JP3579742B2 - Teaching data correction method, correction program and correction device - Google Patents

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a teaching data correction method, a correction program, a recording medium storing the correction program, and a correction device. More specifically, a teaching data correction method, a correction program, and a recording in which the correction program is stored for correcting an error of teaching data by an offline teaching method, particularly an error caused by deformation of a work, created based on CAD data. It relates to a medium and a correction device.
[0002]
[Prior art]
In recent years, as a teaching method for teaching industrial robots to perform predetermined tasks and operations, instead of teaching using actual robots and works, teaching data is created and created using a computer such as a personal computer. Off-line teaching for giving learned teaching data to a robot has attracted attention from the viewpoint of labor saving and productivity improvement.
[0003]
Such off-line teaching can create teaching data by operating only a personal computer, so that it is not necessary to stop the production line when performing teaching work, and it is effective for improving productivity, and for example, Even when there are a large number of operation points, such as a drilling operation for drilling rivet holes in an aircraft wing, there is an advantage that teaching data can be created in a relatively short time and thus the teaching operation can be made more efficient.
[0004]
On the other hand, since the actual work and the robot are not used to teach each work point using actual work and the robot as in direct teaching and teaching playback methods, work installation errors, robot mechanical variations, work Normally, errors occur between the teaching data and the actual position and direction of each work point due to the effects of deflection due to weight and tool mounting errors, and the teaching data is corrected to eliminate such errors. There is a problem that needs to be done.
[0005]
Conventionally, as a method of correcting teaching data in such an off-line teaching method, (1) a predetermined number of reference points are provided on a work, an installation error of the work is calculated by detecting a position of the reference point, and this is used. (2) a method of sensing the position of each work point when performing an operation, and correcting the teaching data based on the result of the sensing (in addition to performing a cutting process). Japanese Patent Application Laid-Open No. H11-114771 discloses a method of correcting NC data in consideration of a predictable distortion of a work generated at the time of teaching. As described in JP-A-5-326364).
[0006]
However, the correction method (1) can only eliminate errors caused by misalignment of the work and cannot correct errors caused by deformation such as thermal expansion and distortion of the work.
[0007]
In the correction method (2), when a large number of work points are present, such as in the case of drilling a rivet hole, a large amount of time is required for a work for sensing the work point and the like. However, there is a problem that it becomes an obstacle to improvement of work efficiency.
[0008]
[Problems to be solved by the invention]
SUMMARY OF THE INVENTION The present invention has been made in view of the above-described problems of the related art, and it is possible to rationally detect and estimate a position of each work point and a shift of a work direction at each work point due to deformation of a work, thereby improving efficiency. It is an object of the present invention to provide a teaching data correction method, a correction program, a recording medium storing the correction program, and a correction device capable of correcting the offline teaching data in an accurate and appropriate manner.
[0009]
[Means for Solving the Problems]
A teaching data correction method according to the present invention is a teaching data correction method for correcting teaching data for causing a robot to perform a predetermined operation, which is created based on design data of a work, and includes two measurement points provided on the work. Location and Work at each point Direction, position data of each detected measurement point and Work at each point Based on the direction data, the actual position data of each work point and Actual work at each point The direction data is estimated, and the position data of each estimated working point and Work at each point The teaching data is corrected based on the direction data.
[0010]
In the teaching data correction method of the present invention, the position data of each detected measurement point and Work at each point Position data and position data created based on workpiece design data Work at each point Calculate each deviation amount from the direction data, and calculate the position deviation amount and Work at each point Assuming that the direction deviation changes according to the positional relationship with each measurement point, the position deviation of each working point and Work at each point The direction deviation amount is estimated from each of the calculated deviation amounts, and the position deviation amount and work The position data of each working point and the Work at each point The direction data is estimated, and the position data of each estimated working point and Work at each point The teaching data is corrected based on the direction data, and the position data and Work at each point Direction data , Position data created based on work design data and work Calculate each deviation amount from the direction data, set a correction reference point corresponding to each work point on a line connecting the measurement points, according to the positional relationship between each work point and the measurement point, Calculate each deviation amount at the correction reference point assuming that the calculated deviation amounts change proportionally on the line segment, and based on the calculated deviation amounts, position data of each working point and Work at each point The direction data is estimated, and the position data of each estimated working point and Work at each point The teaching data is corrected based on the direction data, and the position data and Work at each point Direction data , Position data created based on work design data and work Calculate each deviation amount from the direction data, set a correction reference point corresponding to each work point on a line connecting the respective measurement points according to the ratio of the distance from the measurement point of each work point, Calculate each deviation amount at the correction reference point assuming that the calculated deviation amounts change proportionally on the line segment, and based on the calculated deviation amounts, position data of each working point and Work at each point The direction data is estimated, and the position data of each estimated working point and Work at each point Preferably, the teaching data is corrected based on the direction data.
