JP6608726B2 - Positioning measurement system - Google Patents

Description

  The present invention relates to a positioning measurement system that measures the position of a surface that serves as a reference for positioning a workpiece that is a workpiece, and detects whether the position of the surface has reached a target position.

  Conventionally, measuring machines such as a three-dimensional coordinate measuring machine are known as measuring machines for measuring the shape and dimensions of an object to be measured. Such a measuring machine is provided with a touch probe for detecting contact with an object to be measured as described in Patent Document 1 in order to perform coordinate detection and position detection.

  In a machining center (machine tool), a touch sensor is mounted on the spindle after machining, the work and the touch sensor (touch probe) are moved relative to each other, and the touch sensor is brought into contact with the work surface. Measuring the shape is described in, for example, Patent Document 2.

  In such a machine tool, the surface shape of the workpiece is measured by detecting a position where the touch sensor is in contact with the workpiece (and / or a position away from the workpiece). Furthermore, it is also known to detect whether or not the tool is broken according to the position at which the tool contacts the touch sensor by moving the tool and touching the touch sensor.

  In addition, when measuring a workpiece with a conventional touch probe, it is possible to detect the contact of the probe with the object to be measured by electrical continuity, or the trigger method to detect the structure in which the contact is separated by the probe's contact with the object to be measured. A certain contact sensor type or a displacement sensor type using a displacement amount of the probe due to contact has been used.

JP-A-2015-127713 JP 2013-6257 A

  In the above prior art, since the contact sensor type uses a mechanical contact switch, chattering occurs, contact deterioration occurs, and the contact has a lifetime. Further, advanced processing such as chattering countermeasures and filter processing is difficult, and it is difficult to protect against external electrical stress (such as surge and spark noise). Further, in a machine tool, in order to detect whether or not a tool is broken, it is necessary to have high sensitivity not only in the X and Y directions but also in all the X, Y, and Z directions. .

  On the other hand, since the displacement sensor type does not use a mechanical contact switch, chattering, contact deterioration, and contact life problems do not occur, and continuous position detection is possible. Accuracy can be achieved. However, it is necessary to replace the signal output from the sensor with information in a format that can be effectively used by the machine tool, and an electric circuit or the like for performing the processing is necessarily required, which causes an increase in cost and size.

  The object of the present invention is to solve the above-mentioned problems of the prior art and is sufficient for high-speed movement of a workpiece without being affected by chattering, contact deterioration, contact life, and external electrical stress caused by a mechanical contact switch. Correspondingly, the calculation processing is extremely simple and the positioning measurement system is highly reliable.

  To achieve the above object, the present invention provides a stylus that contacts a reference surface in a positioning measurement system that measures the position of a reference surface of a workpiece and detects whether the position of the reference surface has reached the target position. And a touch probe having a displacement sensor for detecting the position as numerical data, and sampling and detecting the numerical data at N points as past position information using the touch probe, A linear approximation calculation circuit obtained by calculating a position of the next sampling timing as a predicted value by linear approximation by multiplication, and information on whether the predicted value obtained by the linear approximation calculation circuit has reached the target position And a comparison circuit for outputting.

  In the above, it is desirable that the sampling intervals for detecting the numerical data at the N points are equal intervals, and that the calculation process is performed with a time point at which the time is to be predicted as a reference (t = 0).

  Further, it is preferable that the linear approximation calculation circuit and the comparison circuit are processed by a digital circuit having a delay element, an adder, and a multiplier of a constant multiple as components.

  Furthermore, it is desirable that the numerical data is adjusted to be 0 at the target position.

  Further, it is desirable that the straight line approximation calculation circuit performs the arithmetic processing with the absolute value omitted.

  Further, it is preferable that the linear approximation calculation circuit and the comparison circuit are processed by a digital circuit integrated into one.

  Furthermore, it is desirable that the linear approximation calculation circuit and the comparison circuit are configured to initialize the output value of the component to 0 before the first sampling is performed.

  The N is a power of 2 and N = 2λ (λ is a natural number), and it is preferable that the multiplication processing is replaced with a bit shift.

  The touch probe is a displacement-type sensor in which an LVDT sensor is housed in a cover. The touch probe is a hemispherical hemispherical portion that is the upper end of a stylus and is pivotally supported by the cover, and is slidable vertically in the cover. A displacement signal is output according to the relative position of the movable ring, the magnetic core fixed to the movable ring and moving in the vertical direction by the inclination and vertical movement of the hemisphere, and the magnetic core fixed to the cover. It is desirable to have a coil portion that performs.

