JP2009009274A - Numerical controller - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a numerical controller for correcting any errors present between a reference coordinate system for defining a commanded working shape and a coordinate system in actual working. <P>SOLUTION: A translation error and a rotation error are present between the reference coordinate system and an actual coordinate system. A tool length correction vector T is added to a tool tip position Pt commanded on the reference coordinate system, so that a straight axis machine coordinate position Pm can be calculated. Correction quantity ΔD of the tool tip position is calculated from the straight axis machine coordinate position Pm, the applied translation error and the error matrix of the rotation error. The correction quantity ΔD is added to the straight axis machine coordinate position Pm, so that a straight axis correction machine coordinate position Pm' can be calculated, and each straight axis is driven to the straight axis correction machine coordinate position Pm'. The rotation axis is driven to the commanded rotational position. The tool tip position Pt on the reference machine coordinate system and the tool tip position Pt' on the actual machine coordinate system are made equal to each other. As a result, a workpiece can be machined with high precision. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a numerical control device for controlling a 5-axis machine having three linear axes and two rotary axes. In particular, the present invention relates to a numerical control device that corrects errors such as translation errors and rotation errors caused by assembly errors.

The actual machining shape by the machine tool has an error with respect to the command machining shape. This error is also caused by a ball screw pitch error, a machine tool assembly error, deformation due to the weight of each part of the machine such as a column or the work piece, or deformation due to a temperature change. Further, the guide surface or the like is deformed according to the position on the guide surface of the table to which the workpiece (workpiece) is attached, and this deformation causes a processing error. The machining shape is instructed based on the reference coordinate system. However, the coordinate system when the machining is actually performed is shifted from the reference coordinate system, and an error is generated. This error causes a machining shape error.
Non-Patent Document 1, which is a method for correcting these errors, describes a method for correcting translation errors (δx, δy, δz) and rotation errors (α, β, γ) that exist between coordinate systems. Yes.

An error matrix ε _ik representing an assembly error (straightness error and parallelism error) between the i-axis and the k-axis of the machine tool has rotation errors α _ik and β _ik about the X axis, the Y axis, and the Z axis, respectively. , by gamma _ik have been shown to be the following equation (1). Α _ik , β _ik , and γ _ik are assumed to be sufficiently small, and sin (α _ik ) is approximated by α _ik , sin (β _ik ) is approximated by β _ik , and sin (γ _ik ) is approximated by γ _ik . . Moreover, it shows in the homogeneous coordinate system. Note that i and k are arbitrary axes (linear axes X, Y, Z, rotation axes A, B, C) of the machine tool.

また、ｉ軸とｋ軸間の組立誤差（軸心ずれ誤差）を表す誤差行列ε_ikは、Ｘ軸方向、Ｙ軸方向、Ｚ軸方向それぞれの並進誤差δx_ik，δy_ik，δz_ikによって、次の式（２）のようになることが示されている。

Further, an error matrix ε _ik representing an assembly error (axial misalignment error) between the i-axis and the k-axis is expressed by translation errors δx _ik , δy _ik , and δz _{ik in} the X-axis direction, the Y-axis direction, and the Z-axis direction, respectively. It is shown that the following equation (2) is obtained.

２つの座標系間の誤差はこれらの和であるので、一般に２つの座標系間の誤差を表す誤差行列εは、それらの間の回転誤差（α，β，γ）と並進誤差（δx，δy，δz）によって次の式（３）のようになる。

Since the error between the two coordinate systems is the sum of these, generally, the error matrix ε representing the error between the two coordinate systems is the rotation error (α, β, γ) and translation error (δx, δy) between them. , Δz), the following equation (3) is obtained.

つまり、Ｘ、Ｙ、Ｚ３軸の基準座標系に対して実座標系がＸ軸周り、Ｙ軸周り、Ｚ軸周りそれぞれの回転誤差α、β、γとＸ軸方向、Ｙ軸方向、Ｚ軸方向それぞれの並進誤差δx、δy、δzを持っているとした場合、基準座標系上のある点Ｐ(Ｐx，Ｐy，Ｐz)^Tが基準座標系から実座標系への座標変換後に基準座標系上の点Ｐ'(Ｐx'，Ｐy'，Ｐz')^Tとなるとすると、次の式（４）のような関係となる。この関係を図８に示す。なお、各位置も同次座標系で示している。

That is, the actual coordinate system is rotated around the X axis, the Y axis, and the Z axis with respect to the X, Y, and Z3 axis reference coordinate systems, and the X axis direction, the Y axis direction, and the Z axis. Assuming that there are translation errors δx, δy, δz in each direction, a certain point P (Px, Py, Pz) ^T on the reference coordinate system is converted into a reference coordinate system after the coordinate conversion from the reference coordinate system to the real coordinate system. If the upper point P ′ (Px ′, Py ′, Pz ′) ^T is assumed, the relationship is as shown in the following equation (4). This relationship is shown in FIG. Each position is also shown in a homogeneous coordinate system.

点Ｐと点Ｐ’間の基準座標系の直線軸Ｘ，Ｙ，Ｚ軸上の誤差をΔＤ(ΔＸ，ΔＹ，ΔＺ) ^Tとすると、
ΔＤ＝Ｐ’−Ｐ
となる。

When the error on the linear axes X, Y, and Z axes of the reference coordinate system between the point P and the point P ′ is ΔD (ΔX, ΔY, ΔZ) ^T ,
ΔD = P′−P
It becomes.

Conventionally, in a numerical controller, after acceleration / deceleration processing of each axis is performed as shown in FIGS. 9 and 10, correction such as pitch error correction and straightness correction is performed on the mechanical position of each axis by a correction unit. Is going.
FIG. 9 is a schematic block diagram of the operation process of the numerical controller in the case of the tool length correction command. FIG. 10 is a schematic block diagram of the operation process of the numerical control device in the case of the tool tip point control.

  In the case of the tool length correction command shown in FIG. 9, the program block is analyzed by the command analysis unit 1, and the command position based on the tool tip point position of the analysis result is analyzed at the command position by the tool length correction vector addition element 2. A tool length correction vector is added to a certain tool tip point position to obtain a linear axis machine coordinate position (see FIG. 11). The interpolation means 3 interpolates the linear axis machine coordinate position and the rotation axis command position to generate a machine coordinate value of each axis, and each axis (X, Y, Z, B (A), C axis) After acceleration / deceleration is performed by the acceleration / deceleration processing units 4x, 4y, 4z, 4b (a) and 4c, the correction means 5 performs the above-described pitch error correction and straightness on the command position after the acceleration / deceleration processing of each axis. After the degree correction is performed, the servo motors 6x, 6y, 6z, 6b (a), and 6c of each axis are driven.