[0011]
The teaching data correction program according to the present invention is a teaching data correction program that is created based on work design data and corrects teaching data for causing a robot to perform a predetermined operation. Location of two measurement points and Work at each point A position / direction detection procedure for detecting a direction, position data of each detected measurement point and Work at each point Based on the direction data, the actual position data of each work point and Actual work at each point A position / direction estimation procedure for estimating the direction data, and the position data and Work at each point And a correction procedure for correcting the teaching data based on the direction data.
[0012]
In the teaching data correction program of the present invention, the position / direction estimation procedure includes the position data of each measurement point detected by the position / direction detection procedure. work Position data and position data created based on workpiece design data Work at each point Each deviation amount from the direction data is calculated, and assuming that the position deviation amount and the direction deviation amount of each working point change according to the positional relationship with each measurement point, the position deviation amount of each working point and Work at each point Estimate the direction deviation amount, the estimated position deviation amount and work The position data of each working point and the Work at each point The direction data is estimated, and the position / direction estimation procedure is performed by using the position data of each measurement point detected by the position / direction detection procedure. Work at each point Position data and position data created based on workpiece design data work Calculate each deviation amount from the direction data, set a correction reference point corresponding to each work point on a line connecting the respective measurement points according to the positional relationship between each work point and the measurement point, Assuming that each deviation amount changes proportionally on the line segment, each deviation amount at the correction reference point is calculated, and based on each of the calculated deviation amounts, , Position data of each work point and Work at each point The direction data is estimated, and the position / direction estimation procedure is performed by using the position data of each measurement point detected by the position / direction detection procedure. Work at each point Position data and position data created based on workpiece design data work Calculate each deviation amount from the direction data, set a correction reference point corresponding to each work point on a line connecting the measurement points according to the ratio of the distance from the measurement point of each work point, Calculate each deviation amount at the correction reference point assuming that each deviation amount changes proportionally on the line segment, and based on the calculated deviation amounts, position data of each working point and Work at each point Preferably, the direction data is estimated.
[0013]
A recording medium according to the present invention is characterized by storing the teaching data correction program.
[0014]
A teaching data correction device according to the present invention is a teaching data correction device that is created based on design data of a work and corrects teaching data for causing a robot to perform a predetermined operation. Location and Work at each point Detecting means for detecting a direction, position data of each of the detected measurement points, and Work at each point Based on the direction data, the actual position of each work point and Actual work at each point The direction is estimated, and the position data of each of the estimated working points and Work at each point And a correcting means for correcting the teaching data based on the direction data.
[0015]
In the teaching data correction device according to the present invention, the correction means includes position data of each detected measurement point and Work at each point Position data and position data created based on workpiece design data work Calculate each deviation amount from the direction data, and calculate the position deviation amount and Work at each point Assuming that the direction deviation changes according to the positional relationship with each measurement point, the position deviation of each working point and Work at each point Estimate the direction deviation amount, the position deviation amount and work The position data of each working point and the Work at each point The direction data is estimated, and the position data of each estimated working point and Work at each point The teaching data is corrected based on the direction data, and the correction unit detects the position data and the position data of each detected measurement point. Work at each point Direction data , Position data created based on work design data and work Calculate each deviation amount from the direction data, set a correction reference point corresponding to each work point on a line connecting the measurement points, according to the positional relationship between each work point and the measurement point, Calculate each deviation amount at the correction reference point assuming that each deviation amount proportionally changes on the line segment, and based on the calculated deviation amounts, position data of each working point and Work at each point The direction data is estimated, and the position data of each estimated working point and Work at each point The teaching data is corrected based on the direction data, and the correction unit detects the position data and the position data of each detected measurement point. Work at each point Direction data , Position data created based on work design data and work Calculate each deviation amount from the direction data, set a correction reference point corresponding to each work point on a line connecting the respective measurement points according to the ratio of the distance from the measurement point of each work point, Calculate each deviation amount at the correction reference point assuming that each deviation amount changes proportionally on the line segment, and based on the calculated deviation amounts, position data of each working point and Work at each point The direction data is estimated, and the position data of each estimated working point and Work at each point Preferably, the teaching data is corrected based on the direction data.