  The linear approximation calculation circuit and the comparison circuit are preferably processed by a program by a CPU.

  According to the present invention, the numerical data of N points are sampled and detected as past position information, and the workpiece has reached the target position at the next sampling timing position based on the linear approximation by the least square method from the numerical data. Therefore, it is possible to obtain a positioning measurement system that can sufficiently cope with high-speed movement of a workpiece, has an extremely simple arithmetic process, and has high reliability.

The graph which shows the relationship of the time vs. displacement information to the target position concerning one Embodiment of this invention 1 is a schematic block diagram showing arithmetic processing according to an embodiment of the present invention. 1 is a block diagram showing a digital circuit according to an embodiment of the present invention. The block diagram which shows the digital circuit concerning other embodiment by this invention. Furthermore, a block diagram showing a digital circuit according to another embodiment Furthermore, a block diagram showing a digital circuit according to another embodiment Furthermore, a block diagram showing a digital circuit according to another embodiment Furthermore, a block diagram showing a digital circuit according to another embodiment Furthermore, a block diagram showing a digital circuit according to another embodiment Furthermore, a block diagram showing a digital circuit according to another embodiment Furthermore, a block diagram showing a digital circuit according to another embodiment Furthermore, a block diagram showing a digital circuit according to another embodiment Furthermore, a block diagram showing a digital circuit according to another embodiment Furthermore, a block diagram showing a digital circuit according to another embodiment Furthermore, a block diagram showing a digital circuit according to another embodiment Furthermore, a block diagram showing a digital circuit according to another embodiment Furthermore, a block diagram showing a digital circuit according to another embodiment The flowchart concerning other embodiment by this invention. Sectional drawing which shows the touch probe concerning one Embodiment of this invention.

Embodiments of the present invention will be described below in detail with reference to the drawings.
The positioning measurement only detects whether the workpiece has come to a certain position. That is, the work is moved, the reference surface is detected by the stylus of the touch probe, and whether the position is before or after reaching the target position is measured. As a touch sensor system, a contact sensor type and a displacement sensor type are usually used.

  In the contact sensor type, one contact switch is built in the touch probe, and when the work has not reached, the contact switch is turned off (ON), and when the work arrives, it is turned on (OFF). For example, the movement of the workpiece is stopped using the signal.

The displacement sensor type incorporates a sensor that continuously detects the position inside the touch probe, measures the position of the workpiece before the workpiece reaches the target position, and whether the measured position has reached the target position. Is determined numerically. Then, the movement of the workpiece is stopped using the arrival as a signal.
Since the workpiece is moving, if the response due to positioning measurement is slow, the response time is delayed, and an accurate position cannot be obtained. In addition, the displacement sensor type calculates the numerical value at a predetermined sampling interval, so if the sampling interval is larger than the workpiece movement time, the fact that the workpiece has reached the target position is actually reached. The signal is transmitted to the machine tool, for example, to move the workpiece with a delay.

  The positioning measurement system requires chattering and contact deterioration due to mechanical contact switches, contact life, being unaffected by external electrical stress, and being able to withstand and respond to high-speed workpiece movement. Therefore, it is preferable to perform sampling at high speed using a displacement sensor type.

  Also, in order to reduce the inherent error of the sensor (particularly white noise variation), a non-delayed filter is provided so that no error occurs between the workpiece position and the stylus detection position. It shall be suitable for positioning measurement.

  Furthermore, using a small-scale digital circuit or a small microcomputer with a low calculation capability, the position information is acquired from the displacement sensor and output to the machine tool. Therefore, the digital filter processing is required not to cause a delay longer than the sampling interval, to be suitable for positioning a workpiece that moves linearly at a constant speed, and to be extremely simple in arithmetic processing. For this purpose, a plausible straight line is obtained by least square method approximation using past sampling data, and the current position is predicted from the straight line.

  FIG. 1 shows the relationship of time-displacement information to a target position accompanying the movement of a workpiece, and the workpiece moves linearly at a constant speed. The position of the workpiece is detected at a predetermined sampling interval, and is obtained as displacement information of past N-point data v1, v2,..., VN (the larger subscript is the older measurement data). Then, N pieces of data, which are displacement information obtained in time series at regular time intervals, are stored in the memory, and the displacement information data v at the next sampling timing is currently obtained at the sampling timing of v1 using this displacement information. Is obtained as a predicted value, and it is determined whether or not the target position has been reached. Alternatively, it is only necessary to obtain information on whether the target position has been reached or not yet reached without obtaining the data v of the next sampling timing.