  Further, in the case of the tool tip point control, the numerical control device interpolates the tool tip point by the interpolation means 3 as shown in FIG. 10, and the tool length correction vector addition element 2 with respect to the interpolated tool tip point position. After the tool length correction vector is added, a linear axis machine coordinate position is generated, and the acceleration / deceleration processing units 4x to 4c perform acceleration / deceleration on the linear axis machine coordinate position and the interpolated rotation axis position. Various corrections are made for each axis, and the servo motor for each axis is driven.

FIG. 11 shows a case where the correction unit 5 corrects the error ΔD caused by the assembly error described in Non-Patent Document 1 by the conventional numerical control device shown in FIG. 9 (FIG. 10).
In FIG. 11, the end face of the tool head 11 to which the tool 10 is attached is set as the machine coordinate position, and the numerical control device controls the position of the end face of the tool head 11. Therefore, the tool length correction vector T is added to the tool tip point position Ptc, which is the command position, and the linear axis machine coordinate position Pmc to which the tool length correction vector T is added is obtained, and the linear axis machine coordinate position Pmc and the rotation axis After completion of the acceleration / deceleration processing after interpolating the position by the interpolation means 3 to generate the machine coordinate position of each axis and performing acceleration / deceleration by the acceleration / deceleration processing units 4x, 4y, 4z, 4b (a), 4c, respectively The command position is corrected by the error ΔD caused by the above-described assembly error, etc., and the machine coordinate position Pmc ′ is obtained to drive the servo motor of each axis. However, as will be described later, this correction does not correct the tool tip position even if correction by coordinate transformation is performed from the reference coordinate system to the real coordinate system. That is, the command position on the reference coordinate system is different from the tool tip position on the real coordinate system.

In addition to Non-Patent Document 1 described above, Patent Document 1 discloses a method for correcting an assembly error of a machine tool by correcting a machining origin based on a rotation axis error and a command angle with respect to a reference axis. A method for correcting the tilt error and the position error of the rotating shaft caused by the assembly error of the table or the like and the machining error is well known.
Further, Patent Document 2 describes a method for correcting the relative relationship between the tool and the workpiece when there is no mechanical error from the amount of deviation of the rotation axis or the amount of deviation of the spindle turning center. Yes.

JP 2005-59102 A JP 2004-272887 A Supervised by Ichiro Inasaki, “Shape Creation Theory of Machine Tools”, 1st Edition, Yokendo Co., Ltd., July 28, 1997, pp. 76-86

  Even if the conventional correction means 5 shown in FIGS. 9 and 10 performs correction by coordinate transformation to a real coordinate system having an error with respect to the reference coordinate system defining the command machining shape, it is a correction for the machine coordinate position. It is not a correction for the tool tip position. That is, adding the tool length correction vector to the tool tip position to obtain the machine coordinate position of the linear axis and correcting the machine coordinate position does not correct the tool tip position. The mathematical description will be described later. Furthermore, if there is no rotation axis but only a linear axis, the difference between the command position on the reference coordinate system and the tool tip position on the actual coordinate system is almost static, and can be handled by changing the coordinate system origin. However, if there is a rotation axis as in a 5-axis machine, the difference between the command position on the reference coordinate system and the tool tip position on the actual coordinate system is the distance from the rotation axis position or tool tip point position to the rotation center. It changes dynamically depending on etc. Therefore, since correction at the tool tip point position (machining point) is not performed correctly, machining with high accuracy cannot be performed.

The invention described in Patent Document 1 is a method of correcting the machining origin based on the rotation axis error and the command angle with respect to the reference axis, and the reference coordinate system is only used for the rotation axis error with respect to the reference axis. Translation errors (δx, δy, δz) and rotation errors (α, β, γ) existing between the actual coordinate system and the actual coordinate system cannot be corrected accurately. For example, a straightness error that exists between linear axes (such as between the X axis and the Y axis) cannot be corrected by the method disclosed in Patent Document 1.
Similarly, in the correction for correcting the rotational axis deviation amount and the main axis turning center deviation amount described in Patent Document 2, as described above, the translation error (δx) existing between the reference coordinate system and the actual coordinate system. , Δy, δz) and rotation error (α, β, γ) cannot be corrected accurately.
Accordingly, an object of the present invention is to provide a numerical control device capable of correcting all errors existing between a reference coordinate system that defines a command machining shape and a coordinate system during actual machining.

The invention according to claim 1 of the present application provides a reference machine coordinate provided in a numerical control apparatus for controlling a 5-axis processing machine that processes a workpiece attached to a table by three linear axes and two rotary axes. The correction amount (ΔX, ΔY) of the tool tip point position from the translation error (δx, δy, δz) and rotation error (α, β, γ) of the actual machine coordinate system and the linear axis machine coordinate position and the tool length correction vector , ΔZ) ^T , a correction amount calculating means for obtaining ^T , a correction means for obtaining a linear axis correction machine coordinate position based on the correction amounts (ΔX, ΔY, ΔZ) ^T , and each linear axis to the linear axis correction machine coordinate position. Each rotary shaft is provided with means for driving to the commanded rotational position, and performs error correction of the tool tip point position.

  According to a second aspect of the present invention, the correction amount calculation means calculates the correction amount of the tool tip point position for each interpolation period. Furthermore, the invention according to claim 3 further includes a determination unit that determines whether or not the correction amount for the tool tip point position is obtained in the interpolation cycle, and the determination unit determines that the correction amount for the tool tip point position is obtained. The correction amount calculation means obtains and updates and stores the correction amount of the tool tip position, and the correction means obtains the linear axis correction machine coordinate position for each interpolation period based on the stored correction amount. It was.

