JP2009063541A - Geometric quantity measurement method and device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a geometric quantity measurement method capable of reducing errors due to deviation of setting position of sensors in a sensor head and relative inclination between the sensor head and a measurement object. <P>SOLUTION: The geometric quantity at the surface of the measurement object is measured by relatively moving the sensor head provided with at least 3 sensors along the measurement object using the detected signal from each sensor. Preliminarily measuring a straightness to the reference plane 30, the deviation of the center sensor in a measurement direction is calculated from the segment connecting the reference points of both the sensors, and the relation between the relative inclination and deviation is made into a correction table as pair data. At the measurement of geometric quantity, the relative inclination between the measuring object and the sensor head is calculated, and the calculated deviation is taken out while referring the correction table. The geometric quantity of the surface of the measurement object is operated using the deviation taken out and the detection signal obtained from the 3 sensors. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a method and apparatus for measuring geometric quantities such as straightness and flatness, and in particular, calibration for measuring geometric quantities using three or more sensors to measure the geometric quantities of the surface of a measurement object. The present invention relates to a method and a geometric quantity measuring apparatus having the calibration function.

  With reference to FIG. 5, an example of a straightness measurement method will be described as an example of geometric amount measurement. For this type of measurement, a sensor head SH having a configuration in which three sensors I, J, and K for detecting the distance to the measurement target OB are arranged in a straight line is used. The sensor head SH is moved along the measurement target and performs measurement at a fixed interval timing (or a fixed interval moving distance). The detected distance values obtained by the sensors I, J, and K are i, j, and k, respectively. The detection values i, j, and k are distances from the reference points (indicated by black circles) of the sensors I, J, and K to the measurement target, respectively. As shown in FIG. 5, if there are irregularities on the measurement target surface, differences occur in the detection values i, j, and k, and the straightness can be measured from these differences. According to such a three-point measurement, even if the relative positional relationship between the sensor head SH and the measurement target OB changes during the movement of the sensor head SH, that is, even when the sensor head SH approaches or separates, it is affected. There is no.

  The above description is based on the assumption that the three sensors I, J, and K are arranged on a straight line. However, in the installation of the three sensors I, J, K, it is inevitable that a minute shift occurs.

  FIG. 6 exaggerates an example in which the sensor J is misplaced among the three sensors I, J, and K. The deviation amount (offset) of the reference point of the central sensor J from the line segment connecting the reference points of the sensors I and K on both sides of the three sensors is denoted by δ, and this is hereinafter referred to as an installation deviation amount. Since this installation deviation amount δ is a mechanical deviation amount, it does not change. Therefore, if the installation deviation amount δ can be known by some method, the value w that is the amount representing the local shape of the measurement target OB can be known together with the detected values i, j, and k. Then, while moving the sensor head SH, the value w is obtained at an equal interval with the inter-sensor pitch P and accumulated, and the surface shape of the measurement object OB can be known by setting the measurement points at respective grid intersections. .

  In the case of FIG. 6, the installation deviation amount δ can be calculated by the following equation (1).

δ = j− (i + k) / 2 (1)
However, the non-contact sensor head SH as described above is affected by the relative inclination between the sensor head SH and the measurement target OB. That is, if a relative inclination occurs between the sensor head SH and the measurement target OB during the movement of the sensor head SH, the detected value fluctuates slightly compared to the case where there is no relative inclination. The installation deviation amount δ uniquely determined by 1) cannot be used as it is. This is because, when there is a relative inclination between the sensor head SH and the measurement object OB, the acquired installation deviation amount includes a variation due to the relative inclination, which becomes an error factor, which is high. It means that measurement with accuracy cannot be maintained.

  An example of a detector calibration method for measuring straightness is disclosed in Patent Document 1. In this calibration method, it is necessary to rotate a disk having a sufficiently large diameter with respect to the sensor pitch as a calibration target with good reproducibility, and a large-scale and precise device is required. Further, the sensor installation deviation amount (offset error) is determined uniquely by averaging fluctuations during one rotation of the disk, and high accuracy during straightness measurement cannot be expected.

JP 2005-308703 A

  In the present invention, even if the sensor in the sensor head has an installation deviation amount, and there is a case where a relative inclination occurs between the sensor head and the measurement object during measurement, the error caused by the installation deviation amount and the relative inclination. It is an object of the present invention to provide a geometric amount measuring method and a geometric amount measuring apparatus that can eliminate the problem.