  The displacement sensor is a transformer system using a coil, and it is preferable to use an LVDT sensor that converts a coil movement in a magnetic field generated by passing a current through the coil into a detectable electrical signal.

  The LVDT sensor outputs a mutual inductance change as a voltage when a magnetic core provided on the stylus moves in the primary side and secondary side coils. The magnetic core has a structure that moves inside the hollow of the cylindrical coil, and the primary coil is connected to an AC excitation power source, generates a voltage in the secondary coil, and utilizes the fact that a voltage difference occurs depending on the position of the magnetic core. .

If the predicted value v of the current value is sampled measurement data v1, v2,..., VN at times t1, t2,..., TN, this is expressed by v (t) = It is obtained by linear approximation as a (t) + b. The coefficients a and b that minimize the square error of the difference from the measurement data are determined. If the sum of squared differences is I,
The condition for minimizing I is
Can be obtained by solving the normal equation of (Equation 2) obtained by partial differentiation with variables a and b. When expressed using a matrix,
And a and b are obtained, and as a result, when the measurement data sampled at times t1, t2,..., TN are v1, v2,. Is expressed by the following formula.

Here, if sampling is set at an equal interval T and the time point at which the time is to be predicted is a reference (t = 0),
It becomes. The sampling intervals are t1 = -T, t2 = -2T,..., TN = -NT.
As a result, the predicted value v of the current value can be obtained.

  In order to simplify the calculation amount or the digital circuit, N is 2 or more, sampling is always performed at regular intervals, the output value of the displacement sensor at the target position is 0, or the target position has been reached or is still Suppose that it is only necessary to obtain information on whether or not it has reached.

  FIG. 2 is a block diagram showing an outline. What is indicated by reference numeral 12 is a stylus, and reference numeral 2 indicates a workpiece. The output from the displacement sensor 1 is converted into numerical data by the AD converter 4 through the detection circuit 3 as a non-delayed filter that reduces the inherent error of the sensor, and is calculated by the processing circuit 5 to obtain the target Outputs information on whether the position has been reached or not yet reached. The processing circuit 5 includes a linear approximation calculation circuit 5-1 that obtains a predicted value v of a current value, and a comparison circuit 5-2 that compares zero with a target value. However, the processing circuit 5 only needs to obtain information about whether the target position has been reached or not yet reached without directly obtaining the predicted value v of the current value. The amount can be significantly reduced.

Hereinafter, an example of the processing circuit will be shown as an embodiment, and the linear approximation calculation circuit 5-1 may be realized not only by the CPU but also by a digital circuit (digital filter or DSP (Digital Signal Processor)).
<Example 1>
FIG. 3 is obtained by replacing (Equation 6) with a digital circuit as it is, and outputs the same result. In the figure, Z ⁻¹ indicates a delay element of one clock. When the current value vN is output, this value outputs vN−1 input before one sampling. ◯ indicates an adder, and a triangle indicates a calculation process indicating a constant multiple of the subscript.

  Even if the coefficient of the multiplication circuit is adjusted with respect to FIG. 3, the information on whether or not the target position has been reached does not change, and a circuit that outputs the same result as in FIG. 3 can be configured.

<Example 2>
The “comparison circuit with zero” that is the subsequent circuit of “linear approximation calculation circuit” in FIG. 3 only determines whether the output value of “linear approximation calculation circuit” is equal to zero, greater than zero, or less than zero. If it performs, it will become information on whether the target position has been reached or not. In other words, only the sign relationship (positive number, negative number, or 0) is meaningful for the output value of the “linear approximation calculation circuit”, and its absolute value has no meaning. Therefore, the last multiplication coefficient in FIG. 3 can be omitted, and the same operation as in FIG. 3 can be realized as a result even if the circuit of FIG. 4 is used.

<Example 3>
Inserting a circuit that outputs a value with the same absolute value and opposite sign at an arbitrary position in the digital circuit, and adding the results of each to the numerical value at the arbitrary position outputs the same result . Therefore, in FIG. 4, it may be configured by the circuit of FIG. 5 in which a circuit for outputting a value having the same absolute value and the opposite sign is inserted at an arbitrary position and the respective results are added.

<Example 4>
Since the order of addition / subtraction does not affect the result, the order of addition / subtraction in FIG. 5 may be changed as shown in FIG. Further, another circuit that can obtain exactly the same result by the distribution law may be used.

