JPH09128029A - Data interpolation device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To determine interpolation positional data in which an error is small and also speed change is mitigated by the linear interpolation which is provided for the control of an NC machine tool and is easy in calculation by providing plural interpolation arithmetic parts storing different linear interplation methods and properly selecting one of the interpolation methods. SOLUTION: In a positional command data preparation part 2, a first interpolation arithmetic part 10, a second interpolation arithmetic part 11 and a third interpolation arithmetic part 12 storing each different linear interpolation methods are connected via a changeover switch 13. The changeover of this switch 13 is performed by an interpolation method selection judging part 14. The three interpolation arithmetic parts 10 to 12 are connected with a target value table 4. The target positional data stored in the target value table 4 is successively outputted to each positional controller 5 controlling the operation of each axis of an NC machine tool. Therefore, the linear interpolation in which an error is small and speed change is mitigated can be performed without using a high grade interpolation method by properly selecting one of the interpolation arithmetic parts 10 to 12.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data interpolation for controlling a numerically controlled machine tool, which linearly interpolates input discrete position command data to calculate a target position group between two designated points. Regarding the device.

[0002]

2. Description of the Related Art At present, numerical control machine tools (hereinafter referred to as NC machine tools) are generally used in automation of drilling, milling and the like. What is numerical control?
This is to automatically perform complicated operations of machining which are conventionally performed by an operator by giving numerical information composed of numerical values and signs. The numerical information is composed of coordinate data indicating the position of the tool, data instructing the operation sequence of the machine, and the like, and is recorded on a perforated tape called NC tape (specifically, paper tape, magnetic tape, etc.). The machine tool is controlled according to this numerical information and automatically processes the workpiece.

In this NC machine tool, since the position information of the tool is commanded by discrete values, it is given when actually moving the tool linearly or in a curve (circle, arc, parabola, etc.). It is necessary to approximate (interpolate) a point group along a predetermined function curve from one point to the next point, and obtain a target position group between the two points. Normally, linear interpolation is performed to approximate a given two points by a point group along a straight line for reasons such as easy calculation.

FIG. 6 is a block diagram of a control system of a conventional NC machine tool. The numerical information recorded on the NC tape 1 is input to the numerical controller. The position command data creation unit 2 that constitutes the central processing unit in the numerical control device creates discrete position command data from the input numerical information, and the interpolation calculation unit that constitutes the positioning unit in the numerical control device. Output to 3. The interpolation calculation unit 3 linearly interpolates the discrete position command data output from the position command data creation unit 2, and a target value table 4 forming the same positioning unit uses the calculated interpolation positions (target position) for several steps. Group) is output. The target position data stored in the target value table 4 is sequentially output to each position controller 4 that controls the operation of each axis (here, n) of the NC machine tool. Each position controller 4 outputs an operation signal to the shaft motor 7 via the amplifier 6. At this time, the position detection signal from the encoder 8 mechanically attached to the shaft motor 7 is used as the target value table 4
The target position data from is compared, and feedback control is performed so that they match. If the number of axes of the NC machine tool is 2 or more (n ≧ 2), each position controller 4
Continuously associates motions of two or more axes while continuously controlling the corresponding axes.

[0005]

However, in such a conventional apparatus, since there is only one interpolation calculation section and the interpolation calculation can be performed only by one kind of interpolation method programmed in the interpolation calculation section, the interpolation calculation is performed. There are the following problems depending on the selection of the method.

That is, in the linear interpolation method, the remainder is averagely distributed to each step according to the distribution method of the remainder obtained by dividing the total movement amount by the division step (hereinafter, this is referred to as an average division method). Call), a method of putting the remainder in the first half of each step (hereinafter, this will be called a remainder preemption method), and a method of putting the remainder in the second half of each step (hereinafter, This will be referred to as the retrofitting method), and the average division method is usually used. This is because the surplus-first method and the surplus-postfix method have a larger change in speed during interpolation than the average division method. Generally, such linear interpolation (by means of average division)
In the above, there is a problem that an error between the target trajectory and the actual trajectory and a speed change between two front and rear interpolations are large.

In order to reduce the error and alleviate the speed change, it is conceivable to use so-called spline interpolation (a method that uses a spline function as an interpolation formula) instead of linear interpolation, but spline interpolation is extremely difficult to calculate. It is complicated and requires a lot of peripheral data (for example, a certain number of data before and after, many data before and after, dozens of data before and after, dozens of data before and after, etc.) There is a problem in that point, and high-performance equipment is required to address this, leading to higher costs.

