Measurement space calibration method

Abstract

A method for calibrating a measurement space is provided that can accurately calibrate the measurement space without moving a 3D profiler or a calibration jig. [Solution] This method of calibrating the measurement space is a calibration method that uses a 3D profiler and a moving means, in which the calibration jig 30 has multiple calibration parts and the shape and dimensions including the calibration parts are known, and calibration of the measurement space is performed by positioning the reference point of the calibration jig 30 so that it is located at the origin of the measurement space, and by performing a first step of determining the length dimension and tilt angle of one cross-sectional shape of the calibration jig 30, and a second step of determining a homogeneous transformation matrix between the measurement coordinates of the measurement coordinate space and the measurement coordinates of the measurement space from the length dimension and tilt angle determined in the first step.

Description

[Technical Field]

[0001] The present invention relates to a method for calibrating a measurement space, which calibrates a measurement space when performing three-dimensional measurement to measure the three-dimensional shape of a measurement object. [Background technology]

[0002] Conventionally, the light-section method has been known as a method for measuring the three-dimensional shape of an object in a non-contact manner.

[0003] As an example of a method using the light-section method, Patent Document 1 listed below describes a calibration method for a shape measuring device in which at least three sets of light-emitting body groups, each including a plurality of light-emitting points, corresponding to a plurality of feature points in the bottom shape of a cylindrical calibration target whose bottom shape and dimensions are known, are arranged side by side at different positions in the height direction of the cylinder of the calibration target, and the feature points are positioned on the same line as the light-emitting points corresponding to the feature points; the position coordinates of the feature points of the calibration target are identified based on the fact that the cross ratio is invariant with the captured images of each light-emitting body and a line laser and projective transformation, and the device is calibrated.

[0004] Incidentally, when a 3D profiler used in the light-section method is handled by an articulated robot or the like to perform 3D measurements of an object, for example, in the following cases (1) and (2), it is necessary to calibrate the measurement space in which the object is placed as the measurement coordinate system of the robot or the like, and to obtain 3D information including the position of the object within the measurement space.

[0005] That is, (1) when a 3D profiler is held at the tip of a robot or the like and the robot or the like moves the 3D profiler to perform three-dimensional measurements of an object to be measured, it is necessary to align the measurement coordinates of the 3D profiler in the measurement coordinate space with the measurement coordinates of the object to be measured in the measurement space (for example, when the position of the object to be measured is determined by positioning pins and you want to identify the coordinates of a specific part (for example, a reference hole for mounting another part)), or (2) when it is necessary to integrate three-dimensional information measured by multiple articulated robots or the like. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Laid-Open No. 2012-189372 Summary of the Invention [Problem to be solved by the invention]

[0007] In the above cases (1) and (2), there is a demand to calibrate the measurement coordinates of the measurement space in which the measurement object is placed as the measurement coordinates of a robot or the like that handles the 3D profiler.

[0008] However, the measurement coordinates (measurement coordinate origin) that serve as the basis for the 3D profiler's measurement data are invisible, and the only data that can be obtained when the 3D profiler is stationary is the cross-sectional shape data of the object being measured, so the coordinate space registration method using three-point teaching for a typical robot cannot be applied.

[0009] Furthermore, there is no method for registering coordinates while visually observing a calibration target on a monitor, as is the case with so-called vision sensors.

[0010] Therefore, when matching the measurement coordinates of the measurement space to be measured with the measurement coordinates measured by the 3D profiler, for example, three-dimensional measurements are performed by moving the 3D profiler or the calibration jig, and the measurement data is compared with the dimensions of the calibration jig, and calculations are made from the results.

[0011] In this case, the calibration accuracy of the measurement space may not be sufficient, and there is a demand for an improvement in the calibration accuracy of the measurement space.

[0012] Therefore, an object of the present invention is to provide a method for calibrating a measurement space that can accurately calibrate the measurement space without moving the 3D profiler and the calibration jig. [Means for solving the problem]

[0013] In order to achieve the above object, the present invention provides a method for calibrating a measurement space, in which, when performing three-dimensional measurement to measure the three-dimensional shape of a measurement object using a 3D profiler used in a light-section method and a moving means for relatively moving the 3D profiler and a measurement object placed in a measurement space, a calibration jig is placed in the measurement space and the calibration jig is measured with the 3D profiler, and the measurement space is calibrated, wherein when translational deviations in the X direction, Y direction, and Z direction in an XYZ three-dimensional coordinate system between the measurement space and a measurement coordinate space measured by the 3D profiler are respectively ΔX, ΔY, and ΔZ, and rotational deviations around axes along each direction as rotation centers are respectively θX, θY, and θZ, the calibration jig is The calibration system has a plurality of calibration parts in which the length dimensions and tilt angles of the cross-sectional shapes obtained by the light-section method change with respect to the θZ, and the shape and dimensions including the calibration parts are known, and the calibration of the measurement space includes a first step of arranging the calibration jig in the measurement space so that a reference point of the calibration jig is located at the origin of the measurement space, and measuring the calibration jig by the light-section method to obtain the length dimensions and tilt angles of one cross-sectional shape of the calibration jig, and a second step of obtaining a homogeneous transformation matrix between the measurement coordinates of the measurement coordinate space including the 3D profiler and the moving means and the measurement coordinates of the measurement space to be calibrated, from the length dimensions and tilt angles of one cross-sectional shape of the calibration jig obtained in the first step.