The invention according to claim 4 includes means for setting error reference values for translation errors (δx, δy, δz) and rotation errors (α, β, γ), and the translation errors (δx, δy, δz) and The rotation error (α, β, γ) can be changed, and includes means for obtaining a difference between the translation error (δx, δy, δz) and the rotation error (α, β, γ) before and after the change, The judging means determines the tool tip position when the difference between the translation error (δx, δy, δz) or the rotation error (α, β, γ) before and after the change is larger than the set error reference value. It was determined that this is the period for obtaining the correction amount.
According to a fifth aspect of the present invention, there is provided means for setting the number of interpolation cycles, and the judging means judges that the correction amount for the tool tip point position is determined every time the set number of interpolation cycles.
The invention according to claim 6 is characterized in that the tool length correction vector is a reference tool length correction vector T (Tx, Ty, Tz) ^T commanded in the reference machine coordinate system and an actual tool length correction vector in the actual machine coordinate system. T ′ (Tx ′, Ty ′, Tz ′) ^T. According to a seventh aspect of the present invention, the reference tool length correction vector is calculated based on an error for an actual tool length correction vector comprising a translation error (Tδx, Tδy, Tδz) and a rotation error (Tα, Tβ, Tγ). It was determined from the correction vector. According to an eighth aspect of the present invention, the actual tool length correction vector is derived from the reference tool length correction vector by an actual tool length correction vector error comprising a tool direction error (ΔTa) and a tool perpendicular direction error (ΔTrx, ΔTry). I wanted it. In the invention according to claim 9, the actual tool length correction vector is obtained from the reference tool length correction vector by an error for an actual tool length correction vector composed of linear axis direction errors (ΔTz, ΔTx, ΔTy) ^T. .

  Since the error between the reference coordinate system that defines the command machining shape and the actual coordinate system at the time of machining can be corrected by the tool tip point position, which is the machining point, highly accurate machining can be performed.

First, the principle of the present invention and the basic algorithm will be described. FIG. 1 is an explanatory diagram of the principle and basic algorithm of the present invention. In FIG. 1, the code | symbol 10 is a tool and the code | symbol 11 is a tool head. Further, the linear axis machine coordinate position Pm commanded by the numerical control device will be described with reference to FIG.
When translation errors (δx, δy, δz) and rotation errors (α, β, γ) between the reference coordinate system that is the basis for commanding the machining shape and the actual coordinate system that is actually machined are given, From the machine coordinate position Pm (Pmx, Pmy, Pmz) ^T and the current tool length correction vector T (Tx, Ty, Tz) ^T , the correction amount ΔD (ΔX, ΔY of the tool tip point position by the following equation (5) , ΔZ) ^T is obtained.

すなわち、式（３）で示される２つの座標系間の誤差を示す誤差行列εと、現在の直線軸機械座標位置Ｐｍ(Ｐmx，Ｐmy，Ｐmz) ^Tから現在の工具長補正ベクトルＴ(Ｔx，Ｔy，Ｔz) ^Tを減じて得られる工具先端点位置Ｐｔ（Ｐtx，Ｐty，Ｐtz）^Tとにより、工具先端点位置Ｐｔ（Ｐtx，Ｐty，Ｐtz）^Tの補正量ΔＤ(ΔＸ，ΔＹ，ΔＺ) ^Tを求める。なお、各式におけるベクトルや行列は同次座標系で表記しているが、最後の項の０や１が自明の場合は特に記載しない。また、^Tは転置を表すが自明の場合は^Tを記載しない場合がある。特に図面では、この^Tを記載していない。

That is, from the error matrix ε indicating the error between the two coordinate systems represented by Expression (3) and the current linear axis machine coordinate position Pm (Pmx, Pmy, Pmz) ^T , the current tool length correction vector T (Tx, Ty, Tz) A correction amount ΔD (ΔX, ΔY, ΔZ) of the tool tip point position Pt (Ptx, Pty, Ptz) ^T based on the tool tip point position Pt (Ptx, Pty, Ptz) ^T obtained by subtracting ^{T. Find T.} In addition, although the vector and matrix in each formula are described in the homogeneous coordinate system, they are not particularly described when 0 or 1 in the last term is obvious. ^T represents transposition, but ^T may not be described if it is obvious. In particular, this ^T is not described in the drawings.

Then, the linear axis corrected machine coordinate position Pm ′ (Pmx ′, Pmy ′, Pmz ′) ^T is obtained by the following equation (6), and each linear axis is driven to that position. The rotating shaft is driven to the commanded position. As a result, the translation error (δx, δy, δz) and the rotation error (α, β, γ) at the tool tip point position Pt (Ptx, Pty, Ptz) ^T , which is the machining point, are corrected and high-precision machining is performed. Made.

基準座標系に対する実座標系の並進誤差（δx，δy，δz）と回転誤差（α，β，γ）は基準座標系に対する実座標系のすべての誤差を含ませることができる。つまり、回転軸のずれ量、及び／又は主軸旋回中心のずれ量や、直線軸間（Ｘ軸とＹ軸の間など）に存在する真直度誤差も含めて、基準座標系に対する実座標系のすべての誤差を表現することができ、この補正を行うことによって、精度の高い加工を実施できる。

The translation error (δx, δy, δz) and rotation error (α, β, γ) of the real coordinate system with respect to the reference coordinate system can include all errors of the real coordinate system with respect to the reference coordinate system. That is, the actual coordinate system relative to the reference coordinate system, including the amount of deviation of the rotation axis and / or the amount of deviation of the turning center of the spindle and the straightness error existing between the linear axes (such as between the X axis and the Y axis). All errors can be expressed, and by performing this correction, highly accurate machining can be performed.

The tool tip point position Pt (Ptx, Pty, Ptz) ^T is obtained by driving each linear axis to the linear axis corrected machine coordinate position Pm ′ (Pmx ′, Pmy ′, Pmz ′) ^T obtained by the above equation (6). The reason why the translation error (δx, δy, δz) ^T and the rotation error (α, β, γ) are corrected is as follows.

  First, a position conversion matrix for adding the unit matrix E to the error matrix ε and converting the actual machine coordinate value to the reference machine coordinate value is M (M = E + ε). The position conversion matrix M is expressed by the following equation (7).

図１において、基準座標系上の工具先端点位置Ｐｔ(Ｐtx，Ｐty，Ｐtz) ^T、実座標系上の工具先端点位置Ｐｔ'(Ｐtx'，Ｐty'，Ｐtz') ^T、工具長補正ベクトルＴ(Ｔx，Ｔy，Ｔz) ^Tは、次の関係となる。

In FIG. 1, the tool tip point position Pt (Ptx, Pty, Ptz) ^T on the reference coordinate system, the tool tip point position Pt '(Ptx', Pty ', Ptz') ^T on the real coordinate system, and the tool length correction vector T (Tx, Ty, Tz) ^T has the following relationship.