[0016]
[Action]
Since the present invention is configured as described above, the work efficiency can be improved only by providing a small number of measurement points in the preceding work process, and the position of each work point can be rationally determined from the actual measurement data at each measurement point. -Since the direction is estimated, the teaching data can be corrected with appropriate accuracy.
[0017]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, the present invention will be described based on embodiments with reference to the accompanying drawings, but the present invention is not limited to only such embodiments.
[0018]
FIG. 1 shows a schematic configuration of a robot control system to which a teaching data correction method according to an embodiment of the present invention is applied. This robot control system (hereinafter, simply referred to as a system) A drills rivet holes in, for example, an aircraft wing. In order to perform a predetermined operation such as drilling operation using an industrial robot, use a computer such as a personal computer to create teaching data that indicates the position and orientation of the robot at each work point, and use this teaching data. A so-called off-line teaching type control system that controls the robot to perform the predetermined work.
[0019]
That is, the system A uses a CAD (Computer Aided Design) device 10 that designs a work W as a manufacturing target and creates a three-dimensional model thereof, and a three-dimensional model of the work W created by the CAD device 10. It comprises a CAM (Computer Aided Manufacturing) device 20 for creating teaching data by means of a robot, a robot 30, and a vision sensor 40 as main components.
[0020]
The CAD device 10 includes various programs for supporting a work of designing and modeling a work W having required functions and performances, and data indicating the position and direction of each work point from a created three-dimensional model of the work W. (Example: the coordinate of the center position of the rivet hole on the surface of the work W and each component value of the direction vector indicating the drilling direction) are to be generated and output (hereinafter, each of the working points generated by the CAD device 10). The data indicating the position and the data indicating the direction are referred to as theoretical position data and theoretical direction data, respectively).
[0021]
The CAM device 20 is composed of various programs that support creation of teaching data for instructing the position and orientation of the robot 30 at each work point using the theoretical position data and the theoretical direction data from the CAD device 10. Various programs constituting the CAD apparatus 10 and the CAM apparatus 20 are specifically incorporated into a computer such as a personal computer and operate.
[0022]
The robot 30 includes a robot controller 31, a six-axis articulated manipulator 32, and a tool 33 such as a drill mounted on the manipulator 32.
[0023]
The vision sensor 40 includes, for example, a charge coupled device (CCD) imaging device and a pattern recognition / evaluation processing operation unit (both not shown), and captures each measurement point formed on the work W by a predetermined marking method. The position and direction of each measurement point are detected using image information obtained by performing predetermined image processing on the captured image, and the detection signal is output to the robot controller 31.
[0024]
FIG. 2 shows a detailed configuration of the robot controller.
[0025]
The robot controller 31 controls the manipulator 32 according to the teaching data corrected by the teaching data correction unit 34 and the teaching data correction unit 34 that corrects the teaching data created by the CAM device 20 based on the detection signal from the vision sensor 40. And an NC control unit 35 that performs the operation.
[0026]
Hereinafter, the teaching data correction process performed by the teaching data correction unit 34 will be described by taking, as an example, a case of performing a drilling operation of drilling a large number of rivet holes in a work W composed of an aircraft wing. For simplicity, the description will be made in a two-dimensional space, but the actual correction is performed in three dimensions.
[0027]
1. Basic concept of teaching data correction processing
[0028]
As shown in FIG. 3A, in the teaching data correction process, the theoretical position data and the theoretical direction data of each of the working points P1, P2,..., P5 of the work W are converted into the respective coordinate positions in the robot coordinate system xz. P ₁ , P ₂ , ..., P ₅ (Hereinafter, normal subscripts indicate the names of points, and subscripts indicate data) and each direction vector B ₁ , B ₂ , ..., B ₅ For example, when the working points P1 and P5 at both ends are marked by a predetermined method to form measurement points, the coordinate positions of the measurement points P1 and P5, as shown in FIG. And the direction vector are detected by the vision sensor 40, and the detected coordinate positions P ₁ ´, P ₅ 'And each direction vector B ₁ ´, B ₅ (Hereinafter referred to as detection position data and detection direction data), as shown in FIG. 3C, the coordinate positions P of the working points P2, P3, and P4 other than the measurement points P1 and P5. ₂ ´, P ₃ ´, P ₄ 'And direction vector B ₂ ´, B ₃ ´, B ₄ ′, And the estimated coordinate positions P ₂ ´, P ₃ ´, P ₄ 'And each direction vector B ₂ ´, B ₃ ´, B ₄ '(Hereinafter referred to as estimated position data and estimated direction data) and the detected position data and detected direction data of each of the measurement points P1 and P5 described above to correct the teaching data from the CAM device 20. It is said.