[0014] According to the above invention, the calibration jig has a plurality of calibration sections in which the length dimensions and tilt angles of the cross-sectional shape obtained by the light-section method change with respect to ΔX, ΔY, ΔZ, θX, θY, and θZ. Therefore, in the second calibration step, it is possible to determine a homogeneous transformation matrix between the measurement coordinates including the 3D profiler and moving means and the measurement coordinates of the measurement space from the length dimensions and tilt angle of one cross-sectional shape of the calibration jig, and the measurement space can be calibrated with high accuracy without moving the 3D profiler or calibration jig.

[0015] In the method for calibrating a measurement space according to the present invention, the calibration unit may have a pair of inclined portions formed at the upper end in the Z direction and on both sides in the Y direction, and having an inclined surface shape that is inclined at a predetermined angle in the Z direction with respect to a ceiling surface of the calibration jig in a front view and inclined at a predetermined angle in the X direction with respect to the front surface of the calibration jig in a plan view; and a concave portion formed at the upper end in the Z direction between the pair of inclined portions and having a groove shape that is recessed with respect to the ceiling surface in a front view and extends in the X direction in a plan view, or a convex portion formed at the upper end in the Z direction between the pair of inclined portions and that is convex with respect to the ceiling surface in a front view and has a convex shape that extends in the X direction in a plan view.

[0016] According to the above aspect, since the calibration section has a pair of inclined portions and a concave portion, or a pair of inclined portions and a convex portion, the accuracy of determining the homogeneous transformation matrix can be improved in the second calibration step.

[0017] In the method for calibrating a measurement space according to the present invention, in the first step, the 3D profiler may be moved by the moving means so as to bring each translational deviation amount and each rotational deviation amount between the measurement space and the measurement coordinate space closer to zero, and measurement of the calibration jig by the 3D profiler may be repeated.

[0018] According to the above aspect, since the first step has the above configuration, it is possible to reduce errors due to interference between each translational deviation amount and each rotational deviation amount, and to improve the accuracy when determining the homogeneous transformation matrix.

[0019] In the method for calibrating a measurement space according to the present invention, the calibration jigs may include a first jig and a second jig, the calibration jig being arranged so that a reference point of the first jig is positioned at the origin of the measurement space, and the second jig being arranged at a distance that is at least one-third of the length in the X direction of an area actually used in the measurement space, and both the first jig and the second jig may be measured by the 3D profiler in the first step.

[0020] According to the above aspect, the calibration jig is composed of the first and second jigs and has the above layout, and the first step of calibration has the above configuration, so that the homogeneous transformation matrix can be corrected so as to improve the accuracy of the tilt angle at θY and the tilt angle at θZ in the homogeneous transformation matrix in particular.

[0021] In the method for calibrating a measurement space according to the present invention, there may be a plurality of 3D profilers, each of which may be movable by the movement means, and in the first step, all of the 3D profilers may measure the same calibration jig.

[0022] According to the above aspect, in the first step of calibration, all 3D profilers measure the same calibration jig, thereby improving the overall consistency and measurement accuracy of the measurement data of the calibration jig by the 3D profilers in the same measurement coordinate system.

[0023] In the method for calibrating a measurement space according to the present invention, the calibration jig may have a pair of inclined portions formed symmetrically with respect to a line segment extending parallel to the Y direction, and the concave portion formed across the entire area in the X direction, and the number of 3D profilers may be two, with a predetermined profiler being disposed on the front side of the calibration jig and the other profiler being disposed on the back side of the calibration jig in an orientation facing the predetermined profiler.

[0024] According to the above aspect, since the calibration jig has the above shape, the measurement space can be calibrated using two 3D profilers arranged on the front and back sides of the same calibration jig, so the measurement space can be calibrated quickly.

[0025] In a method for calibrating a measurement space according to the present invention, the moving means is an articulated robot, the calibration jigs include a first jig and a second jig, the first jig is positioned so that a reference point of the first jig is located at the origin of the measurement space, and the method has a first measurement coordinate system having its origin at the reference point of the first jig and a second measurement coordinate system having its origin at the reference point of the second jig, and in the first step, either the first measurement coordinate system or the second measurement coordinate system may be selected depending on the distance between the position of the 3D profiler in the measurement space and the installation position of the first jig, or the distance between the position of the 3D profiler in the measurement space and the installation position of the second jig.

[0026] According to the above aspect, in the first step, it is possible to select either the first measurement coordinate system or the second measurement coordinate system depending on the distance between the position of the 3D profiler in the measurement space and the installation position of the first jig, or the distance between the position of the 3D profiler in the measurement space and the installation position of the second jig, thereby reducing the impact of positioning errors of the articulated robot on the accuracy of three-dimensional measurement by the 3D profiler. [Effects of the Invention]

[0027] In the present invention, the calibration jig has multiple calibration sections in which the length dimensions and tilt angles of the cross-sectional shape obtained by the light-section method change with respect to ΔX, ΔY, ΔZ, θX, θY, and θZ. Therefore, in the second calibration step, it is possible to determine a homogeneous transformation matrix between the measurement coordinates including the 3D profiler and moving means and the measurement coordinates of the measurement space from the length dimensions and tilt angle of one cross-sectional shape of the calibration jig, and this makes it possible to accurately calibrate the measurement space without moving the 3D profiler or calibration jig. [Brief explanation of the drawings]

