The coordinates and the end point coordinates on the side of the second existing segment are obtained, and from the coordinate values, the first existing segment and the previous segment are calculated.
It calculates a deviation of the relative position and posture between the serial second existing segments and the assembly segment is characterized by providing an arithmetic unit for outputting to the main controller.

[0007] The invention of claim 2 in order to achieve the above another object, in the arithmetic apparatus, the assembly segment and the first of the calculated from the end point coordinates of the slit light images
Clearance at the boundary between the existing segment and the longitudinal direction of the tunnel
And step and the assembly segment and the second existing segment
Instruments and the tunnel transverse boundary gap and the step of comparing the respective allowable values, the tunnel longitudinal direction boundary portion of the gap, be within the allowable value in step both the tunnel <br/> Le transverse boundary and discriminating means for discriminating a peculiar state clearance is present in excess of tolerance only, if the determination result is obtained, extending an axis extending in the shield axis direction X-axis, the shield radial direction perpendicular to the X axis Z-axis uniaxial, the other one axis orthogonal to the X-axis and Z-axis when the Y-axis based on the calculated value of the clearance of the tunnel transverse to the positioning correction of the set <br/> standing segment Required rolling amount (X
About the axis of rotation) and the Y-axis direction, calculates the movement amount in the Z axis direction, characterized in that is equipped with processing means for outputting.

[0008] The invention of claim 3 for the same purpose are, to the arithmetic unit calculates the value of the gap of the slit light image endpoint coordinate value or al the preparative N'ne <br/> Le transverse boundary of the Based on the calculated value, the assembly segment and the first existing segment caused by the clearance at the boundary portion in the tunnel transverse direction
Preliminary seek relative orientation deviations between the second existing segments, wherein this obtained the assembled segments relative orientation deviations were calculated from the end point coordinate values of the respective slit light image
First existing segment and the second existing segment
Subtracted from the relative position deviation between, characterized in that the result is provided with a posture deviation correcting means for outputting as a final relative position deviation.

For the same purpose, the invention according to claim 4 provides the arithmetic unit with the assembly segment and the image data from the image data.
1st existing segment or said 2nd existing segment
If the slit light image end point at the boundary with is not detected, it is determined whether or not the length of the slit light image is more than a predetermined value. the assembly segment slit light image endpoint coordinates the assembly segment side determined by coarse positioning state before the
Ment side and the first existing segment side or the second
Characterized in that the was provided processing means for certification as both a slit light image endpoint coordinates of the existing segment side.

[00010]

With the first visual means, the assembly segment and the first
It is projected on the boundary between the existing segment and the longitudinal direction of the tunnel.
The slit light image is also assembled by the second visual means.
Crossing the tunnel between the first segment and the second segment
The slit light image projected on the boundary is passed through the image input / output device.
Then, the image data is stored in the image memory. That
At this time, at least the first visual means and the second visual means
Either one has a plurality of light projecting means and a plurality of image pickup means.
The tunnel longitudinal direction boundary and how to cross the tunnel
The slit light image on at least one of the
You can get several. As a result, the tunnel longitudinal direction and tunnel
A total of three or more slit light images in the transverse direction can be obtained.
It And the slits stored in these image memories
When the optical image is processed by the arithmetic unit, the assembly segment
Side slit light image has three or more end point coordinates, the first existing section
Segment and the slit optical image of the second existing segment side
3 or more end point coordinates of the
Calculate the deviation amount of 6 items shown in FIG. 2 from the end point coordinate values
You can

Further, both the clearance and the step of the boundary in the tunnel longitudinal direction between the first existing segment and the assembly segment are within the allowable values, and the boundary across the tunnel in the tunnel transverse direction between the second existing segment and the assembly segment. If there is a gap that exceeds the allowable value only in the section, as described later, based on the calculated value of the gap at the boundary in the tunnel crossing direction, the rolling amount and Y-axis direction required for the alignment correction of the assembly segment , By calculating each movement amount in the Z-axis direction by the calculation device and outputting the result to the main body control device, the positioning correction can be completed by one slit light image signal processing and deviation amount calculation.

According to a third aspect of the present invention, the posture deviation amount correcting means includes the second existing installation regardless of whether the clearance or the step between the first existing segment and the assembly segment at the boundary in the tunnel longitudinal direction is within the allowable value. When there is a clearance at the boundary between the segment and the assembled segment in the tunnel crossing direction, it is possible to calculate the actual relative attitude deviation amount by using the relative attitude deviation amount of both segments caused by the clearance as a correction value.

Furthermore the processing means according to claim 4, by previously seeking the slit light image endpoint coordinates of the assembly segment side in a state before the coarse positioning of the pre-assembled segment pairs
Assembly segment after adjusting the position and orientation of the vertical segment
Even if the slit light image at the boundary between the first existing segment and the second existing segment overlaps and the end point of the slit light image cannot be detected from the image data, the assembly obtained in advance Set from the slit light image end point coordinates on the segment side
It is possible to obtain the slit light image end point coordinates on both the vertical segment and the first existing segment or the second existing segment side.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings.