[0029]
Hereinafter, a specific procedure of the teaching data correction processing in the embodiment will be described.
[0030]
2. Instruction data correction procedure
[0031]
(Procedure 1): Obtain position data and direction data of each work point of the work W from the CAD device 10.
[0032]
(Procedure 2): As shown in FIG. 4, for example, two working points P1, P7 at both ends among a plurality of working points P1, P2,. By marking in advance in the process, two measurement points P1 and P7 are formed.
[0033]
(Procedure 3): Position data P of each measurement point P1, P7 by the vision sensor 40 ₁ ´, P ₇ 'And direction data B ₁ ´, B ₇ 'Is detected.
[0034]
(Procedure 4): Detection position data P acquired in Procedure 3 ₁ ´, P ₇ 'And detection direction data B ₁ ´, B ₇ , The position data P of each of the working points P2, P3..., P6 other than the measurement points P1, P7 ₂ ´, P ₃ ´,…, P ₆ 'And direction data B ₂ ´, B ₃ ', ..., B ₆ ′ Is estimated in consideration of the influence of the displacement of the work W and the distortion of the work W, and the estimated position data P ₂ ´, P ₃ ´,…, P ₆ 'And estimated direction data B ₂ ´, B ₃ ', ..., B ₆ ', The teaching data is corrected. Hereinafter, the gist of the estimation processing of the procedure 4 will be described.
[0035]
2-1. Key points of the estimation process
[0036]
Hereinafter, for the sake of simplicity, the estimation process will be described taking only the influence of the deformation of the work W as a problem.
[0037]
(Point 1): The detected position data and the detected direction data of each of the measurement points P1 and P7 are compared with the theoretical position data and the theoretical direction data, and the detected position data deviates from the theoretical position data (hereinafter, referred to as position deviation amount) and A deviation of the detected direction data from the theoretical direction data (hereinafter, referred to as a direction deviation amount) is calculated. The position deviation amount is calculated as three-dimensional data, and the direction deviation amount is calculated as a vector amount holding two-dimensional data.
[0038]
(Point 2): Assuming that the positional deviation amount and the directional deviation amount of each of the working points P2, P3,..., P6 change according to the positional relationship with each of the measuring points P1, P7, each of the working points P2, P3,. , P6, and the position data and the direction data of each of the working points P2 to P6 are estimated by using the position deviation and the direction deviation to correct the teaching data.
[0039]
When the position deviation amount and the direction deviation amount of each of the working points P2, P3,..., P6 change proportionally according to the distance from each of the measurement points P1, P7, the following procedure is used. The teaching data is corrected by estimating the position deviation amount and the direction deviation amount at P2, P3,..., P6.
[0040]
(Point 3): Correction reference points are set for working points P2, P3,..., P6 other than the measurement points P1, P7. For example, as shown in FIG. 4B, the ratio of the distances between the points P1 to P7 based on the theoretical position data, that is, <P ₁ P ₂ >: <P ₂ P ₃ ＞ ：… ： 〈P ₆ P ₇ 〉 Line segment P ₁ P ₇ , And these internal dividing points R2, R3,..., R6 are set as correction reference points for the corresponding working points P2 to P6. Note that <P ₁ P ₂ >, <P ₂ P ₃ >,…, <P ₆ P ₇ 〉 Is a line segment P ₁ P ₂ , Line segment P ₂ P ₃ , ..., P ₆ P ₇ Indicates the length of each. This is the same in the following.
[0041]
(Point 4): The position deviation amount and the direction deviation amount at each of the correction reference points R2 to R6 are calculated as being proportionally changed according to the distance from each of the measurement points P1 and P7. The position and direction of each of the working points P2 to P6 are estimated using the position deviation amount and the direction deviation amount at the points R2 to R6, and the teaching data is corrected.