[0028] [Figure 1] 1 is a perspective view showing a first embodiment of a method for calibrating a measurement space according to the present invention; [Figure 2] 3A and 3B are schematic explanatory diagrams of the calibration principle in the calibration method for the measurement space. [Figure 3] FIG. 2 is a schematic perspective view of the calibration principle in the calibration method for the measurement space. [Figure 4] FIG. 10 is a side view illustrating a method for calibrating the measurement space. [Figure 5] FIG. 10 is a plan view illustrating a method for calibrating the measurement space. [Figure 6] FIG. 2 is a front view of a calibration jig used in the method for calibrating the measurement space. [Figure 7] FIG. 2 is a plan view of a calibration jig used in the method for calibrating the measurement space. [Figure 8] FIG. 2 is a side view of a calibration jig used in the method for calibrating the measurement space. [Figure 9] 9 is a cross-sectional explanatory view taken along the line IX-IX of FIG. 7 when each parameter is calculated in the first step of calibration in the method for calibrating the measurement space. [Figure 10] FIG. 10 is a first explanatory plan view when each parameter is calculated in the first step of the calibration. [Figure 11] FIG. 10 is a first explanatory front view when each parameter is calculated in the first step of the calibration. [Figure 12]FIG. 10 is a second explanatory plan view when each parameter is calculated in the first step of the calibration. [Figure 13] FIG. 10 is a second front view illustrating the calculation of each parameter in the first step of the calibration. [Figure 14] FIG. 10 is a perspective explanatory view showing a second embodiment of the method for calibrating a measurement space according to the present invention. [Figure 15] FIG. 10 is a perspective view of a calibration jig used in a third embodiment of the method for calibrating a measurement space according to the present invention. [Figure 16] FIG. 10 is a perspective view of a calibration jig used in a fourth embodiment of the method for calibrating a measurement space according to the present invention. [Figure 17] FIG. 10 is a perspective view of a calibration jig used in a fifth embodiment of the method for calibrating a measurement space according to the present invention. [Figure 18] FIG. [Figure 19] FIG. 10 is a perspective view of a calibration jig used in a sixth embodiment of the method for calibrating a measurement space according to the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0029] (First embodiment of the method for calibrating the measurement space) A first embodiment of a method for calibrating a measurement space according to the present invention will be described below with reference to FIGS.

[0030] In the well-known light section method, a 3D profiler is attached to a moving means (for example, an articulated arm of a robot) and measures a measurement object placed in a measurement space.

[0031] 2, during the above measurement, a predetermined measurement coordinate A1 in measurement space A where the measurement object is placed and a measurement coordinate B1 in measurement coordinate space B where the 3D profiler 10 and moving means 20 are placed are usually shifted in the X, Y, and Z directions in the well-known XYZ three-dimensional coordinate system, or in the rotational direction around an axis along each direction. Therefore, it is necessary to calibrate the measurement coordinate A1 in measurement space A so that the measurement coordinate B1 of a robot or the like having the moving means 20 that moves the 3D profiler 10 matches the measurement coordinate A1 (see the lower diagram in FIG. 2).

[0032] Therefore, as shown in Figures 1, 4 and 5, the calibration method for a measurement space according to the present invention (hereinafter also simply referred to as the "calibration method") is a measurement system that performs three-dimensional measurement to measure the three-dimensional shape of a measurement object using a 3D profiler 10 used in the light-section method and a moving means 20 that moves the 3D profiler 10 and a measurement object 1 placed in a measurement space A relatively, and calibrates the measurement space A by placing a calibration jig 30 in the measurement space A and measuring the calibration jig 30 only once while the 3D profiler 10 is stationary.

[0033] 3, the moving means 20 in this embodiment is an articulated robot (particularly an articulated arm of an articulated robot) that holds (handles) the 3D profiler 10 and moves the 3D profiler 10 in the X, Y, and Z directions in an XYZ three-dimensional coordinate system relative to the measurement object 1, or rotates it along each direction. That is, the moving means 20 is an articulated robot with six degrees of freedom: one degree of freedom in the X-axis direction, one degree of freedom in the Y-axis direction, one degree of freedom in the Z-axis direction, one degree of freedom around the X-axis, one degree of freedom around the Y-axis, and one degree of freedom around the Z-axis.

[0034] 3, in this embodiment, when a measurement space A1 in which a measurement object is placed for a combination of a 3D profiler 10 and a handling moving means 20 is determined by an attitude theoretically derived from the physical factors of the moving means 20 (installation position, arm length, relative positional relationship of links, joint angle, etc.) and the connection state with the 3D profiler 10, a deviation occurs between the measurement coordinate system B1 realized in this way and the measurement space A1 in which the target measurement object is placed due to the physical factors of the moving means 20 and the connection error with the 3D profiler 10. Therefore, this calibration method calibrates the measurement coordinate system in the moving means 20 and the measurement coordinate system in which the measurement object is placed so that they coincide with each other. In other words, a homogeneous transformation matrix between the measurement coordinate system B1 and the robot coordinate system is derived and registered while the measurement coordinates A1 (user coordinates) of the measurement space A in which the measurement object is placed coincide with the measurement coordinates B1.

[0035] Furthermore, the robot coordinate system is defined on the base of the robot itself, and all robot trajectories are handled based on this coordinate system, while the user coordinate system is defined within the workspace where measurement and assembly operations are performed, and operations within each space are determined based on this coordinate system. Generally, if the layout of the workspace is physically constrained with high precision to match the coordinate system of the robot itself in order to determine operations corresponding to the operations, this is not practical due to cost issues and sacrifices in operational convenience. Therefore, a common approach is to separate the robot and the workspace, establish physical layout constraints for each, and then connect the relative relationship between the robot and the workspace by calibrating the coordinate system. However, the present invention performs the calibration work required in this process with high efficiency and high precision.

[0036] The 3D profiler 10 is used for the light-section method, in which a line of light (laser light) is irradiated onto the object to be measured from a light source and the reflected light from the object to be measured is captured from a direction different from that of the light source, thereby obtaining profile data of changes in the height, shape, position, etc. of the cross section of the object to be measured.