As shown in FIGS. 7 to 10, the erector body 12 used for automatic segment assembly is installed at the rear part of the shield body 11 having a cylindrical shape. This Electa body 1
2 is roughly classified into an erector ring 13 and a slewing motor 16 which are slewing mechanisms, a suspension beam 21 and a pressing jack 22 which are a pressing (heaving) mechanism, a lateral slide frame 24 and a lateral slide jack 25 which are a slewing fine adjustment mechanism, and a longitudinal slide. It includes a front and rear slide frame 27 and a front and rear slide jack 28 which are moving mechanisms, a spherical frame 29 which is a posture control mechanism for pitching, rolling, yawing and the like, posture control jacks 31, 32 and 33 and a segment grip portion 34.

In the following description, the axis extending in the axial direction of the shield body 12 is the X axis, and the shield body 12 is orthogonal to the X axis.
The one axis extending in the radial direction of is the Z axis, the X axis, and the other one axis orthogonal to the Z axis and the Y axis.

The elector ring 13 is a shield body 11.
It is guided by the outer peripheral guide roller 14 and the side surface guide roller 15 arranged at a plurality of positions, and is rotated by a turning motor 16 attached to the shield body 11 via a pinion 17 and a ring gear 18. Along with this, the following parts supported on the elector ring 13 are also turned to the left and right at the same time.

A suspension beam 21 supported by left and right arms 19 of the elector ring 13 via guide rods 20.
Is a pressing jack 2 mounted between the arm 19 and
By the expansion and contraction of 2, the eclectic ring 13 is moved in the radial direction (Z-axis direction), and along with this, the following parts supported on the suspension beam 21 also move in the same direction.

The horizontal slide frame 24 supported by the suspension beam 21 via the linear bearing 23 is a horizontal slide jack 25 mounted between the suspension beam 21 and the suspension beam 21.
Is horizontally slid on the suspension beam 21 in the Y-axis direction by the expansion and contraction, and along with this, the following respective parts supported on the horizontal slide frame 24 also move in the same direction. This movement is used for fine adjustment of the turning direction of the segment grip portion 34.

A front and rear slide frame 27 supported by a horizontal slide frame 24 through a linear bearing 26.
Is slid back and forth in the X-axis direction on the horizontal slide frame 24 due to the expansion and contraction of the front and rear slide jacks 28 mounted between the horizontal slide frame 24 and the following parts supported on the front and rear slide frame 27. Also moves in the same direction.

The spherical frame 29 incorporated in the spherical guide portion 27a of the front and rear slide frame 27 moves as follows by the expansion and contraction of the two attitude control jacks 31 and 32 mounted between the front and rear slide frame 27. do. In FIG. 9, when the two jacks 31 and 32 are simultaneously extended or contracted, the spherical frame 29 is tilted around the X axis including the spherical center G in the arrow θ direction of FIG.
This movement is used for rolling control of the segment grip portion 34. When either one of the jacks 31 and 32 is extended and the other is contracted, the spherical frame 29 is rotated left and right around the Z axis including the spherical center G in the direction of arrow Ψ in FIG. It is used for yawing control of the grip portion 34.

The segment gripping portion 34 suspended from the central axis 30 of the spherical frame 29 is expanded and contracted by the attitude control jack 33 mounted between the spherical frame 29 and the spherical frame 29 to rotate around the central axis 30 by the arrow φ in FIG. Tilted in the direction, and this movement is used for pitching control of the segment gripping portion 34.

The segment grip 34 comprises a male threaded screw shaft 36 which matches the grout hole 35 of the assembly segment 2. In addition, in the segment grip 34,
A drive motor 37 for rotating the screw shaft 36, and a screw shaft 36
Is equipped with an elevating jack 39 for moving up and down together with a drive motor 37 and a bearing bracket 38, and after aligning the screw shaft 36 with the grout hole 35 of the assembly segment 2 placed under the erector by a positioning sensor (not shown), While the screw shaft 36 is rotated, the screw shaft 36 is projected toward the segment 2 and, after the screw is completely screwed into the grout hole 35, the screw shaft 36 is pulled back until the segment 2 hits the end face of the segment grip portion 34, thereby completing the segment grip. .

In automatically assembling the segments, after the assembly segment 2 is roughly positioned near a predetermined assembly position by the erector body 12, the relative position between the existing segment 1 and the assembly segment 2 is ensured in order to ensure the assembly accuracy. -It is necessary to detect the deviation amount of the posture and perform the positioning correction. FIG. 1 shows the deviation amount detecting means adopted in the present invention to perform this positioning correction.

In FIG. 1, reference numeral 1a denotes an assembly segment 2.
On the other hand, it is an existing structure assembled in a ring shape in the longitudinal direction
Segment 1b is assembled into existing segment 1a
The existing segment 2 is an assembly segment that is joined to the existing segment with a bolt or the like. As shown in the figure, when the assembly segment 2 is roughly positioned in the vicinity of a predetermined assembly position, three slit lights from the light projectors 3a, 3b and 3c, two of which are the existing segment 1a and the assembly segment, At the boundary in the tunnel longitudinal direction of No. 2, the other one irradiates the boundary between the existing segment 1b and the assembly segment 2 in the tunnel transverse direction, and each slit light image (hereinafter referred to as light cut image) A, A ', B, B', C,
C'is imaged by the television cameras 4a, 4b, 4c.
As shown in FIG. 11, three sets of visual devices 10 a, 10 b, and 10 c in which these projectors and a television camera are integrated are attached to the erector body 12.