  According to the present invention, the relative inclination between the sensor head and the measurement target at the time of measurement is easily obtained by two sensors on both sides among at least three sensors installed on the sensor head, and the obtained relative inclination is determined. By dynamically changing the value of the installation deviation amount of the sensor used for measurement (calculation), that is, by correcting the fluctuation amount of the installation deviation amount used for measurement (calculation) according to the acquired relative inclination, The measurement accuracy is improved.

  In the geometric amount measuring method according to the present invention, a sensor head in which at least three sensors for detecting the distance to the measurement target surface are arranged with an offset along the traveling direction is relatively moved to the measurement target surface. Thus, the geometric amount of the measurement target surface is measured using detection signals obtained from the at least three sensors.

  In the geometric amount measuring method according to a preferred aspect of the present invention, measurement is performed in advance on a reference plane having a flat surface, and the center of the at least three sensors from the line segment connecting the reference points of the sensors on both sides is measured. By calculating the deviation amount of the reference point of the sensor and the deviation amount in the measurement direction from the detection values of the three sensors, and changing the relative inclination between the reference plane and the sensor head, A correction table is created by collecting the relationship between the calculated amount of deviation as a pair of data, and the measurement target surface and the sensor are measured each time when measuring the geometric amount for the measurement target. The relative inclination between the head and the head is calculated from the detected values of the sensors on both sides and the distance between them, and the correction table is referred to based on the calculated relative inclination. The calculated shift amount is removed and calculating the geometric volume of the measurement object surface by using a detection signal obtained from the the shift amount extracted the at least three sensors Te.

  In the geometric amount measuring method according to the present invention, when the correction table is created, the relative inclination between the reference plane and the sensor head is set with an axis orthogonal to the arrangement direction of the at least three sensors as a central axis. The reference plane or the sensor head is preferably changed by inclining.

  According to the present invention, there is also provided a sensor head in which at least three sensors for detecting the distance to the measurement target surface are arranged with an offset along the traveling direction, and the sensor head is relative to the measurement target surface. There is provided a geometric quantity measuring apparatus comprising a moving means for moving and a computing means for computing a geometric quantity of the measurement target surface using detection signals obtained from the at least three sensors.

  In the geometric amount measuring apparatus according to the present invention, the calculation means is a deviation amount of a reference point of a central sensor from a line segment connecting reference points of sensors on both sides of the at least three sensors, and relates to a measurement direction. The amount of deviation is calculated from the detection values of the three sensors, and between the relative inclination between the reference plane by the flat surface and the sensor head, and the calculated amount of deviation that changes by changing this. The correction means is created by collecting the relationship of the data as a pair of data, and the calculation means also has a relative inclination between the measurement target surface and the sensor head each time the geometric amount is measured with respect to the measurement target. Is calculated from the detection values of the sensors on both sides and the interval between them, and the calculated deviation amount is extracted by referring to the correction table based on the calculated relative inclination. Characterized by calculating the geometry of the measurement object surface by using a detection signal obtained from the out shift amount between the at least three sensors.

  According to the present invention, the installation displacement amount and the relative inclination of the sensor are affected by dynamically changing the calculated installation displacement amount according to the relative inclination between the sensor head and the measurement target at the time of measurement. It is possible to achieve high-accuracy geometric quantity measurement that will not be received.

  An embodiment of the present invention will be described with reference to FIGS.

  FIG. 1 is an enlarged view showing only the relationship between a sensor head and a measurement object in a geometric quantity measuring apparatus according to the present invention. In particular, it is exaggerated that the sensor has an installation deviation amount (offset), and that a relative inclination occurs between the sensor head and the measurement object during measurement. Here, as in FIG. 5, three sensors I, J, and K are installed in the sensor head SH, and the deviation amount of the reference point of the central sensor J from the line segment connecting the reference points of the sensors I and K on both sides, In particular, the amount of deviation in the measurement direction is δ. In addition, the relative inclination between the line segment connecting the reference points of the sensors I and K and the measurement target OB is θ. The detection values of sensors I, J, and K are i, j, and k.

  For the sensor head SH as shown in FIG. 1, the following calibration work is performed by the following straightness measurement using the tilt stage and the reference plane shown in FIG. 2 before actual measurement. It is preferable that the tilt stage and the reference plane are installed on the measurement object side to calibrate the sensor. Since the sensor position error changes under various conditions such as temperature and humidity, it is desirable to calibrate periodically. In FIG. 2, the tilt stage 20 can tilt a table 21 on which a reference plane 30 having a flat surface on the upper surface is mounted with one axis as a central axis. Here, the one axis is set to be an axis orthogonal to the direction in which the three sensors I, J, and K in the sensor head SH are arranged. It is assumed that the three sensors I, J, and K are installed at an equal pitch P with respect to a line segment that connects the reference points of the sensors I and K.