[0037] Furthermore, the moving means 20 may be an articulated robot or a three-dimensional stage having a plurality of sliding mechanisms, rotating mechanisms, or the like.

[0038] In this calibration method, when the translational deviations in the X, Y, and Z directions in a well-known XYZ three-dimensional coordinate system between measurement space A and measurement coordinate space B measured by the 3D profiler 10 are defined as ΔX, ΔY, and ΔZ, respectively, and the rotational deviations about axes along each direction as the center of rotation are defined as θX, θY, and θZ, respectively, the calibration jig 30 has a plurality of calibration parts whose cross-sectional shapes obtained by the light-section method have different length dimensions and tilt angles with respect to ΔX, ΔY, ΔZ, θX, θY, and θZ, and the shape and dimensions including the calibration parts are known.

[0039] The Y direction is perpendicular to the X direction, and the Z direction is perpendicular to the X and Y directions.

[0040] Furthermore, the calibration jig 30 of this embodiment is generally plate-shaped, having a predetermined depth dimension (thickness dimension, which can also be considered as the length dimension in the depth direction) along the X direction, a predetermined width dimension (which can also be considered as the length dimension in the width direction) along the Y direction, and a height dimension along the Z direction. The X direction is the depth direction of the calibration jig 30, the Y direction is the width direction of the calibration jig 30, and the Z direction is the height direction of the calibration jig 30.

[0041] As shown in Figures 1 and 6 to 8, the calibration jig 30 in this embodiment has a bottom surface 31, a front surface 32 arranged orthogonally on one end side of the bottom surface 31 in the X direction, a back surface 33 arranged orthogonally on the other end side of the bottom surface 31 in the X direction, one side surface 34 arranged orthogonally on one end side of the bottom surface 31 in the Y direction, the other side surface 35 arranged orthogonally on the other end side of the bottom surface 31 in the Y direction, and a ceiling surface 36 located at the upper ends in the Z direction of the front surface 32, the back surface 33, and both side surfaces 34, 35.

[0042] The calibration portion of the calibration jig 30 is formed at the upper end in the Z direction and on both sides in the Y direction, and has a pair of inclined portions 40, 45 that are inclined at a predetermined angle in the Z direction relative to the ceiling surface 36 when viewed from the front (when the front surface 32 is viewed from the X direction) and have an inclined surface shape that is inclined at a predetermined angle in the X direction relative to the front surface 32 when viewed from the top (when the ceiling surface 36 is viewed from the Z direction), and a recessed portion 50 that is formed at the upper end in the Z direction between the pair of inclined portions 40, 45 and has a groove shape that is recessed relative to the ceiling surface 36 when viewed from the front and extends in the X direction when viewed from the top.

[0043] Figure 9 shows a predetermined cross-sectional shape of the calibration jig 30. Referring to Figures 9 and 10 together, in this embodiment, the inclined portion 40 is a flat inclined surface having an approximately trapezoidal shape made up of four side edges 41, 42, 43, and 44: a first side edge 41 located on the front surface 32 and inclined at a predetermined angle θ1 in the Z direction with respect to the ceiling surface 36 in a front view; a second side edge 42 located on the ceiling surface 36 and inclined at a predetermined angle θa in the X direction with respect to the front surface 32 in a plan view; a third side edge 43 located on the back surface 33 and parallel to the first side edge 41; and a fourth side edge 44 located on the side surface 34 and connecting the third side edge 43 and the first side edge 41 (it can also be said to be a flat surface having an approximately trapezoidal shape surrounded by the four side edges 41, 42, 43, and 44).

[0044] In addition, the inclined portion 45 is a flat inclined surface having an approximately trapezoidal shape made up of four side edges 46, 47, 48, and 49: a first side edge 46 located on the front surface 32 and inclined at a predetermined angle θ2 in the Z direction relative to the ceiling surface 36 in a front view; a second side edge 47 located on the ceiling surface 36 and inclined at a predetermined angle θa in the X direction relative to the front surface 32 in a plan view; a third side edge 48 located on the back surface 33 and parallel to the first side edge 46; and a fourth side edge 49 located on the side surface 35 and connecting the third side edge 48 and the first side edge 46 together (it can also be said to be a flat surface having an approximately trapezoidal shape surrounded by the four side edges 46, 47, 48, and 49).

[0045] As shown in FIG. 7, the inclined portion 40 and the inclined portion 45 are in a line-symmetrical relationship with respect to a line segment L1 that passes through the center of the calibration jig 30 in the Y direction and that runs along the X direction.

[0046] On the other hand, the recessed portion 50 in this embodiment is formed in the center of the calibration jig 30 in the Y direction, and is a substantially U-shaped recessed groove that is open upward in the Z direction and at both ends in the X direction, and is composed of a bottom surface 51 formed at a predetermined depth from the ceiling surface 36 and side surfaces 52, 53 formed parallel to each other on both sides of the bottom surface 51 in the Y direction. Moreover, this recessed portion 50 is a uniform recessed groove over the entire area in the X direction from the front surface 32 to the back surface 33 of the calibration jig 30.

[0047] The shape, structure (the shape of the entire calibration jig and the shape of the calibration portion), dimensions, etc. of the calibration jig are not limited to the above-described calibration jig 30, and can be selected as appropriate. However, the shape and dimensions including the calibration portion are assumed to be known. Calibration jigs of other shapes will be described in the embodiments described later.

[0048] Furthermore, although the recessed portion 50 in this embodiment is a recessed groove having an approximately U-shape, the recessed portion may also be a recessed groove having, for example, an approximately V-shape, an approximately U-shape, or an approximately arc-shaped curved surface.