The present invention has been made in view of the above problems of the prior art, and the error is small and the speed change is alleviated by the linear interpolation which is used for the control of the NC machine tool and which is easy to calculate. An object of the present invention is to provide a data interpolating device that can obtain interpolated position data.

[0009]

In order to achieve the above-mentioned object, the invention according to claim 1 linearly interpolates the input discrete position command data for controlling a numerically controlled machine tool, and gives an instruction. In the data interpolating device for calculating the target position group between the two points, the position command data input by the input device is linearly interpolated by the input means for inputting the discrete position command data and the averaging method. A first interpolating means, a second interpolating means for linearly interpolating the position command data input by the input means by a remainder pre-adding method, and a linear interpolating position command data input by the input means by a surplus post-applying method. Third interpolating means for interpolating, and based on the position command data input by the inputting means, the first interpolating means, the second interpolating means, and the third interpolating means. Selection means for selecting any one of, and having a storage means for storing the output from the interpolation means selected by said selecting means.

In the data interpolating apparatus according to the present invention, the input means inputs discrete position command data, and the selecting means performs the first interpolation based on the position command data inputted by the input means. Means, second interpolating means, and third
Select one of the interpolation means. When the first interpolating means is selected by the selecting means, the first interpolating means:
The target command group between two points is calculated by performing linear interpolation on the position command data input by the input means by a method of average division. When the second interpolating means is selected by the selecting means, the position command data input by the second interpolating means inputting means is linearly interpolated by a remainder-preceding method to calculate a target position group between two points. To do. Further, when the third interpolating means is selected by the selecting means, the third interpolating means performs linear interpolation on the position command data input by the inputting means by a method of residual addition, and a target position group between two points. Is calculated. The storage means stores the output from the interpolation means selected by the selection means. That is, three different linear interpolation methods are stored.
Since there are provided different kinds of interpolation calculation units and one of them is appropriately selected, it is possible to perform linear interpolation in which the error is small and the speed change is moderated.

According to a second aspect of the present invention, in the data interpolating apparatus according to the first aspect, the selecting means sequentially calculates the amount of movement between two points from the input position command data, and obtains the present movement. It is characterized in that the first interpolation means is selected when the absolute value of the difference between these amounts is compared with the next movement amount and is less than or equal to a preset judgment value. In the data interpolating device according to the second aspect, the selecting means selects 2 from the position command data input by the input means.
The amount of movement between points is sequentially calculated, the calculated current amount of movement is compared with the next amount of movement, and when the absolute value of the difference is less than or equal to a preset judgment value, the first interpolation means is selected. To do.

According to a third aspect of the present invention, in the data interpolating device according to the first aspect, the selecting means sequentially calculates the amount of movement between two points from the input position command data, and obtains the present movement. Amount is compared with the next movement amount, when the value obtained by subtracting the next movement amount from the current movement amount is greater than or equal to the preset judgment value, and the previous interpolation method was too much retrofitting, It is characterized in that the first interpolation means is selected. In the data interpolating device according to the third aspect of the present invention, the selecting means sequentially calculates the movement amount between two points from the position command data input by the inputting means, and obtains the present movement amount and the next movement amount. The first interpolation means is selected when the value obtained by subtracting the next movement amount from the current movement amount is equal to or greater than the preset determination value and the previous interpolation method is the post-addition method. To do.

According to a fourth aspect of the present invention, in the data interpolating apparatus according to the first aspect, the selecting means sequentially calculates the amount of movement between two points from the input position command data, and obtains the present movement. If the value obtained by subtracting the next movement amount from the current movement amount is greater than or equal to the preset judgment value, and the previous interpolation method was other than the retrofit method, , The second interpolation means is selected. In the data interpolating device according to the present invention, the selecting means successively calculates the movement amount between two points from the position command data input by the inputting means, and obtains the present movement amount and the next movement amount. When the value obtained by subtracting the next movement amount from the current movement amount is equal to or greater than the preset determination value and the previous interpolation method is other than the post-installation method, the second interpolation means is set. select.

According to a fifth aspect of the present invention, in the data interpolating device according to the first aspect, the selecting means sequentially calculates the amount of movement between two points from the input position command data, and obtains the present movement. Amount is compared with the next movement amount, the value obtained by subtracting the current movement amount from the next movement amount is equal to or greater than the preset determination value, and the previous interpolation method is the preemptive method, It is characterized in that the first interpolation means is selected. In the data interpolating apparatus according to the present invention, the selecting means sequentially calculates the movement amount between two points from the position command data input by the input means, and obtains the present movement amount and the next movement amount. When the value obtained by subtracting the current movement amount from the next movement amount is equal to or greater than the preset determination value and the previous interpolation method is the remainder preemption method, the first interpolation means is selected. To do.