  In the case of FIG. 2, since the measurement target is the reference plane 30, the local shape w of the measurement target OB as shown in FIG. 1 is always zero. When the relative inclination between the measurement target (reference plane 30) and the sensor head SH is θ, the relative inclination θ is expressed by the following equation (1) using the detection values i and k of the sensors I and K on both sides and the pitch P between the sensors. 2).

tan θ = (ki) / 2P (2)
Further, the installation deviation amount δ of the sensor J is calculated by using the detected values i, j, and k of the sensors I, J, and K as the above-described equation (1), that is, δ = j− (i + k) / 2. Can be calculated.

In the calibration operation, the relative inclination θ is sequentially changed by the inclination stage 20, and the relative inclination θ is measured and the installation deviation amount δ is calculated each time. Then, by repeating this n times (where n is a positive integer), the relationship between the relative inclination and the calculated amount of installation deviation is expressed by n pairs of calibration data (θ ₁ , δ ₁ ),. , Δi) (i indicates i-th)..., (Θn, δn), and a correction table is created. FIG. 2 shows that the relative inclination θ when the reference plane 30 is inclined clockwise is positive.

  The relationship between the relative inclination acquired in this way and the calculated installation deviation amount is not affected by other factors because the measurement target is the reference plane, and the calculated installation deviation amount is the relative inclination. This is a value that has already been factored in. In other words, it can be said that the calibration amount of the installation deviation amount is dynamically changed according to the relative inclination between the sensor head and the measurement target.

  The correction table created in the above calibration work is stored in a storage device provided in a control device that also functions as a calculation means in the geometric amount measuring apparatus.

  FIG. 3A shows one form of straightness measurement by the geometric amount measuring apparatus. The measurement object OB is mounted on a table 41 that can be conveyed in the front and back direction of the drawing (hereinafter referred to as the length direction). Above the measurement object OB, it can be moved along the guide 42 in the left-right direction of the drawing, that is, in the direction crossing the measurement object OB (hereinafter referred to as the width direction), and in the vertical direction of the drawing, ie, the measurement object OB. A traveling portion (moving means) 43 that can move in a direction perpendicular to the traveling portion 43 is disposed, and a sensor head SH is attached to the traveling portion 43.

  The following operations are executed under the control of the control device.

  In measuring the straightness, the sensor head SH is moved in the width direction from one end side to the other end side of the measurement target OB. And while moving, measurement with three sensors I, J, and K is performed at regular intervals, and detection values i, j, and k are obtained. Here, as described with reference to FIG. 1, the center sensor J has an installation displacement amount δ, and the fact that there is a relative inclination θ between the sensor head SH and the measurement target OB is affected by the above. That's right. For this reason, the relative inclination θi is calculated by the above equation (2), and the detected value j from the sensor J is used by using the installation deviation amount δi read by referring to the correction table based on the calculated relative inclination θi. The distance to the measurement target OB is calculated as (j−δi). The value w is obtained using the values i, (j−δi), and k thus obtained.

  The above measurement is repeated at regular intervals until the sensor head SH reaches the other end side from the one end side of the measurement target OB, and the straightness in the width direction of the measurement target OB is measured by accumulating the value w. can do. Subsequently, the measurement target OB is moved by a certain distance in the length direction by the table 41, and then the above measurement is performed.

  By repeating the measurement as described above, the measurement object OB can be measured at a grid-like intersection as shown by the black dots in FIG. 3B, and measurement is performed using these measurement results. The flatness of the target OB can be measured. In FIG. 3B, the dot interval is increased for easy understanding, but this interval is actually small.

  In FIG. 1, the center sensor J has an installation deviation amount δ. However, if the sensors J and K are both on the same straight line and the sensor I is installed with a deviation from the straight line in the measurement direction. Even so, the deviation amount of the reference point of the central sensor J from the line connecting the reference points of the sensors I and K on both sides is defined as the installation deviation amount δ.