[0049] Furthermore, although the calibration jig 30 in this embodiment has the concave portion 50, it may also have a convex portion. That is, the calibration jig may have a convex portion at the upper end in the Z direction between a pair of inclined portions, which protrudes from the ceiling surface in a front view and forms a ridge extending in the X direction in a plan view.

[0050] The calibration of the measurement space A is performed in the following first and second steps.

[0051] The first step is to place the calibration jig 30 in the measurement space A so that the reference point C (see Figure 9) of the calibration jig 30 is located at the origin of the measurement space A, and to determine the length dimension and inclination angle of one cross-sectional shape of the calibration jig 30 (see Figure 9) obtained by measuring the calibration jig 30 using the light cutting method.

[0052] The length dimension refers to the dimension along the X, Y, and Z directions of the calibration jig or its predetermined cross-sectional shape, while the tilt angle refers to the tilt angle with respect to a plane along a predetermined direction of the calibration jig or its predetermined cross-sectional shape, or the tilt angle with respect to each axis when rotated around an axis along each direction as the center of rotation.

[0053] More specifically, the first step is performed, for example, as follows. Fig. 9 shows a cross-sectional shape (cross-sectional shape at the X cross section) of the calibration jig 30 when cut along a predetermined YZ plane. Note that the X cross section in Fig. 9 is a cross section passing through the center of the inclined portions 40, 45 and the recessed portion 50 in the X direction (a cross-sectional view taken along arrows IX-IX in Fig. 7).

[0054] 9, the coordinate C of the reference point of the calibration jig 30 is (Yc, Zc). The coordinate C (Yc, Zc) here is on the upper end of the calibration jig 30 in the Z direction and passes through the center position in the Y direction.

[0055] Furthermore, in the above-mentioned specified cross section shown in Figure 9, (a) the coordinate P1 of the intersection between the first side edge 41 of the inclined portion 40 and the cross-sectional line 54 of the ceiling surface 36 is defined as (Y1, Z1), (b) the coordinate P2 of the side surface 52 of the concave portion 50 on the ceiling surface 36 is defined as (Y2, Z2), (c) the coordinate P3 of the side surface 53 of the concave portion 50 on the ceiling surface 36 is defined as (Y3, Z3), and (d) the coordinate P4 of the intersection between the first side edge 46 of the inclined portion 45 and the cross-sectional line 55 of the ceiling surface 36 is defined as (Y4, Z4).

[0056] Furthermore, in the plan view of the calibration jig 30 shown in Fig. 10, the inclination angle of the second side edges 42, 47 of the inclined portions 40, 45 with respect to a line segment parallel to the front surface 32 is defined as "θa." In addition, in the above-mentioned specified cross section shown in Fig. 9, the inclination angle of the first side edge 41 of the inclined portion 40 with respect to a line segment parallel to the ceiling surface 36 is defined as "θ1," and the inclination angle of the first side edge 46 of the inclined portion 45 with respect to a line segment parallel to the ceiling surface 36 is defined as "θ2."

[0057] Then, "L1" in the above-mentioned predetermined cross section shown in FIG. 9, that is, the shortest distance in the Y direction between the above-mentioned coordinates P1 and P2, is calculated by the following mathematical formula (1).

[0058]

number

[0059]

number

[0060]

number

[0061] Furthermore, the Y position of the calibration jig 30 in the X cross section (the translational deviation ΔY of the measurement coordinate B1 relative to the measurement coordinate A1) is calculated by the following formula (4).

[0062]

number

[0063]

number

[0064]

number

[0065]

number

[0066] Furthermore, the rotational deviation θZ (the rotational deviation θZ of the measurement coordinate B1 relative to the measurement coordinate A1) around the axis along the Z direction in the cross-sectional shape of the calibration jig 30 is calculated by the following formula (8).

[0067]

number

[0068] In other words, the cross-sectional shape of a calibration jig 30 in a specified X-section in the optical section method, in which the reference point C is positioned at the origin of the measurement space A, may be obtained shifted from the origin of the measurement space A, as shown by the dotted line K1 in Figure 11.

[0069] In this case, the translational deviations ΔY and ΔZ of the measurement coordinate B1 relative to the measurement coordinate A1 can be calculated using the above formulas (4) and (5), respectively. Also, the rotational deviation θX of the measurement coordinate B1 around the axis along the X direction perpendicular to the YZ plane as the rotation center can be calculated using the above formula (6).

[0070] Furthermore, the cross-sectional shape of the calibration jig 30 at a predetermined Z cross section from the planar direction in the light-section method may be obtained shifted from the origin of the measurement space A (here, located at the center in the Y direction on the line segment K0), as shown by the two-dot chain line K2 in Figure 12. In this case, the translational deviation ΔX of the measurement coordinate B1 with respect to the measurement coordinate A1 can be calculated using the above formula (3). Also, the rotational deviation θZ of the measurement coordinate B1 around the axis along the Z direction as the center of rotation can be calculated using the above formula (8).

[0071] Furthermore, the cross-sectional shape of the calibration jig 30 at a predetermined X cross section in the light-section method may be obtained with a deviation from the line segment (here, the line segment K0) passing through the origin of the measurement space A, as shown by the two-dot chain line L3 in Fig. 13. In this case, the amount of rotational deviation θY around the axis along the Y direction as the rotation center in the measurement coordinate system B1 can be calculated using the above formula (7).