According to a sixth aspect of the present invention, in the data interpolating apparatus according to the first aspect, the selecting means sequentially calculates a movement amount between two points from the input position command data, and obtains the present movement. Amount is compared with the next movement amount, and the value obtained by subtracting the current movement amount from the next movement amount is greater than or equal to the preset judgment value, and the previous interpolation method was other than the preemptive method. , The third interpolating means is selected. In the data interpolating device according to the present invention, the selecting means successively calculates the moving amount between the two points from the position command data input by the inputting means, and obtains the present moving amount and the next moving amount. When the value obtained by subtracting the current movement amount from the next movement amount is equal to or greater than a preset determination value and the previous interpolation method is other than the pre-emption method, the third interpolation means is used. select.

[0016]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram of a control system of an NC machine tool including a data correction device of the present invention. In addition,
The same parts as those in FIG. 6 are designated by the same reference numerals.

This NC machine tool has n axes, and each axis is provided with a position controller 5, an amplifier 6, an axis motor 7, and an encoder 8 as in the conventional case. The position controller 5 inputs the target position data from the target value table 4, compares it with the position detection signal fed back from the encoder 8 (the current position of the shaft motor 7) to generate and generate an operation signal. Operation signal amplifier 6
To the shaft motor 7 via. As a result, the shaft motor 7 operates so as to match the target position. Since the target position data in the target value table 4 is sequentially transferred to each position controller 5, each axis moves following the change in the target position. At that time, the number of axes of the NC machine tool is 2 or more (n ≧
In the case of 2), each position controller 5 constantly controls the operation of the NC machine tool while simultaneously associating the motions of two or more axes. Although the position of the shaft motor 7 is detected by the encoder 8 here, it is needless to say that the position is not limited to this. For example, a pulse generator or the like may be used instead of the encoder 8.

A machine program for controlling the NC machine tool is provided by the NC tape 1. The NC tape 1 is a perforated tape on which numerical information is recorded, on which coordinate data indicating the position of the tool, data on the operation sequence of the machine, and the like are recorded according to a predetermined format.
The position information of the tool instructed on the NC tape 1 is discrete, and is given as discrete values. The numerical information recorded on the NC tape 1 is input to the numerical control device and transferred to the position command data creating unit 2 which constitutes the central processing unit in the numerical control device. The position command data creation unit 2 has a function of creating discrete position command data from the input numerical information. The position command data creation unit 2 includes three interpolation calculation units that store different linear interpolation methods,
That is, the first interpolation calculation unit 10, the second interpolation calculation unit 11, and the third interpolation calculation unit 12 are connected via the changeover switch 13. The first interpolation calculation unit 10 stores a so-called average division method, and performs linear interpolation by the average division method on the input discrete position command data. The second interpolation calculation unit 11 stores a so-called remainder pre-allocation method,
Linear interpolation is performed on the input discrete position command data by the remainder pre-addition method. The third interpolation calculation unit 12 stores a so-called surplus retrofitting method, and performs linear interpolation on the input discrete position command data by the surplus retrofitting method. These linear interpolation methods will be described later. The switch 13 is switched by the interpolation method selection determination unit 1
4 is performed. The interpolation method selection determination unit 14 is also connected to the position command data creation unit 2, and the interpolation method selection determination unit 14 inputs the position command data from the position command data creation unit 2 and the input position command data The optimum linear interpolation method is selected based on the above, and the switch 14 is switched according to this result, that is, the position command data creation unit 2
To output position command data from the interpolation calculation units 11 to 12
Has a function of selecting one. The interpolation method selected by the interpolation method selection determination unit 14 is stored in a predetermined internal memory in the interpolation method selection determination unit 14 in association with the two points each time. The determination criteria for selecting the interpolation method in the interpolation method selection determination unit 14 will be described later. The three interpolation calculation units 10 to 12 are connected to the target value table 4.
The target value table 4 includes the selected interpolation calculation units 10-12.
Result, that is, the data of the target position group (interpolation position for a predetermined number of steps) between two points obtained by linearly interpolating the input discrete position command data
It is stored in the form of a table including. Target value table 4
As described above, the target position data stored in
It is sequentially output to each position controller 4 that controls the operation of each axis of the machine tool. The above three interpolation calculation units 10 to 10
12, interpolation method selection determination unit 14, and target value table 4
Constitutes a positioning unit in the numerical controller.