Pt ′ = M ⁻¹ (Pm′−T)
= M ⁻¹ (Pm + ΔD−T)
= M ⁻¹ (Pm−T + ε (Pm−T))
= M ^-1 (E + ε) (Pm-T)
= M ^-1 M (Pm-T)
= Pt
It becomes. That is, the tool center point position Pt on the reference coordinate system (Ptx, Pty, Ptz) ^T and the tool center point position Pt on the real coordinate system '(Ptx', Pty ', Ptz') T are equal.
Accordingly, it is possible to correct all errors at the tool tip point (machining point) by driving to the corrected machine coordinate position Pm ′ (Pmx ′, Pmy ′, Pmz ′) ^T , and perform highly accurate machining. It means that you can.
Here, even if the conventional correction means 5 shown in FIGS. 9 and 10 performs correction by coordinate conversion to a real coordinate system having an error with respect to the reference coordinate system that defines the command machining shape, the tool tip point position is corrected. The fact that there is no correction will also be explained using mathematical formulas. In FIG. 11, the tool tip point position (command position) Ptc (Ptcx, Ptcy, Ptcz) ^T on the reference coordinate system, the tool tip point position Ptc ′ (Ptcx ′, Ptcy ′, Ptcz ′) ^T on the real coordinate system, Tool length compensation vector T (Tx, Ty, Tz) ^T , linear axis machine coordinate position Pmc (Pmcx, Pmcy, Pmcz) ^T , modified linear axis machine coordinate position Pmc ′ (Pmcx ′, Pmcy ′, Pmcz ′) ^T Is the following relationship. Here, ΔD is an error between Pmc and Pmc ′, which is the same as in the equation (5).
Ptc ′ = M ⁻¹ (Pmc′−T)
= M ⁻¹ (Pmc + ΔD−T)
= M ⁻¹ (Pmc−T + εPmc)
= M ⁻¹ (Pmc−T + ε (Pmc−T) + εT)
= M ⁻¹ M (Pmc−T) + M ⁻¹ εT
= Ptc + M ⁻¹ εT
That is, the Ptc 'Ptc is different only M ^-1 εT.
Further, as shown in FIG. 13, the tool length correction vector is the reference tool length correction vector T (Tx, Ty, Tz) ^T commanded by the program in the reference machine coordinate system, but the actual machine coordinate system has an error. Tool length compensation vector T ′ (Tx ′, Ty ′, Tz ′) ^T In other words, the actual tool length correction vector T ′ (Tx ′, Ty ′, Tz ′) ^T in the actual machine coordinate system is programmed in the reference machine coordinate system due to deflection, thermal expansion or expansion, or assembly error due to the tool's own weight. There may be an error with respect to the reference tool length correction vector T (Tx, Ty, Tz) ^T. In this case, the correction amount ΔD (ΔX, ΔY, ΔZ) ^T is calculated by the following equation (8) instead of the equation (5).

Next, the principle of operation in which the present invention is applied to various 5-axis machines will be described.
FIG. 2 is a diagram for explaining the operation principle of the present invention in a tool head rotating machine that is controlled by the numerical controller of the present invention. This rotary tool head type machine includes a linear axis X, Y, and Z axes, a rotary axis B that rotates the tool 10 around the Y axis, and a rotary axis C that rotates around the Z axis. . Here, the actual table position of the table 12 to which the workpiece is attached has an error with respect to the reference table position. The machine coordinate system corresponding to the actual table position is the actual machine coordinate system, and the machine coordinate system corresponding to the reference table position is the reference machine coordinate system. The actual machine coordinate system has a translation error (δx, δy, δz) and a rotation error (α, β, γ) with respect to the reference machine coordinate system. The tool 10 is inclined by the rotation axes B and C, and the tool length correction vector T (Tx, Ty, Tz) ^T has a value corresponding to the inclination of the tool 10.

In this case, the translation error (δx, δy, δz) and rotation error (α, β, γ) of the actual machine coordinate system with respect to the reference machine coordinate system are given, and the current linear axis machine coordinate value Pm (Pmx, Pmy, Pmz). ) ^T and the current tool length correction vector T (Tx, Ty, Tz) From the ^T and the actual tool length correction vector error, the correction amount ΔD (ΔX, ΔY, ΔZ) seek ^T. In the tool head 11, the point where the B-axis rotation center axis and the C-axis rotation center axis intersect is the linear axis machine coordinate position Pm (Pmx, Pmy, Pmz) ^T.

Then equation (6) linear scale modification machine coordinate position Pm by '(Pmx', Pmy ', Pmz') T a determined to that location for driving each linear axis. Further, the B-axis and C-axis of the rotation shaft are driven to the commanded rotation position. As a result, the tool tip point position Pt (Ptx, Pty, Ptz) ^T on the reference machine coordinate system and the tool tip point position Pt ′ (Ptx ′, Pty ′, Ptz ′) ^T on the actual machine coordinate system become equal. . As a result, it is possible to perform highly accurate machining at the tool tip point (machining point) on the workpiece placed on the actual table 12.

The correction amount of the tool tip point position may be obtained for each interpolation cycle. However, the translation error (δx, δy, δz) and the rotation error (α, β, γ) can be changed and before the change. And the difference between the translation error (δx, δy, δz) and the rotation error (α, β, γ) after the change, and the translation error (δx, δy, δz) and the rotation error before and after the change ( When the difference between [alpha], [beta], and [gamma] is larger than a set error reference value, the correction amount of the tool tip position can be obtained and updated.
Further, the number of times may be set, and the correction amount of the tool tip position may be obtained and updated every set number of interpolation cycles.

FIG. 3 is a diagram for explaining the operating principle of the present invention in a table rotating machine that is controlled by the numerical controller of the present invention. This table rotation type machine has X, Y, and Z axes as linear axes, a rotation axis A axis that rotates the table 12 to which the workpiece is attached, around the X axis, and a rotation axis C axis that rotates around the Z axis. Is provided. Also in this example, the actual table position has an error with respect to the reference table position. The machine coordinate system corresponding to the actual table position is the actual machine coordinate system, and the machine coordinate system corresponding to the reference table position is the reference machine coordinate system. The actual machine coordinate system has a translation error (δx, δy, δz) and a rotation error (α, β, γ) with respect to the reference machine coordinate system. Since the tool is not tilted, only Tz has a non-zero value for the tool length correction vector T (Tx, Ty, Tz) ^T , and Tx and Ty are zero.

In this case, the translation error (δx, δy, δz) and the rotation error (α, β, γ) of the actual machine coordinate system with respect to the reference machine coordinate system are given, and the current machine coordinate value Pm (Pmx, Pmy, Pmz) ^T And the current tool length correction vector T (Tx, Ty, Tz) ^T and the actual tool length correction vector error, the correction amount ΔD (ΔX, ΔY, ΔZ of the tool tip point position according to Equation (5) or Equation (8). ) ^{Find T} Then, the corrected machine coordinate value Pm ′ (Pmx ′, Pmy ′, Pmz ′) ^T is obtained by the equation (6), and the linear axes X, Y, and Z axes are driven to the positions. Further, the C-axis and A-axis of the rotation shaft are driven to the commanded rotation position. As a result, the tool tip point position Pt (Ptx, Pty, Ptz) ^T on the reference machine coordinate system is equal to the tool tip point position Pt ′ (Ptx ′, Pty ′, Ptz ′) ^T on the actual machine coordinate system. . As a result, it is possible to perform highly accurate machining at the tool tip point (machining point) on the workpiece placed on the actual table 12.