  When a plurality of branches are made from a certain circuit portion, the same result is output even if the completely same circuit is configured redundantly. Therefore, a circuit that outputs the same result as in FIG. 6 can be configured.

  A circuit in which N sampling delay circuits are connected in series and an N sampling delay circuit output the same result. Therefore, a circuit that outputs the same result as that of FIG. 6 can be configured by replacing a portion in which N circuits for delaying one sampling are connected in series with a circuit for delaying N sampling.

<Example 5>
A circuit that delays one sampling and an arbitrary circuit output the same result even if the order is changed. Therefore, the circuit for delaying one sampling and the arbitrary circuit connected in series immediately before or after may be replaced as shown in FIG.

  Since the same result is output even if the same circuit is configured redundantly, a circuit that outputs the same result as in FIG. 7 can be configured. Further, the circuit may be configured by changing the order of the N-1 stage and the circuit for delaying one sampling.

<Example 6>
Further, even if the outputs of a plurality of circuits are added after being delayed by one sampling, the same result is output even if they are added to those delayed by one sampling after being added. Therefore, it may be configured as shown in FIG.

<Example 7>
In FIG. 9 with SW-A added to FIG. 8, if SW-A is tilted upward from the start, the same result as FIG. 8 is output. Further, even if SW-A is on the lower side in FIG. 9, if Q coincides with Q ′ at a certain time regardless of the state up to that point, then Q and Q ′ coincide with each other all the time thereafter. Therefore, in FIG. 9, when starting, SW-A is set to the upper side, and after all the element values of the circuit are determined, SW-A is switched to the lower side before the next sampling is performed. You may comprise so that it may be made.

  In FIG. 9, SW-A is set to the lower side from the time of startup, and after the values of all elements of the circuit are determined, the output values of all elements are initialized to 0 before the next sampling is performed. Naturally, Q and Q ′ have the same value (zero). Therefore, in FIG. 9, SW-A is set to the lower side at the time of start-up, and after the values of all elements of the circuit are determined, the output values of all elements are initialized to 0 before the next sampling is performed. You may comprise so that it may become. The output value may be initialized before the first sampling is performed.

<Example 8>
FIG. 10 shows a case where the output value of the circuit component is initialized to 0 in FIG. In the following embodiments, the output values of all circuit components are always initialized to 0 at the time of start-up (before the first sampling is performed). Moreover, the order of addition / subtraction may be changed, and you may comprise redundantly.

<Example 9>
A circuit in which N sampling delay circuits are connected in series and an N sampling delay circuit output the same result, and may be configured as shown in FIG.

<Example 10>
12 changes the order of addition and subtraction of FIG. 11 and outputs the same result as FIG. 11, so it may be configured as shown in FIG. Similarly to the description of FIG. 9, all circuit components may be configured to always output 0 at the time of startup.

<Example 11>
Since the output values of all the circuit components are initialized to 0 at the time of startup, the configuration as shown in FIG. 13 may be used. Since there are two component elements Z 1 ^-N that output the same result, the same result is output as the entire circuit even if they are combined into one, and the configuration shown in FIG. 14 may be used. Moreover, since the constant multiplication is the same even if divided, it may be configured as shown in FIG.

<Example 12>
If N is limited to a power of 2, that is, N = 2λ (λ is a natural number), FIG. 15 can be configured as shown in FIG. Also, since multiplying by 2k (k is a natural number) and shifting k bits to the left have the same result, FIG. 16 can be configured as shown in FIG. 17, and the arithmetic operation of multiplication is replaced with bit shift. Simpler and faster processing can be realized.

<Example 13>
FIG. 18 is a flowchart of a program that outputs the same result as FIG. 17, and the digital circuit of FIG. 17 may be programmed and executed by the CPU. In addition, the part which becomes the same value as the value of the variable used in FIG. 18 is also described in FIG. Also, in the program expression, the symbols v, w, and << are the measured values that v is sampled most recently (that is, the previous one), w is the measured value sampled 2λ (= N) times ago, and << is the left Bit shift operation symbol.

  FIG. 19 is a cross-sectional view showing the structure of a touch probe used in the positioning measurement system, which is a displacement sensor in which an LVDT sensor is stored in a cover 10. In FIG. 19, the LVDT sensor outputs a change in mutual inductance as a voltage when the magnetic core 15 moves in the vertical direction by the stylus 12 in the coil portion 11 provided with the primary side and secondary side coils. The lower end of the stylus 12 is in contact with the workpiece.