The image data from the television cameras 4a, 4b, 4c are selected by the camera switch 5 and stored in the image memory 7 via the image input / output device 6, and the contents of the image memory 7 are input / output. The image is displayed on the image monitor 8 via the device 6.

The arithmetic unit 9 image-processes the data stored in the image memory 7 to obtain the coordinates of the end points of each light section image, and the relative values of the existing segments 1a and 1b and the assembly segment 2 are calculated from the coordinate values. The position / orientation deviation amount is calculated and output to the control device (main body control device) of the erector main body.

The image processing, the detection of the end points of the light section image, the calculation of the deviation amount, etc. in the arithmetic unit 9 will be described below.

The light section images of the three points are shown in FIG.
As shown in the display images of (b) and (c), two slit light images A, A ', B, B', C, and C'are input. In these light section images, ΔX ₁ , ΔX ₂ , ΔY
₃ indicates a clearance, and ΔY ₁ , ΔY ₂ , and ΔX ₃ indicate steps.
Therefore, each image data is binarized according to a normal image processing method, and only the slit light portion is divided into white and the others are divided into black. Since this binary image is replaced with the original image in the image memory, by scanning the image memory, the end points a, a ', b, b', c, c'of the respective light-section images are displayed on the image memory. The coordinates of are easily found.

In this embodiment, after the segment gripping is completed, the erector body 12 and the assembled segment 2 are in a state of being integrated via the segment gripping portion 34. In addition, FIG.
As shown in FIG. 3, since the visual devices 10a, 10b, 10c that integrate the light projector and the television camera are mounted on the erector body, it is considered that the relative positional relationship between the visual devices 10a, 10b, 10c and the assembly segment 2 is unchanged. obtain.

Therefore, the coordinate system (X,
The relative position / orientation of the existing segments 1a, 1b and the assembly segment 2 will be obtained using Y, Z). Na
In addition, this Fig. 5 shows the segment shape for ease of explanation.
The shape is shown as a rectangular parallelepiped.

The origin O of the X, Y, and Z coordinates is taken as the end point of the assembly segment 2, and the end points of the light section image are a, a ', b, respectively.
b ′, c, c ′, and the coordinates of these points on the image memory are a (Pax, Pay), a ′ (Pax ′, Pay ′), b.
( Pbx , Pby), b '(Pbx', Pby '), c (Pcx,
Pcy) and c '(Pcx', Pcy '). Also X, Y,
The absolute coordinates of each point with respect to the Z coordinate system are a (ax, ay,
az), a '(ax', ay ', az'), b (bx,
by, bz), b '(bx', by ', bz'), c
(Cx, cy, cz), c '(cx', cy ', cz
′),

As is apparent from FIG. 5, ax = 0, ay
= La, az = 0, bx = 0, by = Lb, bz = 0,
cx = Lc, cy = 0 and cz = 0. Where La
, Lb, Lc are distances from the coordinate origin to the end points a, b, c, and are constants uniquely determined when the visual device is installed.

Using the coordinates on the image memory, the following relationship holds.

Ax '= (Pax-Pax'). Multidot.P, ay '
= (Pay−Pay ′) · ΔP + La, az ′ = (Pay−P
ay ′) · ΔP · tan ξ, bx ′ = (Pbx−Pbx ′) · Δ
P, by '= (Pby-Pby'). Multidot.P + Lb, bz '=
(Pby−Pby ′) · ΔP · tanξ, cx ′ = (Pcx−Pc
x ′) · ΔP + Lc, cy ′ = (Pcy−Pcy ′) · Δ
P, cz ′ = (Pcy−Pcy ′) · ΔP · tan ξ where:
ΔP is the length per pixel on the image memory, and ξ is the angle between the slit light and the central axis of the camera sensor (see FIG. 6).

In FIG. 6, the slit light from the projector 3 is incident on the camera sensor central axis at an angle ξ. Let F be the intersection of the slit light and the inner surface of the assembly segment.
The image forming lens 41 and the image sensor 42 of the television camera 4 are arranged on the normal line of, and the intersection point F is located at the point D on the sensor surface.
The image of is connected. Also, slit light and existing segment 1
It is shown that the image of the intersection F'is connected to the point E on the sensor surface, where F'is the intersection with the inner surface.

The deviation amount of the six items is obtained as follows, depending on the relationship between the coordinates of each point on the image memory and the absolute coordinates of each point.

The yawing amount Ψ is tan Ψ = (bx′−ax ′) / (by′−ay ′) rolling amount θ is tan θ = (bz′−az ′) / (by′−ay ′) pitching amount φ Tan φ = (cz ′ −az ′) / (cx ′ −ax ′) The sliding, turning, and heaving amounts can be obtained for each of the three points, but after the yawing, rolling, and pitching amounts have been corrected, , The sliding amount ΔX is ΔX =
ax ', turning amount ΔY is ΔY = cy', heaving amount ΔZ
Is a value represented by ΔZ = bz ′.