[0072] Next, the second step of calibration will be described. That is, the second step of calibration of the measurement space A involves determining a homogeneous transformation matrix between the measurement coordinates B1 of the measurement coordinate space B including the 3D profiler 10 and the moving means 20 and the measurement coordinates A1 of the measurement space A to be calibrated, from the length dimension and tilt angle of one cross-sectional shape of the calibration jig 30 determined in the first step (for example, the cross-sectional shape in the X cross section).

[0073] In this second step, the unknowns required to derive the homogeneous transformation matrix T are six parameters: translational deviation ΔX, translational deviation ΔY, translational deviation ΔZ, rotational deviation θX, rotational deviation θY, and rotational deviation θZ, which are calculated in the first step using the above equations (3), (4), (5), (6), (7), and (8), respectively.

[0074] The homogeneous transformation row matrix T can be calculated using the above six parameters by the following equation (9).

[0075]

number

[0076] In the above formula (9), R11 to R33 are rotation matrix parts determined by θX, θY, and θZ, and ΔX, ΔY, and ΔZ are translation vectors, and the product of these is the above formula (9).

[0077] Furthermore, in this embodiment, in the first step, the 3D profiler 10 is moved by the moving means 20 so that the translational deviation amounts ΔX, ΔY, ΔZ and the rotational deviation amounts θX, θY, θZ between the measurement space A and the measurement coordinate space B approach zero, and measurement of the calibration jig 30 by the 3D profiler 10 is repeated.

[0078] Although this embodiment uses one 3D profiler 10, it is also possible to use multiple 3D profilers 10. That is, it is also possible to use multiple 3D profilers 10, each of which can be moved by a moving means 20, and all of the 3D profilers 10 may measure the same calibration jig 30 in the first step.

[0079] (Action and effect) Next, the effects of the method for calibrating the measurement space having the above-described configuration will be described.

[0080] First, the calibration jig 30 is placed in the measurement space A so that the reference point C of the calibration jig 30 is located at the origin of the measurement space A, and the length dimension and tilt angle of one cross-sectional shape of the calibration jig 30 obtained by measuring the calibration jig 30 using the light-section method are determined (first step). In this embodiment, in this first step, the six parameters ΔX, ΔY, ΔZ, θX, θY, and θZ are calculated using the above formulas (3) to (8), as described above.

[0081] Then, from the length dimension and tilt angle of one cross-sectional shape of the calibration jig 30 obtained in the first step, a homogeneous transformation matrix between the measurement coordinate B1 in the measurement coordinate space B including the 3D profiler 10 and the moving means 20 and the measurement coordinate A1 in the measurement space A to be calibrated is obtained (second step). In this embodiment, in this second step, the homogeneous transformation matrix T shown in the above formula (9) can be obtained from the six parameters calculated in the first step.

[0082] In this calibration method, the calibration jig 30 has a plurality of calibration portions (here, a pair of inclined portions 40, 45 and a concave portion 50) whose length dimensions and inclination angles of the cross-sectional shape obtained by the light-section method change with respect to ΔX, ΔY, ΔZ, θX, θY, and θZ.

[0083] Therefore, in the second calibration step, it is possible to calculate a homogeneous transformation matrix T between the measurement coordinates B1 in the measurement coordinate space B, which includes the 3D profiler 10 and the moving means 20, and the measurement coordinates A1 in the measurement space A, from the length dimension and tilt angle of one cross-sectional shape of the calibration jig 30. As a result, the measurement space A can be calibrated with high accuracy without moving the 3D profiler 10 and the calibration jig 30.

[0084] Furthermore, when calibrating the measurement space A, there is no need to move the 3D profiler 10 and the calibration jig 30, and it is only necessary to measure the calibration jig 30 once, so calibration of the measurement space A can be performed easily and quickly.

[0085] In this embodiment, the calibration unit is made up of a pair of inclined portions 40, 45 and a groove-like recessed portion 50 formed between them. In this case, the accuracy of determining the homogeneous transformation matrix T in the second calibration step can be improved. Note that the same effect can be obtained when the calibration unit is made up of a pair of inclined portions and a protruding portion formed between them.

[0086] Furthermore, in this embodiment, in the first step of calibration, the 3D profiler 10 is moved by the moving means 20 so that the translational deviation amounts ΔX, ΔY, ΔZ and the rotational deviation amounts θX, θY, θZ between the measurement space A and the measurement coordinate space B approach zero, and measurement of the calibration jig 30 by the 3D profiler 10 is repeated.

[0087] According to the above aspect, since the first step of calibration has the above configuration, it is possible to reduce errors due to interference between the translational deviation amounts ΔX, ΔY, ΔZ and the rotational deviation amounts θX, θY, θZ (to converge the errors to 0), and to improve the accuracy when determining the homogeneous transformation matrix T.

[0088] Furthermore, when a plurality of 3D profilers 10 are used in a predetermined configuration as described above, that is, when a plurality of 3D profilers 10 are used and each 3D profiler 10 is movable by a moving means 20, and when all the 3D profilers 10 are configured to measure the same calibration jig 30 in the first step, the following effects are achieved.

[0089] In other words, in the first step of calibration, all 3D profilers 10 measure the same calibration jig 30, thereby improving the overall consistency and measurement accuracy of the measurement data of the calibration jig 30 by the 3D profilers 10 in the same measurement coordinate system.

[0090] (Second embodiment of the method for calibrating the measurement space) A second embodiment of the method for calibrating a measurement space according to the present invention is shown in Fig. 14. Note that parts that are essentially the same as those in the previous embodiment are given the same reference numerals and their description will be omitted.

[0091] In this second embodiment, as shown in FIG. 14, two jigs are used.

