JP2902285B2 - 3D position control system - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a position control system using a three-dimensional measuring device, and more particularly to a three-dimensional position control system in which a preliminary registration process of a probe vector for correcting a spatial error is improved.

[0002]

2. Description of the Related Art Spatial errors (geometric errors) of a coordinate measuring machine include parallel errors such as axis scale errors and straightness errors in a rectangular coordinate system in FIG. 4 and pitch errors and yawing errors. There is a rotation error. The types and numbers of these spatial errors are as follows. [Parallel error] Each axis scale error: 3 Each axis horizontal straightness error: 3 Each axis vertical straightness error: 3 [Rotation error] Each axis pitching error: 3 Each axis yawing error: 3 Each axis rolling error: 3 Right angle error between each axis: 3

The calculation of the error correction amount δ x is performed based on the following equation (1). In equation (1), the underline means a vector expression, ei is the parallel error at each axis fulcrum (also the center of rotation), θi is the rotation error at each axis fulcrum, and Ci A is from each axis fulcrum Ci to the probe tip A. A heading vector, Ci B is a vector heading from each shaft fulcrum Ci to the center B of the lower end of the Z-axis spindle, P is a probe vector heading from B to A, n is the number of fulcrum points, and "x" is a vector cross product.

[0004]

(Equation 1) Ci A = Ci B + P

The reference position for correcting the spatial error is determined by, for example, the center position B of the lower end of the z-axis spindle 21 as shown in FIG. It is. Therefore, each time a probe to be actually measured is determined, it is necessary to correct the positional relationship before and after the probe attitude is changed by measurement using the reference sphere each time, and to specify a probe vector for a new attitude. The correction of the positional relationship by changing the posture of the probe is necessary to handle the case where the probe is in the vertical posture and the case where the probe is in the horizontal posture as if they were measured using the same probe. That is, the center position of the reference sphere was measured in advance when the probe was in the vertical posture, the center position of the reference sphere was also measured when the posture was changed, and the difference between the two was used as the correction amount, so that the measurement was performed with the same probe. The same result will be obtained. On the other hand, the probe vector is represented by a vector P from the reference position B to the tip position A of the probe 22 as shown in FIG. Must be specified in advance.

As for the probe vector designation, for example, when the posture of the probe is changed, it is of course possible for a person to measure the probe vectors for three axes each time with a mono-sashi and to manually input the keys by hand, but this completely automates the measurement. Can not. Therefore, it is desired that the probe vector be automatically calculated at the same time as the probe tip position measurement using the reference sphere 23, which is also required for normal measurement, and that the probe vector be pre-registered.

The basic expression of the probe vector P is a subtraction formula of P = [reference sphere center coordinate measured at point B] − [reference sphere center coordinate measured at point A] (1). The reference sphere center coordinate is usually 4
It is obtained by a sphere calculation process based on measurement of points or more. If a probe whose standard state such as type, orientation, probe, angle, etc. is known in advance (this is referred to as a probe initial configuration), the probe vector is defined as P0 and the above equation (1) is transformed as follows. Is done. [Reference sphere center coordinate measured at point B] = P0 + [reference sphere center coordinate measured at point A] (2) Therefore, if the reference sphere measurement is performed only once according to the initial configuration of the probe, measurement is performed at point B. Reference sphere center coordinates can be registered. Thereafter, no matter what kind of probe is used, if the reference sphere measurement is performed at the point A, the probe vector can be automatically calculated based on the equation (1).

[0008]

When the probe vector is automatically calculated by equation (1) when the probe is changed, the probe vector used when measuring four points of the reference sphere at point A is a new probe vector. Has not yet been determined, the previous old probe vector will be used, and this will cause an error. To solve this, after the new probe vector is determined, the reference sphere measurement using the new probe vector may be performed again. However, in this case, the reference sphere measurement must be performed twice each time the probe is changed.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a three-dimensional position control system having a probe vector designating means that simplifies a reference sphere measuring procedure.

[0010]

SUMMARY OF THE INVENTION The present invention comprises a three-dimensional measuring machine to which a replaceable and rotatable probe is mounted, and a center coordinate of a reference sphere measured at a reference position of the measuring machine for spatial error correction. in the three-dimensional position control system obtains the probe vector by subtracting the reference sphere center coordinates measured by the probe tip position, which is the pre-registration from changes of
Pre-register new probe vector corresponding to the selected probe
The old probe base corresponding to the probe before the change.
From the coordinate position of the center of the reference sphere measured using
Reference sphere center coordinates to be measured using lobe vector
Calculating means for calculating the position, wherein the calculating means
When measuring the reference sphere center coordinates using the lobe vector
The used position data is converted to a space for correcting the spatial error.
The initial approximate solution of the position variable of the error correction function
Convert the similar solution to the spatial error function using the old probe vector
To obtain a spatial error, and this spatial error is calculated as the position data.
Data to obtain the position data correction value.
The difference value obtained by subtracting the data correction value from the initial approximate solution is a predetermined value.
It is determined whether the convergence condition is satisfied.
If not, add the difference value to the initial approximate solution and
Similar iterative operation until the convergence condition is satisfied as an approximate solution
Is equivalent to measuring with the new probe vector
Is calculated .

[0011]

According to the present invention, the reference sphere measurement is omitted once by automatically calculating the reference sphere center coordinates using the new probe vector in the preliminary registration of the probe vector at the time of probe replacement using a simple inverse correction operation. can do. In other words, the probe vector designation processing required for spatial error correction can be incorporated into the reference sphere measurement processing for position correction before and after the change of the probe attitude, which is conventionally required, with the same number of steps as before. By specifying the probe vector, error correction at the probe tip becomes possible.