Therefore, these values are output as deviation amounts to be corrected from the arithmetic unit 9 to the main body control unit, and the erector main body 12 shown in FIGS. 7 to 10 is output according to each deviation amount.
By operating the position / posture control jack of
The position / orientation of the assembly segment 2 can be corrected.

As described above, basically, it is possible to calculate the deviation amount of 6 items by measuring 3 points per segment, but as shown in FIGS. 12 (a), 12 (b) and 12 (c). There are a plurality of types of segments (2a, 2b, 2c) of different sizes and shapes in one ring, and the assembly order does not mean that the segments are assembled one after another in one direction. As shown in d), (b), and (e), since the parts are assembled from the lower part to the left and right sides alternately, the measurement location changes depending on the assembly direction. Therefore, the segments 2a and 2b need to be measured at at least four points. Further, the segment 2c shown in (c) of the figure is called a key segment, which is the last one incorporated in one ring and is made smaller than the other segments, so the measurement points are different from other segments. I have to change. Considering these points, the visual device has at least 7
A pair is required.

The above is an extension of the embodiment of the invention described in claim 1. According to this embodiment, all the relative position / orientation deviation amounts between the existing segment and the assembly segment can be detected. However, in the peculiar case as shown in FIG. It is necessary to perform light section image signal processing and deviation amount calculation at least twice. That is, as shown in FIG. 13, the existing segment 1
Clearances ΔX ₁ , Δ obtained from the light section images A, A ′, B, B ′ at the boundary between the tunneling direction of a and the assembly segment 2.
X ₂ and the steps ΔY ₁ and ΔY ₂ are both 0 (or within an allowable value), and the clearance ΔY ₃ and the step ΔX are obtained from the light-section images C and C ′ at the boundary between the existing segment 1 b and the assembly segment 2 in the tunnel transverse direction. of the _three, the step [Delta] X ₃ is within tolerance, when only the gap [Delta] Y ₃ exceeds the allowable value, in normal processing, the arithmetic unit 9 outputs only amount deviation corresponding to [Delta] Y ₃ (turning amount [Delta] Y) In response to this, the main body control device causes the lateral slide jack 2 of the erector main body shown in FIG.
Only 5 is operated. By the operation of the lateral slide jack 25, the assembly segment 2 moves in the Y direction in the coordinate system (X, Y, Z) shown in FIG. 13B. When the assembled segment after the movement is shown by a dotted line 2'in FIG. 13B, the clearance between the assembled segment 2'and the existing segment 1b is 0.
However, since the segment has a curved surface, this time the assembly segment 2'and the existing segment 1
As the step between b increases, a step also occurs between the assembly segment 2 and the existing segment 1a. Therefore,
In order to correct these newly generated deviation amounts, the step detection by the light section image and the deviation amount calculation are performed again, and the resulting output is used to output the rolling jack 3 shown in FIGS.
1, 32 and the pressing jack 22 must be operated, and in the end, positioning correction takes a long time.

Claim 2 made to improve this point.
Examples of the described invention will be described below.

First, in a peculiar state as shown in FIG. 13, that is, the clearance and the step between the existing segment 1a and the assembly segment 2 at the boundary in the tunnel longitudinal direction are both 0 (or within the allowable value), and the existing segment 1b is assembled. the detection method of the gap [Delta] Y ₃ when the gap [Delta] Y ₃ only exceeds the permissible value in the tunnel transverse boundary of the segment 2 are present will be described with reference to FIG. 14. In FIG. 14 (a),
The coordinate system (X, Y, Z) is a coordinate system having the midpoint of the assembly segment 2 as the coordinate origin, as in FIG. 13, and the slit light projector 3c and the television camera 4c are on the normal line passing through the end point c of the assembly segment 2. Is installed at a position rotated by the central angle α from the coordinate origin (see FIG. 1). With such a configuration, when the light-section images C and C'at the boundary portion in the tunnel crossing direction between the existing segment 1b and the assembled segment 2 in a peculiar state as shown in FIG.
(B) of FIG. A clearance (distance between the end points c and c ′) ΔY ₃ between the assembled segment 2 and the existing segment 1b is detected from the light section images C and C ′. Since the segment has a curved surface, there is also a step ΔX _3, but in reality the clearance ΔY ₃ is small, so the step ΔX ₃ is almost equal to 0, and only the clearance ΔY ₃ exceeding the allowable value is present. Will be detected.

Next, a method of calculating the deviation amount to be corrected in such a case will be described with reference to FIG.

In FIG. 15A, the assembly segment 2 and the existing segment 1b in the state of FIG. 13 are indicated by arcs PQ and ST. Considering a coordinate system (x, y) where the radius of the segment is R, the central angle to the assembled segment end point P is α, and the center O ′ of the segment arc is the coordinate origin.
The segment arc is naturally represented by x ² + y ² = R ² .
If the existing segment end point is S, the line segment PS = Δ
Y ₃ and <PO'S = η = ΔY ₃ / R (radian)
It is represented by. Furthermore, the coordinates of the end point S are S (R · sin
(Α + η), −R · cos (α + η)).