  FIG. 4 shows an example of calibration data actually collected by the above measurement. Triangulation laser displacement meters were used for the sensors I, J, and K, and an optical flat was used for the reference plane 30. A typical example of the tilt stage 20 is a so-called goniometer. Due to the relative inclination between the sensor head and the measurement object, a clear variation tendency can be seen in the collected installation deviation values. As a result, in the case of measuring the straightness, the relative inclination between the sensor head and the measurement target is calculated each time when measuring at a constant interval, and the value of the installation deviation amount corresponding to the relative inclination is calculated in the correction table. By determining with reference and performing correction, it is possible to realize highly accurate measurement as compared with the method using the installation deviation amount uniquely determined from the above-described equation (1). This can be realized no matter how the posture or the interval with respect to the measurement object changes as the sensor head undulates or tilts during the movement.

  In the above description, the case where the reference plane is inclined has been described. However, the sensor head may be inclined. Moreover, although the case where there are three sensors has been described, four or more sensors are also applicable.

  The present invention is suitable for measuring geometric quantities such as straightness and flatness of an object to be ground in combination with an ultra-precision surface grinder for tables such as stone and iron that require ultra-precision finishing. ing.

FIG. 1 is an enlarged view showing only the relationship between a sensor head and a measurement object in a geometric quantity measuring apparatus according to the present invention. FIG. 2 is a diagram for explaining a calibration operation performed before geometric amount measurement according to the present invention, and is a diagram showing a relationship between the tilt stage and the reference plane and the sensor head used in the calibration operation. It is. FIG. 3A shows one form of straightness measurement by the geometric quantity measuring device, and FIG. 3B explains flatness measurement performed using the straightness measurement result of FIG. 3A. FIG. FIG. 4 shows an example of calibration data actually collected by geometric amount measurement according to the present invention. FIG. 5 is a diagram for explaining an example of a straightness measurement method as an example of geometric amount measurement. FIG. 6 exaggerates an example in which one of the three sensors in the sensor head shown in FIG. 5 is misplaced.

Explanation of symbols

I, J, K Sensor SH Sensor head OB Measurement object 20 Inclined stage 21 Table 30 Reference plane 41 Table 42 Guide 43 Traveling section

Claims (3)

Obtained from the at least three sensors by moving relative to the measurement target surface a sensor head in which at least three sensors for detecting the distance to the measurement target surface are arranged with an offset along the traveling direction. In the method of measuring the geometric amount of the measurement target surface using the detected signal,
Measurement is performed in advance on a reference plane having a flat surface, and the deviation of the reference point of the central sensor from the line segment connecting the reference points of the sensors on both sides of the at least three sensors, and the deviation in the measurement direction The amount is calculated from the detection values of the three sensors, and the relationship between the relative inclination between the reference plane and the sensor head and the calculated deviation amount that changes by changing the relative inclination is obtained. Create a correction table by collecting data as pairs,
When measuring the geometric amount for the measurement target, each time the measurement is performed, the relative inclination between the measurement target surface and the sensor head is calculated from the detection values of the sensors on both sides and the interval therebetween, and the calculation is performed. The calculated deviation amount is extracted with reference to the correction table based on the relative inclination, and the geometric amount of the measurement target surface is obtained using the extracted deviation amount and detection signals obtained from the at least three sensors. A geometric amount measuring method characterized by calculating In creating the correction table, the reference plane or the sensor head is inclined with respect to a relative inclination between the reference plane and the sensor head, with an axis perpendicular to the arrangement direction of the at least three sensors as a central axis. The geometric amount measuring method according to claim 1, wherein the geometric amount measuring method is changed. A sensor head in which at least three sensors for detecting the distance to the measurement target surface are arranged with an offset along the traveling direction; a moving means for relatively moving the sensor head to the measurement target surface; In a geometric quantity measuring apparatus comprising a calculation means for calculating a geometric quantity of the surface to be measured using detection signals obtained from at least three sensors,
The arithmetic means is a deviation amount of a reference point of a central sensor from a line segment connecting reference points of sensors on both sides of the at least three sensors, and a deviation amount relating to a measurement direction is detected by the three sensors. In addition, the relationship between the relative inclination between the reference plane by the flat surface and the sensor head and the calculated deviation amount that changes by changing the relative inclination is collected and created as a pair of data. Correction table,
The calculation means also calculates the relative inclination between the measurement target surface and the sensor head from the detection values of the sensors on both sides and the distance between the measurement target surface and the sensor head each time the geometric amount is measured with respect to the measurement target. The calculated deviation amount is extracted with reference to the correction table based on the calculated relative inclination, and the geometry of the measurement target surface is obtained using the extracted deviation amount and the detection signals obtained from the at least three sensors. A geometrical amount measuring apparatus characterized by calculating a stoichiometric amount.

Images