FIG. 4 is a diagram illustrating the principle of operation of the present invention in a mixed machine that is a control target of the numerical controller of the present invention. This mixed type machine includes an X, Y, and Z axis linear axes, a rotation axis C axis that rotates the table 12 to which the workpiece is attached, around the Z axis, and a rotation axis B axis that rotates the tool 10 around the Y axis. It is equipped with.
The actual table position of the table 12 has an error with respect to the reference table position. The machine coordinate system corresponding to the actual table position is the actual machine coordinate system, and the machine coordinate system corresponding to the reference table position is the reference machine coordinate system. The actual machine coordinate system has a translation error (δx, δy, δz) and a rotation error (α, β, γ) with respect to the reference machine coordinate system. Since the tool 10 is tilted about the rotation axis B about the Y axis, Tx and Tz have non-zero values for the tool length correction vector T (Tx, Ty, Tz) ^T and Ty is zero.

In this case, the translation error (δx, δy, δz) and rotation error (α, β, γ) of the actual machine coordinate system with respect to the reference machine coordinate system are given, and the current linear axis machine coordinate position Pm (Pmx, Pmy, Pmz). ) From the ^T , the current tool length correction vector T (Tx, Ty, Tz) ^T, and the error for the actual tool length correction vector, the correction amount ΔD (ΔX, ΔY of the tool tip point position according to the equation (5) or (8) , ΔZ) ^T is obtained. Then, the linear axis corrected machine coordinate position Pm ′ (Pmx ′, Pmy ′, Pmz ′) ^T is obtained by Expression (6), and the linear axes X, Y, and Z axes are driven to that position. Further, the C-axis and A-axis of the rotation shaft are driven to the commanded rotation position. As a result, the tool tip point position on the reference machine coordinate system is equal to the tool tip point position on the actual machine coordinate system. As a result, it is possible to perform highly accurate machining at the tool tip point (machining point) on the workpiece placed on the actual table.
FIG. 14 shows that in the numerical control apparatus of the present invention, the actual tool length correction vector is based on the reference tool length based on the error for the actual tool length correction vector consisting of translation errors (Tδx, Tδy, Tδz) and rotation errors (Tα, Tβ, Tγ). It is operation | movement principle explanatory drawing in the case of calculating | requiring from a correction vector. Here, if the tool head coordinate system in the reference machine coordinate system is expressed by (Xt, Yt, Zt) coordinates, the tool head coordinate system becomes (Xt ', Yt', Zt ') coordinates in the actual machine coordinate system, and these coordinates It shows that the errors between the systems are translation errors (Tδx, Tδy, Tδz) and rotation errors (Tα, Tβ, Tγ). In this case, the actual tool length correction vector T ′ (Tx ′, Ty ′, Tz ′) ^T is obtained by calculating from the reference tool length correction vector T (Tx, Ty, Tz) ^{T as} follows. In addition, in FIG. 14, it is the figure of the tool head in a tool head rotary type machine as an Example. Therefore, the tool head has two rotation axes.

図１５は、本発明の数値制御装置において実工具長補正ベクトルは工具方向誤差（ΔＴa）と工具直角方向誤差（ΔＴrx，ΔＴry）から成る実工具長補正ベクトル用誤差によって前記基準工具長補正ベクトルから求める場合の動作原理説明図である。この場合、実工具長補正ベクトルＴ’（Ｔx’，Ｔy’，Ｔz’）^Tは基準工具長補正ベクトルＴ（Ｔx，Ｔy，Ｔz）^Tから次のように計算して求める。ここで、(eax，eay，eaz^)Tは工具方向の単位ベクトル、(erxx，erxy，erxz) ^T，(eryx，eryy，eryz) ^Tは工具直角方向の単位ベクトルである。(erxx，erxy，erxz) ^T，(eryx，eryy，eryz) ^Tは互いに独立とする。なお、図１５では実施例として混合型機械における工具ヘッドの図としている。したがって工具ヘッドには１軸の回転軸がある。

FIG. 15 shows that in the numerical control apparatus of the present invention, the actual tool length correction vector is derived from the reference tool length correction vector by the actual tool length correction vector error composed of the tool direction error (ΔTa) and the tool perpendicular direction error (ΔTrx, ΔTry). It is operation | movement principle explanatory drawing in the case of calculating | requiring. In this case, the actual tool length correction vector T ′ (Tx ′, Ty ′, Tz ′) ^T is obtained by calculating from the reference tool length correction vector T (Tx, Ty, Tz) ^{T as} follows. Here, (eax, eay, eaz ^{) T} is a unit vector in the tool direction, and (erxx, erxy, erxz) ^T and (eryx, eryy, eryz) ^T are unit vectors in the tool perpendicular direction. (erxx, erxy, erxz) ^T and (eryx, eryy, eryz) ^T are independent of each other. In addition, in FIG. 15, it is the figure of the tool head in a mixed type machine as an Example. Therefore, the tool head has a single rotation axis.

図１６は、本発明の数値制御装置において実工具長補正ベクトルは工具先端点における直線軸方向誤差（ΔＴz，ΔＴx，ΔＴy）^Tから成る実工具長補正ベクトル用誤差によって前記基準工具長補正ベクトルから求める場合の動作原理説明図である。この場合、実工具長補正ベクトルＴ’（Ｔx’，Ｔy’，Ｔz’）^Tは基準工具長補正ベクトルＴ（Ｔx，Ｔy，Ｔz）^Tから次のように計算して求める。なお、図１６では実施例としてテーブル回転型機械における工具ヘッドの図としている。したがって工具ヘッドには回転軸はない。

FIG. 16 shows that in the numerical control apparatus of the present invention, the actual tool length correction vector is derived from the reference tool length correction vector by the error for the actual tool length correction vector including the linear axis direction error (ΔTz, ΔTx, ΔTy) ^{T at the} tool tip point. It is operation | movement principle explanatory drawing in the case of calculating | requiring. In this case, the actual tool length correction vector T ′ (Tx ′, Ty ′, Tz ′) ^T is obtained by calculating from the reference tool length correction vector T (Tx, Ty, Tz) ^{T as} follows. In addition, in FIG. 16, it is the figure of the tool head in a table rotary type machine as an Example. Therefore, the tool head has no rotation axis.