[0042]
(Point 5): Line segment P connecting measurement points P1 and P7 when estimating the position and direction of each of work points P2 to P6. ₁ P ₇ Derivation of a component of the amount of deviation in a direction perpendicular to the direction (hereinafter, referred to as a vertical deviation component) is determined by specifically considering the material of the work W and the like. For example,
[0043]
(Key point 5.1) When the whole work W expands uniformly, the vertical deviation components of the work points P2 to P6 are calculated by directly reflecting the elongation rate between the measurement points P1 and P7.
[0044]
(Point 5.2) The elongation rate of the work W is P ₁ P ₇ In the case where the direction is different between the direction parallel to the vertical direction and the direction perpendicular to the vertical direction and the two are in a proportional relationship, the elongation percentage in the parallel direction is multiplied by a proportional coefficient to calculate the vertical deviation component of each of the working points P2 to P6. Calculate.
[0045]
(Point 5.3) The elongation percentage between each measurement point P1 and P7 is the line segment P ₁ P ₇ In the case where it is inversely proportional to the elongation percentage in the direction perpendicular to the vertical direction, the vertical deviation component of each of the working points P2 to P6 is calculated based on the reciprocal of the elongation percentage between the measurement points P1 and P7.
[0046]
(Procedure 5.4) The elongation rate between each measurement point P1 and P7 is a line segment P ₁ P ₇ If the elongation percentage in the direction perpendicular to the direction is inversely proportional to a proportionality coefficient other than 1, the reciprocal of the elongation percentage between each of the measurement points P1 and P7 is multiplied by the proportionality coefficient. The vertical deviation component of P6 is calculated.
[0047]
Hereinafter, the estimation process will be described using each specific example.
[0048]
3. Specific examples of working point estimation
[0049]
As shown in FIG. 5A, the theoretical position data and the theoretical direction data of each of the working points P11, P12, and P13 of the workpiece W are represented by coordinates P. ₁₁ , P ₁₂ , P _Thirteen And direction vector B ₁₁ , B ₁₂ , B _Thirteen And the position and direction of the intermediate work point P12 are estimated using the work points P11 and P13 at both ends as measurement points.
[0050]
The theoretical distance between the measurement points P11 and P13 is represented by a value L. ₁ , The theoretical distance between the points P11 and P12 is m1, and the theoretical distance between the points P12 and P13 is m2. The correction reference point of the working point P12 is represented by a symbol R12, and the theoretical distance between the points P11 and R12 is represented by a value (t × L ₁ ) (Where t: 0 <t <1), the following equation (1) holds from the above point 3.
[0051]
m1: m2 = t: 1-t (1)
[0052]
At this time, the theoretical distance between the points P12 and R12 is the value L. ₂ It is assumed that
[0053]
Here, as shown in FIG. 3B, when the position of the work W is simply shifted to the position indicated by the dotted line W ′, the teaching data is corrected by a known process based on translation and rotation. It is possible to do. Accordingly, in the following description, a procedure for estimating the position data and the direction data of each work point by taking into account only the error caused by the deformation of the work W, ignoring the position shift of the work W will be described.
[0054]
In the following description, a situation in which a workpiece is deformed in a plane will be described in order to simplify the description, but actual distortion occurs three-dimensionally, and both position and direction deviation have a three-dimensional component, This method corresponds to this three-dimensional distortion.
[0055]
3-1. When the deformation of the work W is only expansion or contraction
[0056]
As shown in FIG. 5C, the positions of the measurement points P11 and P13 after the deformation of the work W are ₁₁ ´, P _Thirteen ', But the direction data of each of the measurement points P11 and P13 is the original B ₁₁ , B _Thirteen Consider the case where it remains. At this time, assuming that the elongation ratio between the measurement points P11 and P13 is a value α (α: 0 <α <1 for reduction, 1 <α for expansion), a function using α as an argument f (α) is calculated according to the key point 5 described above.
[0057]
Thereby, the actual position P of the working point P12 ₁₂ 'Is the measurement position P of the measurement points P11 and P13 ₁₁ ´, P _Thirteen The point R which internally divides the space between m1: m2 ₁₂ Distance from f ′ (α) × L ₂ Estimated as distant points. If the work W expands or contracts uniformly in this example, f (α) = α and the point P ₁₂ Is estimated.
[0058]
In addition, since the estimated direction data at the work point P12 is considered to have no twist in the workpiece W, the original B ₁₂ Will be left as is.