The input means is the NC tape 1 and the position command data creating section 2, the first interpolating means is the first interpolation computing section 10,
The second interpolation means is constituted by the second interpolation calculation section 11, the third interpolation means is constituted by the third interpolation calculation section 12, the selection means is constituted by the changeover switch 13 and the interpolation method selection judgment section 14, and the storage means is constituted by the target value table 4. There is.

FIG. 2 is an explanatory view of the above-mentioned three kinds of linear interpolation methods, and FIG. 3 is a graph showing the results of FIG. Here, the commanded movement amount (distance) between the two points is 12,
The case where the number of division steps is 8 is illustrated.

The average division method executed by the first interpolation calculation unit 10 is a value obtained by dividing the given total movement amount between two points by the number of division steps (1 in this example), This is a method to distribute the remainder to each step on average (Fig. 2
(A)). The remainder pre-allocation method executed by the second interpolation calculation unit 11 is a value obtained by dividing the given total movement amount between two points by the number of division steps (1 in this example), which is not divisible. This is a method in which the remainder is put together in the first half of each step (see FIG. 2B). Also, the third
The surplus retrofitting method executed by the interpolation calculation unit 12 is a value obtained by dividing the given total movement amount between two points by the number of division steps as the basic movement amount (1 in this example), and the remainder that cannot be divided is This is a method of putting them together in the latter half of each step (see FIG. 3C). FIG. 3 is a graph showing each result shown in FIG. 2 in the form of distance with respect to the division step. Since each step is processed in the same time, as can be seen from the figure, the extra-heading method makes the moving speed of the first half faster than that of the second half (because the moving distance of the first half is larger). According to the retrofit method, the moving speed of the latter half is faster than that of the first half (because the moving distance of the latter half is larger), and according to the average division method, moving between two points while repeating constant speed change. become.

FIG. 4 is a flow chart showing the operation of the interpolation method selection judgment unit 14. The determination criteria used when selecting the interpolation method in the interpolation method selection determination unit 14 are shown in this flowchart.

First, the interpolation method selection judgment unit 14 inputs the position command data scattered from the position command data generation unit 2 (step S1), and based on the data of the two points before and after given by this, the interpolation between the two points is performed. The amount of movement K (j) of S is calculated sequentially and the result is stored in a predetermined internal memory (step S
2).

In step S3, from the internal memory storing the result of step S2, the amount of movement K (n) between the two points for which linear interpolation is to be performed (current movement amount) and the two points for the next linear interpolation are performed. Amount of movement (called the next movement) K (n
+1) is read and compared with each other to determine whether the current movement amount K (n) is, for example, the next movement amount K (n + 1) or more.

If the current movement amount K (n) is greater than or equal to the next movement amount K (n + 1) as a result of the determination in step S3,
A value (K (n) -K (n + 1)) obtained by subtracting the next movement amount K (n + 1) from the current movement amount K (n) is, for example, larger than the predetermined judgment value D. It is determined whether or not (step S4), and as a result of this determination, K (n) -K (n + 1)
If is less than or equal to the determination value D, the average division method is selected and the changeover switch 13 is set in the first interpolation calculation unit 10 (step S8).

The reason is as follows. K (n) -K (n + 1)
Is less than or equal to the determination value D, the difference between the current movement amount K (n) and the next movement amount K (n + 1) is not so large, so the nodes of both movements (that is, the two commands before and after are commanded). To slow down the speed change before and after the command point, which is common to the current move and the next move, among the points, use the average division method that performs linear interpolation so that the speed change averages between the two points. Because it is enough.

As a result of the determination in step S4, K
If (n) -K (n + 1) is larger than the judgment value D, the internal memory that stores the result of the past interpolation method selection is further accessed to determine whether the previous interpolation method is a postscript method. Is determined (step S5). As a result of this judgment, when the previous interpolation method is the excessive post-addition method, the average division method is selected (step S8), and the previous interpolation method is not the excessive post-attachment method, that is, the average division method or the extra pre-addition method. In this case, the surplus prepositioning method is selected (step S9), and the changeover switch 13 is set to the first interpolation calculation section 10 and the second interpolation calculation section 11, respectively.