FIG. 5 is a principal block diagram of the numerical controller according to the first embodiment of this invention. The first embodiment is an example of a numerical control device that performs numerical control in accordance with the tool length correction command shown in FIG. 9. Compared with FIG. 9, an input data unit 7 that inputs data to the correction means 5 is provided. It differs in that it has been added.
FIG. 6 is a principal block diagram of the numerical controller according to the second embodiment of the present invention. This second embodiment is an example of a numerical control device that performs numerical control by the tool tip point control shown in FIG. 10. Compared with FIG. 10, an input data unit 7 for inputting data to the correction means 5 is provided. It differs in that it has been added.

The input data unit 7 includes a translation error (δx, δy, δz) and a rotation error (α, β, γ) of the actual machine coordinate system with respect to the reference machine coordinate system, a tool length correction vector T (Tx, Ty, Tz) ^T. Furthermore, an error reference value for determining the timing for updating the tool tip point position correction amount ΔD (ΔX, ΔY, ΔZ) ^T and the number of interpolation cycles are input to the correction means 5. Here, the tool length correction vector is a reference tool length correction vector T (Tx, Ty, Tz) ^T commanded by a program in the reference machine coordinate system, but an actual tool length correction vector T ′ () having an error in the actual machine coordinate system. If Tx ′, Ty ′, Tz ′) ^T , the tool length compensation vector T (Tx, Ty, Tz) ^T and the reference tool length compensation vector T (Tx, Ty, Tz) ^T and the actual tool length Correction vector T ′ (Tx ′, Ty ′, Tz ′) An error for ^T is input to the correction means 5.

The translation error (δx, δy, δz), rotation error (α, β, γ) and actual tool length correction vector error of the actual machine coordinate system with respect to the reference machine coordinate system are obtained in advance corresponding to the machine coordinate position. When the machining is started, it is stored in the memory of the numerical controller. Further, the tool length correction vector T (Tx, Ty, Tz) ^T has been conventionally obtained when command analysis processing and interpolation processing are performed, and is used. In the first embodiment shown in FIG. 5, the command analysis unit 1 has conventionally performed an operation of adding the tool length correction vector T (Tx, Ty, Tz) ^T to the tool tip point position. The obtained tool length correction vector T (Tx, Ty, Tz) ^T is used. Further, in the second embodiment shown in FIG. 6, the process of adding the tool length correction vector T (Tx, Ty, Tz) ^T to the tool tip position interpolated by the interpolation means 3 has been conventionally performed. The tool length correction vector T (Tx, Ty, Tz) ^T obtained by this process is used.
Furthermore, the translation error (δx, δy, δz), the rotation error (α, β, γ), and the actual tool length correction vector error are stored in the memory of the numerical controller when starting machining as described above. In addition to the case of being provided from the outside of the numerical control device so as to be stored in the interface memory of the numerical control device through a communication line from a device outside the numerical control device such as a personal computer as shown in FIG. There is also.

Further, an error reference value for determining the timing for updating the tool tip point position correction amount ΔD (ΔX, ΔY, ΔZ) ^T and the number of interpolation cycles are set in advance. The correction amount ΔD (ΔX, ΔY, ΔZ) ^T may be updated every interpolation cycle in which the movement command is output, but the translation error and the rotation error are not large values. Since the correction amount ΔD (ΔX, ΔY, ΔZ) ^T for correcting these does not necessarily need to be updated every interpolation period, this update timing is set.

When the error reference value is set, the difference between the input translation error (δx, δy, δz) and the stored translation error (δx, δy, δz), or the input rotation error (α, β, γ) When the stored difference in rotation error (α, β, γ) exceeds the set error reference value, the correction amount ΔD (ΔX, ΔY, ΔZ) ^T for correcting the tool tip position is updated. An error reference value is set in order to update the amount ΔD (ΔX, ΔY, ΔZ) ^T.
The number of interpolation cycles is to set the timing for updating the tool tip point position correction amount ΔD (ΔX, ΔY, ΔZ) ^T as the number of interpolation cycles, and if this number of interpolation cycles is set to “1”. The correction amount ΔD (ΔX, ΔY, ΔZ) ^T is updated every interpolation cycle. If “n” is set, the correction amount ΔD (ΔX, ΔY, ΔZ) ^T is updated every n interpolation cycles. It is set so that.

In the case of the first embodiment shown in FIG. 5, the command analysis unit 1 analyzes the program block, and the tool length correction vector addition element 2 corrects the tool length with respect to the command position of the tool tip point position of the analysis result. The vector T (Tx, Ty, Tz) ^T is added to obtain a linear axis machine coordinate position Pm (Pmx, Pmy, Pmz) ^T to which the tool length correction vector is added. Linear axis machine coordinate position Pm (Pmx, Pmy, Pmz) T and generates a machine coordinate position of each axis performs interpolation by the interpolation means 3 with respect to the commanded rotational position of the rotary shaft, each shaft (X, Y , Z, B (A), C axis) acceleration / deceleration processing units 4x, 4y, 4z, 4b (a), 4c, respectively, and the linear axis command machine after the acceleration / deceleration processing of each axis is completed. An error correction process is executed by the correction means 5 on the coordinate position Pm (Pmx, Pmy, Pmz) ^T to obtain a corrected linear axis machine coordinate position Pm ′ (Pmx ′, Pmy ′, Pmz ′) ^T. Each linear axis X, Y, Z axis is the linear axis machine coordinate position Pm ′ (Pmx ′, Pmy ′, Pmz ′) ^T , and the rotation axis B (A) axis and C axis are commands after the acceleration / deceleration processing is completed. The servo motors 6x, 6y, 6z, 6b (a), 6c of each axis are driven to the rotational position.

In the case of the second embodiment shown in FIG. 6, the command analysis unit 1 analyzes the program block, the command tool tip point position of the analysis result is interpolated by the interpolation means 3, and the tool tip point position is interpolated. The tool length correction vector addition element 2 adds the tool length correction vector T (Tx, Ty, Tz) ^T to the linear axis machine coordinate position Pm (Pmx, Pmy, Pmz) ^T , and the linear axis machine coordinate position Pm (Pmx, Pmy, Pmz) T and interpolated after the rotational position of the rotary shaft was performed each deceleration at each deceleration processing section 4X～4c, linear axis machine coordinate position Pm (Pmx, Pmy, Pmz ) An error correction process is performed on the ^T by the correcting means 5 to obtain a corrected machine coordinate position Pm ′ (Pmx ′, Pmy ′, Pmz ′) ^T to obtain servo motors 6x, 6y, 6z, 6b for each axis. (A), 6c is driven. Note that the command position after acceleration / deceleration processing output from the acceleration / deceleration processing units 4b (a) and 4c is output as it is for the B (A) axis and C-axis of the rotation axis.