From this state, when the assembly segment is rotated by η by the rolling jacks 31 and 32 shown in FIGS. 7 to 10, it is rotated like an arc P'Q 'shown in FIG. 15 (b). The center of the arc P′Q ′ moves to O ″, and as is clear from the figure, the coordinates of the center O ″ are O ″ (− R · sin
η, −R (1 −cos η)). On the other hand, the coordinates of the assembly segment end point P ′ are shown in the same figure as P ′ (− R · sin η +
R ・ sin (α + η), -R (1-cos η) -R ・ cos
(Α + η)). Then, the distances Δx and Δy in the x and y directions between the line segments P′Q are expressed by the following mathematical expressions.

Δx = R · sin (α + η) + R · sin η−R · sin (α + η) = R · sin η Δy = R · cos (α + η) + R (1−cos η) + R · cos (α + η) = R (1-cos η) Therefore, when the assembly segment is moved by Δx and Δy, the end point P ′ and the end point Q coincide with each other. Also, Δx, Δy
By moving only the center O ″ and the center O ′, the assembled segment and the existing segment exist on the same arc. From the above, when only the clearance ΔY ₃ is detected, the assembly is completed. Segment η = ΔY ₃ / R
After rolling by (radian), R · sin η in the Y direction and R in the Z direction of the coordinate system (X, Y, Z) shown in FIG.
It can be seen that it is sufficient to move by (1-cos η).

In order to perform such positioning correction, calculation
FIG. 16 shows the processing contents added to the processing flow of the device 9.
You First, in step 51, the existing segment 1a and the assembly segment are assembled.
The clearance between the two segments ΔX₁, ΔX₂, Step ΔY₁, Δ
Y₂Is within the allowable value, and if NO, then the
Normal processing at step 53 (ΔX₁, ΔX₂, ΔY₁, ΔY₂
Deviation calculation for correcting the above). In step 51
If YES is determined, in step 52, the existing segment
Clearance ΔY between component 1b and assembly segment 2₃ Is the allowable value
Judge whether it is within the range, and if YES, end the processing with this.
However, in the case of NO, η and Δ described above in step 54.
The deviation amount of x and Δy is calculated, and the calculation result is obtained in step 55.
Is output. Based on the output of this result, the main body control device is shown in FIG.
-Rolling jack 3 of the erector body shown in FIG.
1, 32, horizontal slide jack 25 and push jack
22 is operated to complete the positioning correction.

Next, the posture deviation amount correcting means of claim 3 will be concretely described. First, the light section image A shown in FIG.
The relative positions of the existing segments 1a, 1b and the assembly segment 2 from A ', B, B', C, C'using the reference coordinate system (X, Y, Z) on the assembly segment 2 shown in FIG. The amount of dynamic posture deviation is determined.

With respect to the origin O of the reference coordinate system (X, Y, Z), the existing segments 1a, 1b and the assembly segment 2 are arranged.
The absolute coordinates of the end points a, a ', b, b', c, c'of the above light section image are represented by a (ax, ay, az), a '(ax', ay
', Az'), b (bx, by, bz), b '(bx
', By', bz '), c (cx, cy, cz),
If c '(cx', cy ', cz'), the absolute coordinate values of these points can be calculated from the coordinate values of the points on the image memory as described above. Now, on the assembly segment 2, vectors u and b connecting b and a with b as the base point,
Assuming a vector v connecting between c, existing segment 1a,
A vector u'which connects b'and a'with b'as a base point on 1b and a vector v'which connects b'and c'are assumed (only vectors u'and v'are shown in FIG. 17). . When these two vectors u, v and u ′, v ′ are converted into unit vectors and their outer products are obtained, a vector orthogonal to the two vectors is obtained. The obtained vector is converted into a unit vector, for example, the unit product of u and v'is again obtained, and the unit vector is converted to obtain three unit vectors orthogonal to each other on b and b '.
These three unit vectors are orthogonal three axes (x, y,
z), (x ', y', z ') (in FIG. 7, x',
Only y ′ and z ′ are shown), and the relationship between these three orthogonal axes and the reference coordinate system (X, Y, Z) is given by the following mathematical formula 1.

[0051]

[Equation 1]

On the other hand, as shown in FIG. 17B, X,
Let θ ′, φ ′, Ψ ′ be the rotation angle deviations of the x ′, y ′, and z ′ axes with respect to the Y and Z axes, and similarly, the rotation angle deviations of the x, y, and z axes with respect to the X, Y, and Z axes. Let θ be the θ, φ, and Ψ, then the relationship of the following Equation 2 is established.

[0053]

[Equation 2]

From the equations 1 and 2, on the assembly segment side
Is θ = n₂, Φ = −n₁, Ψ = m₁ As well as existing
On the segment side, θ '= n₂′, Φ = −n₁′, Ψ ＝
m₁The rotation angle with respect to the reference coordinate system (X, Y, Z) as'
Deviation is required. Therefore, the assembly segment 2 and the existing segment
The relative posture deviation amount (rotation deviation amount) of the components 1a and 1b is Δθ = θ−θ ′, Δφ = φ−φ ′, ΔΨ = Ψ−Ψ ′.
Can be obtained.