The reason is as follows. K (n) -K
If (n + 1) is larger than the judgment value D, the current movement amount K
(n) is considerably larger than the next movement amount K (n + 1), and at least in order to moderate the speed change before and after the nodes of both movements (that is, the later command point as described above). Two
The movement speed in the latter half of the division step between points must be small.However, if the previous interpolation method was too late, the movement speed in the latter half of the division step was high in the previous interpolation, so Node between movement and current movement (that is, the previous of the two commanded points that move back and forth)
From the viewpoint of easing the speed change before and after, it is better to increase the moving speed in the first half of the division step between the two points, but this increases the moving speed before and after the node. Therefore, the error becomes large. Therefore, as a harmony point between the two, an average division method in which the two points are averagely divided and the velocity change is not so much is selected. On the other hand, in the same case, when the previous interpolation method is a method other than the post-remainder method (that is, the average division method or the remainder-preceding method), the movement speed of the latter half of the division step in the previous interpolation is not so fast, The balance as described above is not necessary so much, and in order to reduce the moving speed in the latter half of the dividing step between the two points, a surplus prepositioning method in which the moving speed in the first half is faster than in the latter half is selected.

Furthermore, if the current movement amount K (n) is smaller than the next movement amount K (n + 1) as a result of the determination in step S3, the next movement amount K (n + 1) to the current movement amount K It is judged whether or not the value (K (n + 1) -K (n)) obtained by subtracting (n) is larger than the judgment value D (step S6). As a result of this judgment, K (n + 1)- If K (n) is less than or equal to the determination value D, the average division method is selected and the changeover switch 13 is set in the first interpolation calculation unit 10 (step S8).

The reason is as follows. K (n + 1)-
When K (n) is less than or equal to the determination value D, the current movement amount K (n) and the next movement amount K (n + 1) are calculated as in the case of executing step S8 according to the determination result of step S4 described above. Since the difference is not so large, the interpolation by the average division method is sufficient to make the speed change before and after the node of both movements (that is, the latter one of the two command points before and after). Because.

Further, as a result of the determination in step S6, K
If (n + 1) -K (n) is larger than the judgment value D, the internal memory that stores the result of the past interpolation method selection is further accessed to determine whether the previous interpolation method is the pre-emptive method. Is determined (step S7). As a result of this judgment, when the previous interpolation method is the residual pre-addition method, the average division method is selected (step S8), and the previous interpolation method is not the preliminary pre-attachment method, that is, the average division method or the residual post-attachment method. If so, the post-removal method is selected (step S10)
The changeover switches 13 are set to the first interpolation calculation section 10 and the third interpolation calculation section 12, respectively.

The reason is as follows. K (n + 1) −
If K (n) is larger than the judgment value D, the current movement amount K
(n) is much smaller than the next movement amount K (n + 1), and in order to make the speed change before and after the node of both movements (that is, the subsequent command point) at least, divide it between the two points. The movement speed in the latter half of the step must be fast, but when the previous interpolation method was too much of a pre-emptive method, the movement speed in the latter half of the division step was slow in the previous interpolation, so From the viewpoint of reducing the speed change before and after the node (that is, the previous command point) with the movement, it is better to slow down the movement speed in the first half of the division step between the two points, but this is Since the moving speed before and after the node becomes slow, the error becomes large. Therefore, as a harmony point between the two, an average division method in which the two points are averagely divided and the velocity change is not so much is selected. On the other hand, in the same case, when the previous interpolation method is a method other than the remainder pre-attachment method (that is, the average division method or the remainder addition method), the movement speed of the latter half of the division step in the previous interpolation is not so slow, The above-mentioned balance is not necessary so much, and in order to increase the moving speed in the latter half of the dividing step between the two points, a post-addition method in which the moving speed in the first half is slower than that in the latter half is selected.

The interpolation methods selected in step S8, step S9, and step S10 are stored in a predetermined internal memory in association with the corresponding points, and are used for the determination in steps S5 and S7. It

When the processing of step S8, step S9, or step S10 is completed, it is judged whether or not a series of processing operations is to be completed depending on whether or not an end signal is input (step S11), and if not completed, step S1. Then, the above series of processing is repeated.

FIG. 5 is a diagram showing an example of an actual operation according to the above example. As can be seen from the figure, compared to the conventional method in which the linear interpolation is performed only by the average division method for the curve (target locus) that smoothly connects the commanded points (nodes), three types of interpolation methods are used. The error with respect to the target locus is generally smaller in the case of the present invention in which the optimum interpolation method is selected and linear interpolation is performed. The speed change before and after each node is also gentle.