[0012]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows a system configuration of an embodiment. The coordinate measuring machine 11 is connected to a measuring machine control device 12, and the control device 12 is connected to a host computer 13. As shown in FIG. 2, a rotatable probe 22 can be attached to the lower end of the z-axis spindle 21 at the tip of the coordinate measuring machine 11. As described above, the parameter of the spatial error correction is a value at the reference position B, and the probe vector P is necessary to obtain the correction amount at the probe tip position A.

In this embodiment, when a probe to be used is determined, the probe vector P is automatically designated and pre-registered by four-point measurement using the reference sphere 23 instead of key input. The basic principle is as described above. The information of the point by the reference sphere measurement is sent to the host computer 13, where the calculation (sphere calculation) for obtaining the center coordinates and the probe vector calculation are performed. Further, in this case, in this embodiment, the correct reference sphere based on the new probe vector is used without performing the reference sphere measurement using the new probe vector from the reference sphere center coordinates measured at the probe tip position A using the old probe vector. Calculation means for automatically determining the center coordinate position is incorporated.

Specifically, the calculation means includes a spatial error correction function δ (x) = f including the probe vector P as a variable.
It is realized by an algorithm as shown in FIG. 3 using (x ′, P). Xt, x ', P, δx, x used in FIG.
c and Δx are all vectors. Note that the spatial error correction process needs to be applied to the trajectory accuracy at the time of movement control, and therefore needs to be performed by the CMMC 12. Therefore, the main inverse correction calculation in FIG. 3 also needs to be performed by the CMMC. The input 4-point position data xt used for obtaining the coordinates of the reference sphere center position measured at the probe tip position A is transferred from the host computer to the CMMC together with a new probe vector (new P) (S0). Then, using the four-point position data xt as an initial approximate solution of the position variable x '(S1),
The spatial error f (x ', old P) is calculated using this and the old probe vector (old P) (S2). Next, the error is added to the value before correction to calculate a correction value xc = x '+ δx (S3). Next, this correction value xc is converted to an initial approximate solution x
Δx is obtained by subtracting from t (S4), and a convergence determination is made as to whether this Δx is smaller than a convergence determination constant K (S4).
Five). More specifically, for three axes x, y, and z, |
Δx | ≦ k, | Δy | ≦ k, and | Δz | ≦ k are convergence conditions here.

If the convergence condition is not satisfied, .DELTA.x is added to the initial approximate solution (S6), and the process returns to step S2 to repeat the same operation until the convergence condition is satisfied. After performing a simple inverse correction by such an iterative method to obtain a value before correction, a new error vector is used to calculate a spatial error δx (S7), and a calculation is performed to obtain a correction value xc (S8). . Then, the host computer receives the four-point data from the CMMC and performs a sphere operation at the four points (S9).
As a result, the correct reference sphere center coordinates at the probe tip position A equivalent to the measurement using the new probe vector can be obtained without performing the reference sphere measurement.

[0016]

As described above, according to the present invention, the reference sphere measurement can be performed by automatically calculating the reference sphere center coordinates using the new probe vector in the preliminary registration of the probe vector using the simple inverse correction operation. A three-dimensional position control system can be obtained in which the probe vector can be automatically specified in the conventional procedure by omitting once and error correction at the probe tip can be performed.

[Brief description of the drawings]

FIG. 1 shows a system configuration according to an embodiment of the present invention.

FIG. 2 is a diagram for explaining probe tip error correction.

FIG. 3 shows an algorithm for probe vector pre-registration of the embodiment.

FIG. 4 shows a kinematic chain representation of a coordinate measuring machine.

[Explanation of symbols]

11: Coordinate measuring machine, 12: Control device, 13: Host computer, 21: Spindle, 22: Probe, 23
... Reference sphere, A: Probe tip position, B: Reference position.

Claims (2)

(57) [Claims] 1. A three-dimensional measuring machine to which a replaceable and rotatable probe is attached.
The three-dimensional position control system that obtains the probe vector by subtracting the reference sphere center coordinate measured at the probe tip position from the reference sphere center coordinate measured at the reference position of the measuring device and pre-registers the probe vector corresponds to the changed probe. New probe vector
The old profile corresponding to the probe before the change
Is the reference sphere center coordinate position measured using the motion vector
In the reference sphere to be measured using the new probe vector
Calculating means for calculating the center coordinate position, wherein the calculating means uses the old probe vector in the reference sphere.
The position data used in measuring the center coordinates is stored in the space
Near initial position variable of spatial error correction function for error correction
Using the old probe vector as the similar solution and this initial approximate solution
The spatial error function is given to the spatial error function,
Add the position error to the position data and correct the position data
And subtract the position data correction value from the initial approximate solution.
Determines whether the calculated difference value satisfies the predetermined convergence condition
If the convergence condition is not satisfied,
The difference value is added and the following approximate solution is satisfied.
The same iterative operation is performed until the new probe vector
A three-dimensional position control system, which calculates a reference sphere center coordinate equivalent to that measured by using a sphere . 2. An error correction amount δ for correcting the spatial error.
The three-dimensional position control system according to claim 1, wherein the calculation of x is performed based on the following equation. Record Ci A = Ci B + P where underline denotes a vector representation, ei parallel error at each pivot point, .theta.i rotation error at each pivot point, Ci A is directed to the probe tip A from the pivot point Ci vector , Ci B are each fulcrum C
a vector from i to the center B of the lower end of the Z-axis spindle, P
Is the probe vector from B to A, n is the number of fulcrums, and "x" is the vector cross product.