By the way, in the state where the assembled segment 2 is roughly positioned, as shown in FIGS. 18 (a) and 18 (b), there is a clearance L in the turning direction at the boundary portion between the adjacent existing segment 1b and the tunnel crossing direction. In case of TV camera 4c
The distance difference K in the Z direction is detected as a relatively large value between the end points c and c ′ of the light section image captured by the above because of the arc shape of the segment. On the other hand, between the end points a, a ′ and b, b ′ of the light section image at the boundary between the existing segment 1a and the assembly segment 2 in the tunnel longitudinal direction,
The distance difference in the Z direction is detected as a relatively small value. Therefore, in the calculation result of the relative attitude deviation amount described above, Y for the existing segments 1a, 1b of the assembly segment 2 is calculated.
The pitching amount Δφ about the axis is detected as a value larger than the actual value. When the angle between the end points c and c ′ with respect to the segment center O ′ is η, the rolling amount Δθ around the X axis with respect to the existing segments 1a and 1b of the assembly segment 2 is detected by η larger than the actual value. In this way, when the attitude deviation amount is obtained only from the coordinates of each light-section image end point, the pitching amount Δφ due to the influence of the clearance L in the turning direction.
Since the rolling amount Δθ is detected as a value larger than the actual value, the actual posture deviation amount cannot be obtained unless the posture deviation caused by the clearance L is corrected.

As a correction method, first, the TV camera 4
The value of the clearance L between c and c ′ shown in FIG. 18 is obtained from the light section image obtained by c. This value of L is expressed as L≈R · η, where R is the segment inner diameter. Therefore, the rolling amount δ around the X axis due to the clearance L in the turning direction is obtained as δ = η≈L / R. Further, when the angle between the mounting position of the television camera 4c and the Z axis with respect to the segment center O'is α, the pitching amount ω around the Y axis due to the turning direction clearance L is ω = R {cos α-cos
It is calculated as (α + L / R)} / Lcx. Where L
cx is the distance in the X direction between the mounting position of the television camera 4c and the end of the assembly segment 2 on the side of the existing segment 1a. Therefore, the actual posture deviation amount can be obtained by subtracting the above-mentioned δ and ω as the correction amounts from the posture deviation amount obtained from the above-mentioned light section image end point coordinates. That is, the actual posture deviation amount (rolling amount Δθ and pitching amount Δφ) is as follows.

Δθ = θ−θ′−δ Δφ = φ−φ′−ω The above-described procedure for correcting the posture deviation amount is performed at step 61 in FIG.
~ 65. The calculation device 9 in FIG. 1 performs this posture deviation amount correction calculation, and outputs the result to the main body control device as the final posture deviation amount.

FIG. 20A is experimental data showing the relationship between the number of control of positioning correction and the attitude deviation amount when the above correction is not performed and when the above correction is performed, respectively. In (), the posture deviation amount does not fall within the allowable range unless the control is repeated about four times, but (b)
Then, all the posture deviation amounts are within the permissible range by the control performed twice.

Next, the processing means of claim 4 will be concretely described. After rough positioning of the assembled segment, slit light is irradiated to each boundary portion between the existing segment and the assembled segment in the tunnel longitudinal direction and the tunnel transverse direction, and when the end point coordinates of each slit light image (light cut image) are obtained, FIG.
As shown in FIG. 3B, the two slit light images A and A'on the assembled segment side and the existing segment side overlap,
The end points may not be detected by normal image processing. This state indicates that the end points of the slit light images A and A'match, but the position / orientation deviation amount cannot be calculated unless the end point coordinates are obtained even at one location, so rough positioning of the assembly segment is performed. You have to start over, and it takes extra time.

The processing means of claim 4 solves this problem, and its procedure will be described below with reference to the flowcharts of FIGS. 21 and 22. As shown in FIG. 21, when the erector preliminarily grips the assembly segment, the assembly segment is irradiated with slit light, and the light section images A, A ′, B, B ′, C, and C ′ shown in FIG. The coordinate values of the end points a, b and c on the assembly segment side (probably Ao, Bo and Co) are obtained (step 71). FIG. 23A shows a state in which the end point a of the light-section images A and A'on the assembly segment side is obtained, and the same applies to the light-section images B, B ', C and C'. Then, the assembled segment is roughly positioned near the existing segment (step 72). In this coarsely positioned state, the images of the light-section images A, A'B, B ', C, C'of the boundaries between the existing segment and the assembled segment are input to detect the respective end points (step 73). Then, it is judged whether or not there is an end point for the light section images A, A'B, B ', C, C' (steps 74, 75, 7).
6) If all of the end points a, a ', b, b', c, c'exist, the deviation amount is calculated based on the detected end point coordinate values and the result is output (step 80, 8
1). If it is determined in any of steps 74, 75, and 76 that there is no end point, steps 77, 78, and 7
9 is reprocessed. As shown in FIG. 22, in these reprocessings, first, the length of the light section image (XL in FIG. 23B) is obtained (step 81). The length of the light section image on the image memory can be easily obtained by image scanning. Next, it is determined whether or not the length of this light section image is a predetermined value or more (step 82). For example, when the length XL of the light section image is detected as the sum of the lengths of the slit light images A and A'on the assembled segment side and the existing segment side as shown in FIG.
It is considered that the end points of these two slit light images coincide with each other, and the coordinate value Ao of the end point a on the assembly segment side obtained in advance is recognized as the end point coordinate values on both the assembly segment side and the existing segment side (step 83). ). When the length of the light section image is less than or equal to a predetermined value, for example, when the deviation between the assembled segment and the existing segment is large and the slit light image on the side of the existing segment does not enter the field of view of the camera, stop the control once, The rough positioning is restarted (step 84).