  At the upper end of the stylus 12, a hemispherical hemispherical portion 13 is pivotally supported by the cover 10, and is pressed downward by a spring 14 from a movable ring 16 that is slidable in the vertical direction within the cover 10. The magnetic core 15 is fixed to the movable ring 16, and the hemispherical portion 13 tilts when tilted in the X and Y directions perpendicular to the paper surface. Therefore, the magnetic core 15 moves by the vertical component of the inclined hemispherical part 13. On the other hand, since the coil portion 11 is fixed to the cover 10, a voltage difference corresponding to the relative position with the magnetic core 15 occurs and is converted into a displacement signal.

  Further, when the stylus 12 moves in the Z direction which is the vertical direction, the movable ring 16 similarly moves in the vertical direction, that is, the magnetic core 15 moves to obtain a displacement signal. Accordingly, the touch probe shown in FIG. 19 can convert the displacement of the stylus 12 in the X, Y, and Z directions into a displacement signal in the vertical direction.

  This touch probe is a displacement sensor continuously detected by an LVDT sensor, and has high sensitivity in all directions of X, Y, and Z. Further, unlike the contact sensor type, the displacement signal can be detected by converting it into a speed, and an abnormal approach or the like can also be detected. Therefore, if this touch probe is combined with the signal processing of the positioning system described above, it can be made more accurate and stable. Further, the output signal of the LVDT sensor can be digitized and numerically calculated at high speed, and accumulated error components such as errors of each device due to temperature changes and errors of each device in electrical characteristics can be eliminated.

  Moreover, it has high sensitivity in all directions of X, Y, and Z, predicts that it will reach the target position, and does not cause a delay due to sampling. Therefore, in a machining center (machine tool) or the like, it is also suitable for detecting whether or not the tool is broken before the tool contacts the workpiece.

DESCRIPTION OF SYMBOLS 1 Displacement sensor 2 Workpiece 3 Detection circuit 4 AD converter 5 Processing circuit 5-1 Linear approximation calculation circuit 5-2 Comparison circuit 10 Cover 11 Coil part 12 Stylus 13 Hemisphere part 14 Spring 15 Magnetic core 16 Movable ring

Claims (10)

In a positioning measurement system that measures the position of a reference plane of a workpiece and detects whether the position of the reference plane has reached a target position,
A touch probe having a stylus that contacts the reference surface and a displacement sensor that detects the position as numerical data;
The numerical data of N points is sampled and detected as past position information using the touch probe, and the position of the next sampling timing is calculated as a predicted value by linear approximation by the least square method from the numerical data. A linear approximation calculation circuit;
A comparison circuit that outputs information on whether the predicted value obtained by the linear approximation calculation circuit has reached the target position;
A positioning measurement system comprising:   2. The positioning according to claim 1, wherein the sampling intervals for detecting the numerical data at the N points are equal intervals, and the calculation processing is performed with a time point at which the time is to be predicted as a reference (t = 0). Measuring system.   3. The positioning measurement system according to claim 1, wherein the linear approximation calculation circuit and the comparison circuit are arithmetically processed by a digital circuit including a delay element, an adder, and a multiplier of a constant multiple. .   The positioning measurement system according to any one of claims 1 to 3, wherein the numerical data is adjusted to be 0 at the target position.   The positioning measurement system according to any one of claims 1 to 4, wherein an absolute value is omitted in the linear approximation calculation circuit and an arithmetic process is performed.   6. The positioning measurement system according to claim 1, wherein the linear approximation calculation circuit and the comparison circuit are arithmetically processed by a digital circuit integrated into one.   7. The linear approximation calculation circuit and the comparison circuit are configured to initialize output values of components to 0 before the first sampling is performed. The positioning measurement system described.   8. The positioning measurement system according to claim 1, wherein N is a power of 2 and N = 2λ (λ is a natural number), and a multiplication operation process is replaced with a bit shift. 9. The touch probe is a displacement sensor in which an LVDT sensor is stored in a cover,
A hemispherical hemisphere that is the upper end of the stylus and is pivotally supported by the cover;
A movable ring capable of sliding in the vertical direction within the cover;
A magnetic core fixed to the movable ring and moving in the vertical direction by the inclination of the hemisphere and movement in the vertical direction;
A coil portion that is fixed to the cover and outputs a displacement signal according to a relative position with the magnetic core;
The positioning measurement system according to any one of claims 1 to 8, further comprising:   The positioning measurement system according to any one of claims 1 to 9, wherein the straight line approximation calculation circuit and the comparison circuit are processed by a program by a CPU.