[0092] 14, the calibration jig is composed of two jigs, a first jig 30A and a second jig 30B, and the calibration jig is placed so that the reference point of the first jig 30A is located at the origin of the measurement space A, and the second jig 30B is placed at a distance of at least one-third of the length in the X direction of the area actually used in the measurement space A. Then, in the first step, the 3D profiler 10 is configured to measure both the first jig 30A and the second jig 30B. The first jig 30A and the second jig 30B have the same shape and dimensions.

[0093] In this embodiment, as shown in FIG. 14, the calibration jig is composed of two jigs, a first jig 30A and a second jig 30B, and the calibration jig is positioned so that the reference point of the first jig 30A is located at the origin of the measurement space A. The second jig 30B is positioned at a distance of at least 1 / 3 of the length in the X direction of the area actually used in the measurement space A (for example, the range of movement of the 3D profiler 10 by the moving means 20). In this embodiment, the 3D profiler 10 is configured to measure both the first jig 30A and the second jig 30B in the first step.

[0094] According to the above aspect, the calibration jig is composed of the first jig 30A and the second jig 30B and has the above layout, and the first step of calibration has the above configuration, so that the homogeneous transformation matrix can be corrected so as to improve the accuracy of the tilt angle at θY and the tilt angle at θZ in the homogeneous transformation matrix in particular.

[0095] Furthermore, in this embodiment, the moving means 20 is an articulated robot, and the first jig 30A is disposed so that a reference point of the first jig 30A is located at the origin of the measurement space A, and a first measurement coordinate system having the reference point of the first jig 30A as its origin, and a second measurement coordinate system having the reference point of the second jig 30B as its origin, are provided. Then, in a first step of calibration, either the first measurement coordinate system or the second measurement coordinate system is selected depending on the distance between the position of the 3D profiler 10 in the measurement space A and the installation position of the first jig 30A, or the distance between the position of the 3D profiler 10 in the measurement space A and the installation position of the second jig 30B.

[0096] According to the above aspect, in the first calibration step, it is possible to select either the first measurement coordinate system or the second measurement coordinate system depending on the distance between the position of the 3D profiler 10 in the measurement space A and the installation position of the first jig 30A, or the distance between the position of the 3D profiler 10 in the measurement space A and the installation position of the second jig 30B, thereby reducing the impact of positioning errors of the articulated robot on the accuracy of the three-dimensional measurement by the 3D profiler 10.

[0097] (Third embodiment of the method for calibrating the measurement space) A third embodiment of the method for calibrating a measurement space according to the present invention is shown in Fig. 15. Note that parts that are essentially the same as those in the previous embodiment are given the same reference numerals and their description will be omitted.

[0098] In this third embodiment, the shape of the calibration jig is different from that of the previous embodiments. That is, a calibration jig 30C in this third embodiment has a concave groove formed from a circular arc-shaped curved surface of a predetermined depth from a ceiling surface 36, which serves as a concave portion 50C. Note that this concave portion 50C is the same concave groove throughout the entire area in the X direction from the front surface 32 to the back surface 33 of the calibration jig 30C.

[0099] In the case of a calibration method using such a calibration jig 30C, the same effects as those of the calibration method of the above embodiment can be obtained.

[0100] (Fourth embodiment of the method for calibrating the measurement space) A fourth embodiment of the method for calibrating a measurement space according to the present invention is shown in Fig. 16. Note that parts that are essentially the same as those in the previous embodiment are given the same reference numerals and their description will be omitted.

[0101] In this fourth embodiment, the shape of the calibration jig is different from that of the previous embodiments. That is, a calibration jig 30D in this fourth embodiment has a generally V-shaped recessed groove formed therein, which is made up of a pair of left and right side surfaces 52, 53 that are inclined obliquely outward so as to approach each other from a ceiling surface 36, and this recessed groove forms a recessed portion 50D. Note that this recessed portion 50D is the same recessed groove throughout the entire area in the X direction from the front surface 32 to the back surface 33 of the calibration jig 30D.

[0102] In the case of a calibration method using such a calibration jig 30D, the same effects as those of the calibration method of the above embodiment can be obtained.

[0103] (Fifth embodiment of the method for calibrating the measurement space) A fifth embodiment of the method for calibrating a measurement space according to the present invention is shown in Figures 17 and 18. Note that parts that are essentially the same as those in the previous embodiment are given the same reference numerals and their description will be omitted.

[0104] In this fifth embodiment, the shape of the calibration jig is different from that of the previous embodiments. That is, in a calibration jig 30E in this fifth embodiment, a pair of inclined portions 40, 45 are formed line-symmetrically with respect to a line segment L2 (see FIG. 18) extending parallel to the Y direction, and a recessed portion 50 is formed over the entire area in the X direction. The inclined portions 40, 40 share a third side edge 43, and the inclined portions 45, 45 share a third side edge 48.

[0105] There are multiple 3D profilers 10. As shown in Fig. 18, this embodiment uses two 3D profilers 10. One 3D profiler 10 is disposed on the front surface 32 side of the calibration jig 30E, and the other 3D profiler 10 is disposed on the back surface 33 side of the calibration jig 30E, facing the one 3D profiler 10.

[0106] In addition, two or more 3D profilers may be placed, for example, on the front and back sides of the calibration jig (two or more 3D profilers in equal numbers on the front and back sides), or different numbers of 3D profilers may be placed on the front or back sides of the calibration jig.

[0107] In the case of a calibration method using such a calibration jig 30E, the same effects as those of the calibration method of the above embodiment can be obtained.