FIG. 7 is a flowchart showing an error correction algorithm for correcting an error of the actual machine coordinate system with respect to the reference machine coordinate system executed by the correction means 5 in the first and second embodiments described above. In this flowchart, the correction amount ΔD (ΔX, ΔY, ΔZ) ^T is updated based on the set error reference values (δx0, δy0, δz0) and (α0, β0, γ0).

The processor of the numerical controller executes the processing shown in FIG. 7 for each interpolation period.
First, the translation error (δx, δy, δz), rotation error (α, β, γ) and actual tool length correction vector error for the tool tip position commanded by the program are read out and stored in the register. Tool length correction vector T (Tx, Ty, Tz) ^T is read (step S1). Next, it is determined whether or not the correction amount ΔD (ΔX, ΔY, ΔZ) ^T of the tool tip point position is to be updated (step S2). In this embodiment, this determination is made based on the set error reference values (δx0, δy0, δz0), (α0, β0, γ0). First, the translation error (δx, The translation error (δxp, δyp, δzp) stored in the register is subtracted from δy, δz), and it is determined whether the absolute value of each element exceeds the set translation error reference value (δx0, δy0, δz0). To do. Further, the rotation error (αp, βp, γp) stored in the register is subtracted from the rotation error (α, β, γ) read in step S1, and the absolute value of each element is used as a reference for the set rotation error. Judge whether the value (α0, β0, γ0) is exceeded. If any of the translation error or rotation error difference elements exceeds the set reference value, the tool tip position correction amount ΔD (ΔX, ΔY, ΔZ) ^T is updated and the process proceeds to step S3. . If none of them exceeds the set reference value, the process proceeds to step S5.
Note that the translation error (δxp, δyp, δzp) and rotation error (αp, βp, γp) stored in the register are initially set to “0” at the initial setting of machining. In addition, the correction amount ΔD (ΔX, ΔY, ΔZ) ^{T of} the tool tip point position stored in the register is initially set to “0” as an initial setting.

If “No” in step S2, that is, if neither the difference in translation error nor the difference in rotation error exceeds the set reference value, the process proceeds to step S5, where each axis (X, Y, Z, B (A), C axis) linear axis machine coordinate position Pm (Pmx, Pmy, Pmz) ^T accelerated / decelerated by acceleration / deceleration processing units 4x, 4y, 4z, and tool tip point position stored in register Based on the quantity ΔD (ΔX, ΔY, ΔZ) ^T , the calculation of the equation (6) is performed to obtain the linear axis corrected machine coordinate position Pm ′ (Pmx ′, Pmy ′, Pmz ′) ^T (step S5). The linear axis corrected machine coordinate position Pm '(Pmx', Pmy ' , Pmz') outputs a ^T X, Y, servomotors 6x the Z-axis, 6y, and drives the 6z. For the B (A) and C axes of the rotary shaft, the command position after the acceleration / deceleration processing output from the acceleration / deceleration processing units 4b (a) and 4c is output as it is, and the servo motor 6b of the rotary shaft is output. (A), 6c is driven.

  On the other hand, if “Yes” in step S2, that is, if either the translation error difference or the rotation error difference exceeds the set reference value, the translation error (δx, δy, δz), rotation error (α, β, γ), translation error (δxp, δyp, δzp), rotation error (αp, βp, γp) at the previous change in order to determine the change in error in step S2. Is stored in the register and updated (step S3).

Next, the linear axis machine coordinate position Pm (Pmx, Pmy, Pmz) ^T subjected to acceleration / deceleration processing and the translation error (δx, δy, δz), rotation error (α, β, γ), tool length correction vector T (Tx, Ty, Tz) ^T and the actual tool length correction vector error are used to perform the calculation of Expression (5) or Expression (8), and the tool tip point position correction amount ΔD ( [Delta] X, [Delta] Y, [Delta] Z) ^T is obtained, stored in the register, and the correction amount [Delta] D ([Delta] X, [Delta] Y, [Delta] Z) ^T is updated (step S4). Thereafter, the process proceeds to step S5, and the process of step S5 described above is executed.

As described above, when neither the difference in translation error nor the difference in rotation error exceeds the set reference value, the processing of steps S1, S2, and S5 is performed to determine the tool tip point position stored in the register. On the basis of the correction amount ΔD (ΔX, ΔY, ΔZ) ^T , the calculation of the equation (6) is performed, and the linear axis corrected machine coordinate position Pm ′ (Pmx ′, Pmy ′, Pmz ′) ^T is obtained and output. The X, Y, and Z axis servo motors 6x, 6y, and 6z are driven. In addition, the rotational axes B, A, and C are output as they are as commanded rotational positions subjected to acceleration / deceleration processing, and the servomotors 6b, 6a, and 6c of the respective rotational axes are driven. Each time either the translation error difference (change amount) or the rotation error difference (change amount) exceeds the set reference value, the processing of steps S1, S2, S3 to S5 is performed, and the tool tip point position is determined. Correction amount ΔD (ΔX, ΔY, ΔZ) ^T is updated, and by this correction amount ΔD (ΔX, ΔY, ΔZ) ^T , the linear axis corrected machine coordinate position Pm ′ (Pmx ′, Pmy ′, Pmz ′) ^T It will be required.