In the arithmetic unit 9, the above steps 71 to 84 are carried out.
By providing the processing functions up to the above, the slit light images on the assembled segment side and the existing segment side overlap, and even if the end point cannot be detected by the normal processing, it is possible to obtain accurate end point coordinates without falling into control failure.

[0062]

According to the first aspect of the invention, all the relative position / posture deviation amounts of the existing segment and the assembly segment can be accurately obtained by the optical cutting method, and the position correction of the assembly segment is sufficient. Therefore, high-precision segment automatic assembly is possible.

In addition to this, according to the invention of claim 2, both the clearance and the step of the boundary portion in the tunnel longitudinal direction of the existing segment and the assembled segment are within the allowable value, and the allowable value is exceeded only in the boundary portion in the tunnel transverse direction. In the unique case where there is a gap, the optical cutting image signal processing and the deviation amount calculation, which are required at least twice in the normal processing, are required only once, and the time required for the positioning correction can be shortened.

According to the third aspect of the present invention, when there is a clearance at the boundary between the existing segment and the assembled segment in the tunnel crossing direction, the apparent posture deviation (rolling amount and pitching amount) due to the clearance is subtracted. Since the actual posture deviation amount is calculated and the positioning correction is performed based on the result, the number of times of the positioning correction control is reduced as shown in FIG. 20, the required time can be shortened, and the calculated posture deviation amount is the actual posture deviation amount. When the deviation amount is larger than the deviation amount, the segments do not collide with each other at the time of posture correction, and the damage of the segment or the eclectic mechanism portion is not caused, which is safe.

Further, according to the invention described in claim 4, even when the slit light images on the existing segment side and the assembled segment side are overlapped and the end point cannot be detected by the normal processing, the end point coordinates are not re-executed without performing the rough positioning again. Therefore, the time required for segment positioning can be further shortened.

[Brief description of drawings]

FIG. 1 is a schematic diagram of deviation amount detection means that is a main part of the present invention.

FIG. 2 is an explanatory diagram of a position / orientation deviation amount of a segment.

FIG. 3 is an explanatory diagram of a conventional technique.

FIG. 4 is an explanatory view of a light section image and its positional relationship in the present invention.

FIG. 5 is an explanatory diagram of a segment coordinate system.

FIG. 6 is an explanatory diagram of a visual device used in the present invention.

FIG. 7 is a cutaway front view of an erector body installed in a shield machine.

8 is a sectional view taken along line VIII-VIII of FIG.

9 is a sectional view taken along line IX-IX in FIG. 7.

10 is a cross-sectional view taken along line XX of FIG.

FIG. 11 is a diagram showing a positional relationship between a segment grip portion of an erector body and a visual device.

FIG. 12 is an explanatory diagram of types of segments and an assembly order.

FIG. 13 is a view showing a peculiar state in which a gap exists only in a tunnel transverse direction boundary portion between an existing segment and an assembled segment, (a) is a plan view, and (b) is a front view.

FIG. 14 is an explanatory diagram of clearance detection in the state of FIG.

15 is an explanatory diagram of a deviation amount calculation in the state of FIG.

16 is a flowchart of deviation amount calculation in the state of FIG.

FIG. 17 is an explanatory diagram of a posture deviation amount calculation using a vector.

18A and 18B are explanatory views of an influence of a turning direction clearance between an existing segment and an assembly segment on rolling and pitching deviation, FIG. 18A is a front view, and FIG.

FIG. 19 is a flowchart of a posture deviation amount correction calculation for the influence of the turning direction clearance.

FIG. 20 is a diagram showing the relationship between the number of times of control and the posture deviation amount when the posture deviation amount is not corrected and when the posture deviation amount is not corrected.

FIG. 21 is a flowchart showing an example of a procedure for detecting a light-section image end point and calculating a deviation amount.

FIG. 22 is a flowchart of reprocessing in FIG.

FIG. 23 is a diagram showing a relationship between a light section image before rough positioning and a light section image after rough positioning.