[0108] Furthermore, in this embodiment, as described above, multiple 3D profilers 10 (two in this case) are used, with one 3D profiler 10 being placed on the front 32 side of the calibration jig 30E and the other 3D profiler 10 being placed on the back 33 side of the calibration jig 30E, facing the one 3D profiler 10, so that the measurement space A can be quickly calibrated using the same calibration jig 30E.

[0109] (Sixth embodiment of the method for calibrating the measurement space) A sixth embodiment of the method for calibrating a measurement space according to the present invention is shown in Fig. 19. Note that parts that are essentially the same as those in the previous embodiments are given the same reference numerals and their description will be omitted.

[0110] The calibration jig 30F of the sixth embodiment is formed with a pair of inclined portions 40, 45 similar to the calibration jig 30E of the fifth embodiment, and is also formed with a recessed portion 50D which is an approximately V-shaped recessed groove similar to the calibration jig 30D of the fourth embodiment.

[0111] In the case of a calibration method using such a calibration jig 30F, the same effects as those of the calibration method of the above embodiment can be obtained.

[0112] Furthermore, the present invention is not limited to the above-described embodiments, and various modified embodiments are possible within the scope of the gist of the present invention, and such embodiments are also included in the scope of the present invention. [Explanation of symbols]

[0113] 10 3D profiler, 20 moving means, 30, 30A, 30B, 30C, 30D, 30E, 30F calibration jig, 40, 45 inclined portion, 50, 50C, 50D, A measurement space, A1 measurement coordinate, B measurement coordinate space, B1 measurement coordinate.

Claims

1. A method for calibrating a measurement space, when performing three-dimensional measurement to measure a three-dimensional shape of a measurement object using a 3D profiler used in a light-section method and a moving means for relatively moving the 3D profiler and a measurement object placed in a measurement space, by placing a calibration jig in the measurement space and measuring the calibration jig with the 3D profiler, where translational deviation amounts in the X direction, Y direction, and Z direction in an XYZ three-dimensional coordinate system between the measurement space and the measurement coordinate space measured by the 3D profiler are ΔX, ΔY, and ΔZ, respectively, and rotational deviation amounts about axes along the respective directions as centers of rotation are θX, θY, and θZ, respectively, the calibration jig has a plurality of calibration portions in which the length dimensions and tilt angles of a cross-sectional shape obtained by a light-section method change with respect to ΔX, ΔY, ΔZ, θX, θY, and θZ, and the shape and dimensions including the calibration portions are known, The calibration of the measurement space is a first step of arranging the calibration jig in the measurement space so that a reference point of the calibration jig is positioned at the origin of the measurement space, and measuring the calibration jig by a light section method to determine a length dimension and a tilt angle of one cross-sectional shape of the calibration jig; a second step of calculating, from the length dimension and tilt angle of one cross-sectional shape of the calibration jig calculated in the first step, measurement coordinates of the measurement coordinate space including the 3D profiler and the moving means, and a homogeneous transformation matrix between the measurement coordinates of the measurement space to be calibrated.

2. The calibration unit a pair of inclined portions formed at an upper end portion in the Z direction and on both sides in the Y direction, each having an inclined surface shape that is inclined at a predetermined angle in the Z direction with respect to a ceiling surface of the calibration jig in a front view and that is inclined at a predetermined angle in the X direction with respect to a front surface of the calibration jig in a plan view; 2. The method for calibrating a measurement space according to claim 1, further comprising: a concave portion formed at the upper end in the Z direction between a pair of the inclined portions, which is recessed relative to the ceiling surface in a front view and has a groove-like shape extending in the X direction in a planar view; or a convex portion formed at the upper end in the Z direction between a pair of the inclined portions, which is convex relative to the ceiling surface in a front view and has a convex shape extending in the X direction in a planar view.

3. 3. The method for calibrating a measurement space according to claim 1, wherein in the first step, the 3D profiler is moved by the moving means so as to bring each translational deviation amount and each rotational deviation amount between the measurement space and the measurement coordinate space closer to zero, and measurement of the calibration jig by the 3D profiler is repeated.

4. the calibration jig comprises a first jig and a second jig; Positioning the calibration jig so that a reference point of the first jig is located at the origin of the measurement space; the second jig is disposed at a distance of at least one-third of the length in the X direction of an area actually used in the measurement space, 3. The method for calibrating a measurement space according to claim 1, wherein in the first step, both the first jig and the second jig are measured by the 3D profiler.

5. a plurality of the 3D profilers are provided, and each of the 3D profilers is movable by the moving means; 3. The method for calibrating a measurement space according to claim 1, wherein in the first step, all of the 3D profilers measure the same calibration jig.

6. the calibration jig has a pair of inclined portions formed line-symmetrically with respect to a line segment extending parallel to the Y direction, and the concave portion or the convex portion formed over the entire area in the X direction; 3. The method for calibrating a measurement space according to claim 2, wherein a plurality of 3D profilers are provided, a predetermined 3D profiler is disposed on the front side of the calibration jig, and other 3D profilers are disposed on the back side of the calibration jig in an orientation facing the predetermined 3D profiler.

7. the moving means is an articulated robot, the calibration jig comprises a first jig and a second jig; Positioning the first jig so that a reference point of the first jig is located at the origin of the measurement space; a first measurement coordinate system having an origin at a reference point of the first jig and a second measurement coordinate system having an origin at a reference point of the second jig; 3. The method for calibrating a measurement space according to claim 1, wherein in the first step, either the first measurement coordinate system or the second measurement coordinate system is selected depending on a distance between a position of the 3D profiler in the measurement space and a position where the first jig is installed, or a distance between a position of the 3D profiler in the measurement space and a position where the second jig is installed.

Images