In the example shown in FIG. 7, the correction amount ΔD (ΔX, ΔY, ΔZ) ^T of the tool tip position is set as the update timing, and the translation error difference (change amount) or rotation error difference (change amount) is set. If the interpolation cycle exceeds the reference value, but the correction amount ΔD (ΔX, ΔY, ΔZ) ^T of the tool tip point position is updated every interpolation cycle, the processing in steps S2 and S3 is eliminated and the step What is necessary is just to make it transfer to step S4 from S1. When the correction amount ΔD (ΔX, ΔY, ΔZ) ^T of the tool tip point position is updated every set number of interpolation cycles, a counter (1 increment) that counts the interpolation cycle instead of steps S2 and S3. Counter is set to determine whether the counter value is equal to or greater than the set value. When the counter value is equal to or greater than the set value, the counter is cleared to “0”, the process proceeds to step S4, and the processes of steps S4 and S5 When the counter does not reach the set value, the process of step S4 is not executed, and only the process of step S5 is executed.
Furthermore, in step S2, it is also possible to make a determination on the error for the actual tool correction vector. For example, when the error for the actual tool correction vector is composed of a tool direction error (ΔTa) and a tool right angle direction error (ΔTrx, ΔTry), if NO is determined in step S2 in FIG. 7, a further set tool direction error reference value is set. The determination is made based on (ΔTa0) and the tool perpendicular direction error reference value (ΔTrx0, Δtry0). First, the tool direction error (ΔTap) stored in the register is subtracted from the tool direction error (ΔTa) read in step S1, and it is determined whether the absolute value of the element exceeds the set tool direction error reference value (ΔTa0). . Also, the tool perpendicular direction error reference value (ΔTrx0) in which the absolute value of each element is set by subtracting the tool perpendicular direction error (ΔTrxp, ΔTryp) stored in the register from the tool perpendicular direction error (ΔTrx, ΔTry) read in step S1. , ΔTry0). If any of the elements of the difference between the tool direction error and the tool perpendicular direction error exceeds the set reference value, the process proceeds to step S5, but the tool tip point position correction amount ΔD (ΔX, ΔY, ΔZ) ^T The process proceeds to step S3. If none of them exceeds the set reference value, the process proceeds to step S5. In step S3, the current tool direction error (ΔTa) and the tool perpendicular direction error (ΔTrx, ΔTry) are stored in the register ΔTap, the registers ΔTrxp, and ΔTryp.

It is explanatory drawing of the principle of this invention and a basic algorithm. It is operation | movement principle explanatory drawing when this invention is applied to a tool head rotary machine. It is operation | movement principle explanatory drawing when this invention is applied to a table rotary type machine. It is operation | movement principle explanatory drawing when this invention is applied to a mixed type machine (machine in which a tool head and a table rotate together). It is a principal part block diagram of the numerical control apparatus of the 1st Embodiment of this invention. It is a principal part block diagram of the numerical control apparatus of the 2nd Embodiment of this invention. It is a flowchart which shows the algorithm of the error correction process by the correction | amendment means in each embodiment of this invention. A point on the reference coordinate system and a point on the reference coordinate system after coordinate conversion from the reference coordinate system to the real coordinate system when the real coordinate system has a rotation error and a translation error with respect to the reference coordinate system. FIG. It is a general | schematic block diagram of the operation | movement process of the conventional numerical control apparatus in the case of a tool length correction | amendment command. It is a general | schematic block diagram of the operation | movement process of the conventional numerical control apparatus in the case of tool tip point control. It is explanatory drawing at the time of correct | amending the error which arises by an assembly error etc. with the conventional numerical control apparatus. It is explanatory drawing in case a translation error ((delta) x, (delta) y, (delta) z), a rotation error ((alpha), (beta), (gamma)), and the error for actual tool length correction | amendment vectors are given from the outside of a numerical controller. It is explanatory drawing when the actual tool length correction vector in an actual machine coordinate system has an error. It is explanatory drawing in case the error for actual tool length correction | amendment vectors consists of a translation error and a rotation error. It is explanatory drawing in case the error for actual tool length correction | amendment vectors consists of a tool direction error and a tool orthogonal direction error. It is explanatory drawing in case the error for actual tool length correction | amendment vectors consists of a linear axis direction error.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Command analysis part 2 Tool length correction | amendment vector addition element 3 Interpolation means 4x-4c Acceleration / deceleration processing part 5 Correction means 6x-6c Servo motor 7 Input data part 10 Tool 11 Tool head 12 Table

Claims (9)

In a numerical control apparatus for controlling a 5-axis processing machine that processes a workpiece attached to a table by three linear axes and two rotary axes,
From the translation error (δx, δy, δz) and rotation error (α, β, γ) of the real machine coordinate system to the given reference machine coordinate system, the linear axis machine coordinate position and the tool length correction vector, the tool tip position correction amount (ΔX, ΔY, ΔZ) and the correction amount calculating means for calculating the ^T, the correction amount (ΔX, ΔY, ΔZ) a correction means for obtaining a linear scale modification machine coordinate position based on the ^T, each linear axes the line A numerical control device having means for driving each rotational axis to the commanded rotational position to the axis correction machine coordinate position.   The numerical control device according to claim 1, wherein the correction amount calculation means calculates a correction amount of the tool tip position for each interpolation period.   And determining means for determining whether the correction amount of the tool tip point position is obtained in an interpolation cycle, and the correction amount calculating means is configured to determine whether the correction amount of the tool tip point position is obtained in the determination means. 2. The numerical control apparatus according to claim 1, wherein a correction amount of the tip point position is obtained, updated and stored, and the correction means obtains the linear axis corrected machine coordinate position for each interpolation period based on the stored correction amount.   Means for setting error error reference values of translation error (δx, δy, δz) and rotation error (α, β, γ), and the translation error (δx, δy, δz) and rotation error (α, β, γ) ) Can be changed, and has means for obtaining a difference between the translation error (δx, δy, δz) and the rotation error (α, β, γ) before and after the change, and the determination means is changed before and after the change. When the difference between the subsequent translation error (δx, δy, δz) or rotation error (α, β, γ) is larger than the set error reference value, it is determined as a cycle for obtaining the correction amount of the tool tip position. The numerical controller according to claim 3.   4. The numerical control apparatus according to claim 3, further comprising means for setting the number of interpolation cycles, wherein the determination unit determines that a cycle for obtaining a correction amount for the tool tip position for each set number of interpolation cycles. The tool length correction vector includes a reference tool length correction vector T (Tx, Ty, Tz) ^T commanded in the reference machine coordinate system and an actual tool length correction vector T ′ (Tx ′, Ty ′) in the actual machine coordinate system. , Tz ′) ^T , the numerical control device according to claim 1.   7. The numerical value according to claim 6, wherein the actual tool length correction vector is obtained from the reference tool length correction vector by an error for an actual tool length correction vector comprising a translation error (Tδx, Tδy, Tδz) and a rotation error (Tα, Tβ, Tγ). Control device.   7. The numerical controller according to claim 6, wherein the actual tool length correction vector is obtained from the reference tool length correction vector based on an error for an actual tool length correction vector including a tool direction error (ΔTa) and a tool perpendicular direction error (ΔTrx, ΔTry). 7. The numerical controller according to claim 6, wherein the actual tool length correction vector is obtained from the reference tool length correction vector by an error for an actual tool length correction vector consisting of a linear axis direction error (ΔTz, ΔTx, ΔTy) ^T.

Images