[Explanation of symbols]

1a, 1b Existing segment 2 Assembly segment 3a, 3b, 3c Projector 4b, 4b, 4c TV camera 5 Camera switcher 6 Image input / output device 7 Image memory 9 Computing device 11 Shield body 12 Elector body 34 Segment gripping portion 51,52 Claims Steps 54 and 55 corresponding to the determining unit of Item 2 Steps 61 to 65 corresponding to the processing unit of Claim 2 Steps 71 to 79, 81 to 83 corresponding to the attitude deviation amount correcting unit of Claim 3 Processing of Claim 4 Steps that correspond to means

Front page continuation (72) Inventor Shinsaku Tsutsui 650 Kintatemachi, Hitachi Construction Machinery Co., Ltd. Tsuchiura, Ibaraki Prefecture Tsuchiura Plant Co., Ltd. (72) Inventor Hikari Yamamoto 650 Jinmachi, Tsuchiura City Ibaraki Hitachi Construction Machinery Co., Ltd. Tsuchiura Plant (72) Inventor Yasuo Mori, Tsuchiura Plant, Hitachi Construction Machinery Co., Ltd., 650 Kintate-cho, Tsuchiura City, Ibaraki Prefecture (72) Inventor Hajime Ozawa, 650 Jinmachi, Tsuchiura-shi, Ibaraki Hitachi Construction Machinery Co., Ltd., Tsuchiura Plant (72) Inventor Yoshio Nakajima 502 Jinritsucho, Tsuchiura-shi, Ibaraki, Institute of Mechanical Research, Hiritsu Manufacturing Co., Ltd. (72) Inventor, Masakatsu Fujie 502, Jinritsu-cho, Tsuchiura, Ibaraki, Institute of Mechanical Research, Ltd. (56) References Special Kaihei 1-263509 (JP, A)

Claims (4)

[Claims] 1. An erector body installed in a shield machine, an assembly segment roughly positioned near the assembly position by the erector body, and a ring in a tunnel longitudinal direction.
Existing segment assembled into a
Deviation amount for detecting the deviation amount of relative position / posture with the second existing segment assembled into one existing segment
The detection means and the control signal corresponding to the detected deviation amount are
In a segment automatic assembling apparatus including a main body control device for outputting to an erector main body to perform positioning correction of the assembling segment, in the deviation amount detecting means, the assembly segment and the first existing segment are combined.
A projector that irradiates slit light to the boundary in the longitudinal direction of the tunnel.
Stage and an imaging hand that captures a slit light image by this light projecting means
First visual means comprising steps and said assembly segment
Boundary across the tunnel with the second existing segment
And a light projecting means for irradiating the portion with slit light.
And a second image forming means for picking up a slit light image.
A visual sense means on the main body of the erector to reduce the number of the first visual means and the second visual means.
Either one of a plurality of light projecting means and a plurality of imaging means
And an image pickup unit that constitutes the first visual unit and the second unit.
And an image output device and an image memory for storing a signal from the imaging means as the input image data forming the visual means, the assembly of the image data stored in the image memory the image processing to the respective slit light image End point coordinates on the segment side
And the end point coordinates on the side of the first existing segment and the second
Find the end point coordinates on the existing segment side and from the coordinate values
Segment is characterized by providing an arithmetic unit for outputting calculates the deviation of the relative position and orientation between the first existing segment and said second existing segments and the assembled segments to the main control device Automatic assembly equipment. Wherein said computing device, the assembled segment and the first of the calculated from the end point coordinates of the slit light images
Clearance at the boundary between the existing segment and the longitudinal direction of the tunnel
And step and the assembly segment and the second existing segment
Instruments and of the tunnel transverse boundary gap and the step of comparing the respective allowable values, the clearance of the tunnel longitudinal direction boundary portion, be within the allowable value in step both the tunnel <br/> Le transverse boundary discriminating means for discriminating specific conditions existing gap exceeding the allowable value only part, if the determination result is obtained, the axis extending in the shield axis direction X-axis, the shield radial direction perpendicular to the X axis Z axis 1 axis extending, when the other one axis orthogonal to the X axis and the Z-axis and Y-axis based on the calculated value of the clearance of the tunnel transverse positioning correction of the set <br/> standing segment Required rolling amount (X
2. The automatic segment assembly apparatus according to claim 1, further comprising processing means for calculating and outputting the amount of rotation about the axis) and the amount of movement in each of the Y-axis direction and the Z-axis direction. Wherein the computing device, the calculated clearance value of the endpoint coordinate value or al the tunnel transverse boundary of the slit light image, the gap between the tunnel transverse boundaries on the basis of the calculated value The resulting assembly segment and the
First existing segment and the second existing segment
The relative orientation a deviation amount, before calculating the relative position deviations that this calculated from the end point coordinate values of the respective slit light images of the
Assembly segment and first existing segment and front
The segment according to claim 1, further comprising posture deviation amount correcting means for subtracting the relative posture deviation amount from the second existing segment and outputting the result as a final relative posture deviation amount. Automatic assembly equipment. 4. The arithmetic unit is configured to extract the set from image data.
Vertical segment and the first existing segment or the first segment
When the slit light image end point at the boundary with the existing segment 2 is not detected, it is determined whether or not the length of the slit light image is equal to or more than a predetermined value. As a result, when it is determined that the length is equal to or more than the predetermined value, , The slit light image end point coordinates on the side of the assembled segment obtained in advance before the rough positioning of the assembled segment are the side of the assembled segment and the side of the first existing segment.
2. The automatic segment assembly apparatus according to claim 1, further comprising processing means for recognizing the slit light image end point coordinates on both sides of the second existing segment.