Therefore, according to the present embodiment, the three interpolation calculation units 10 to 10 storing different linear interpolation methods, respectively.
12 is provided, and the interpolation method selection determination unit 14 is configured to select one of them appropriately in consideration of the change in the moving amount and the moving speed before and after each instructed point, so that the optimum linear interpolation is always performed. The position command data can be linearly interpolated by the method, and the accuracy is improved compared to the conventional case where the linear interpolation is performed only by the average division method.
The speed change before and after each node becomes gentle. That is,
The tool can be moved more accurately and more smoothly (see FIG. 5).

Moreover, since the above effect is obtained by linear interpolation, it is not necessary to use spline interpolation, the calculation is greatly simplified as compared with the case of using spline interpolation, and it is possible to sufficiently cope with existing performance equipment. There is no cost increase.

[0038]

As described above, according to the first aspect of the present invention, three kinds of interpolating operation units storing different linear interpolation methods are provided and one of them is appropriately selected. It is possible to perform linear interpolation with a small error and a reduced speed change without using a high-level interpolation method such as.

Further, according to each of the inventions described in claims 2 to 6, since the interpolation method is selected in consideration of the change in the moving amount and the moving speed before and after each commanded point, the command is issued. When performing linear interpolation between two points, the optimum interpolation method can always be selected.

[Brief description of the drawings]

FIG. 1 is a block diagram of a control system of an NC machine tool including a data correction device of the present invention.

FIG. 2 is an explanatory diagram of three linear interpolation methods.

FIG. 3 is a graph showing the results of FIG.

FIG. 4 is a flowchart showing the operation of an interpolation method selection determination unit.

FIG. 5 is a diagram showing an actual operation example according to the present invention.

FIG. 6 is a block diagram of a control system of a conventional NC machine tool.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... NC tape (input means) 2 ... Position command data creation part (input means) 4 ... Target value table (storage means) 10 ... First interpolation calculation part (first interpolation means) 11 ... Second interpolation calculation part (first 2 interpolation means) 12 ... third interpolation calculation section (third interpolation means) 13 ... changeover switch (selection means) 14 ... interpolation method selection determination section (selection means)

Claims (6)

[Claims] 1. A data interpolating device for controlling a numerically controlled machine tool, which linearly interpolates input discrete position command data to calculate a target position group between two designated points. Input means for inputting position command data, first interpolating means for linearly interpolating the position command data input by the input means by a method of average division, and position input by the input means by a method of remainder heading Second interpolating means for linearly interpolating the command data, third interpolating means for linearly interpolating the position command data input by the input means by a method with a remainder, and based on the position command data input by the input means A selection means for selecting any one of the first interpolation means, the second interpolation means, and the third interpolation means; A data interpolating device comprising: a storage unit that stores the output from the selected interpolating unit. 2. The selecting means successively calculates the movement amount between two points from the input position command data, compares the obtained movement amount and the next movement amount, and obtains the absolute value of the difference between them. When it is less than or equal to a preset judgment value, the first
2. The data interpolating device according to claim 1, wherein the interpolating means is selected. 3. The selecting means successively calculates the amount of movement between two points from the input position command data, compares the calculated current amount of movement with the next amount of movement, and calculates the next amount of movement from the current amount of movement. 2. The first interpolation means is selected when the value obtained by subtracting the movement amount is equal to or greater than a preset determination value and the previous interpolation method is a post-replacement method. Data interpolator. 4. The selecting means successively calculates the amount of movement between two points from the input position command data, compares the calculated current amount of movement with the next amount of movement, and calculates the next amount of movement from the current amount of movement. The second interpolation means is selected when the value obtained by subtracting the movement amount is equal to or greater than a preset determination value and the previous interpolation method is other than the post-replacement method. The described data interpolator. 5. The selecting means sequentially calculates the amount of movement between two points from the input position command data, compares the calculated current amount of movement with the next amount of movement, and calculates the current amount of movement from the next amount of movement. 2. The first interpolating means is selected when the value obtained by subtracting the movement amount is equal to or greater than a preset determination value and the previous interpolation method is a surplus advance method. Data interpolator. 6. The selecting means sequentially calculates the amount of movement between two points from the input position command data, compares the calculated current amount of movement with the next amount of movement, and calculates the current amount of movement from the next amount of movement. The third interpolating means is selected when the value obtained by subtracting the movement amount is equal to or greater than a preset determination value and the previous interpolation method is other than the pre-prediction method. The described data interpolator.