[0059]
3-2. When the work W is twisted
[0060]
If the detected direction data at each of the measurement points P11 and P13 is different from the theoretical direction data, the direction data of the working point P12 is also corrected accordingly. That is, the detection direction data at each of the measurement points P11 and P13 is represented by B ₁₁ ´, B _Thirteen And the angle difference θ around the y-axis, where each direction deviation amount forms the z-axis. ₁ , Θ ₂ , The directional deviation θ of the working point P12 at the correction reference point R12. _a Is calculated by the following equation (2).
[0061]
θ _a = (1-t) x θ ₁ + T × θ ₂ (2)
[0062]
At this time, assuming that the detected position data of each of the measurement points P11 and P13 is not different from the theoretical position data, the estimated working point P ₁₂ 'And correction reference point R ₁₂ Is the value L ₂ It is presumed to remain.
[0063]
4. Correction example when there are multiple working points between each measuring point
[0064]
4-1. When the work W expands or contracts
[0065]
As shown in FIG. 6A, there are a plurality of working points P22 and P23 between the measurement points P21 and P24, and the ratio of the distance between the points P21 to P24, that is, <P ₂₁ P ₂₂ >: <P ₂₂ P ₂₃ >: <P ₂₃ P ₂₄ > Is a: b: c (a, b, c: a> 0, b> 0, c> 0 and a + b + c = 1).
[0066]
In a state where the work W is not deformed, the correction reference points R22 and R23 of the work points P22 and P23 are defined by the line segment P such that the following equation (3) is satisfied. ₂₁ P ₂₄ Set above.
[0067]
<P ₂₁ R ₂₂ >: <R ₂₂ R ₂₃ >: <R ₂₃ P ₂₄ > = A: b: c (3)
[0068]
Further, each theoretical direction data of each point P21, P22, P23, P24 is converted into a direction vector B. ₂₁ , B ₂₂ , B ₂₃ , B ₂₄ And the direction vector B of each working point P22, P23 ₂₂ , B ₂₃ Are the angles around the y-axis that the z-axis makes ₂₁ , Θ ₂₂ And the line segment R ₂₂ P ₂₂ And line segment R ₂₃ P ₂₃ Are the angles around the y-axis that the z-axis makes ₂₃ , Θ ₂₄ And the line segment R ₂₂ P ₂₂ And line segment R ₂₃ P ₂₃ Is the value of L ₂₁ , Value L ₂₂ And
[0069]
Here, as shown in FIG. 3B, the position of each measurement point P21, P24 is changed to the point P by the expansion of the work W. ₂₁ ´, P ₂₄ ′, The elongation rate α of the work W between the measurement points P21 and P24 is calculated by the following equation (4).
[0070]
α = <P ₂₁ 'P ₂₄ '> / <P ₂₁ P ₂₄ ＞ (4)
[0071]
At this time, the line segment P ₂₁ 'P ₂₄ ′ Is internally divided into a: b: c and the correction reference point R of each of the working points P22 and P23. ₂₂ ´, R ₂₃ 'Is set. Position P of working point P22 after deformation of work W ₂₂ 'Is the correction reference point R ₂₂ Angle θ from ´ ₂₃ Distance in the direction of α × L ₂₁ And the position P of the work point P23 after the deformation of the work W ₂₃ Is an angle θ from the correction reference point R23 ′. ₂₄ Distance in the direction of α × L ₂₂ It is estimated that
[0072]
Also, a direction vector B indicating each direction data of each point P21, P22, P23, P24 ₂₁ ´, B ₂₂ ´, B ₂₃ ´, B ₂₄ 'Is each direction vector B of the theoretical direction data ₂₁ , B ₂₂ , B ₂₃ , B ₂₄ And are respectively the same.
[0073]
4-2. When the work W is twisted
[0074]
As shown in FIG. 7A, there are a plurality of working points P32 and P33 between the measurement points P31 and P34, and the ratio of the distance between the points P31 to P34, that is, <P ₃₁ P ₃₂ >: <P ₃₂ P ₃₃ >: <P ₃₃ P ₃₄ > Is a: b: c (a, b, c: a> 0, b> 0, c> 0 and a + b + c = 1).
[0075]
In a state where the work W is not deformed, the correction reference points R32 and R33 of the working points P32 and P33 are determined by the line segment P such that the following equation (5) is satisfied. ₃₁ P ₃₄ Set above.
[0076]
<P ₃₁ R ₃₂ >: <R ₃₂ R ₃₃ >: <R ₃₃ P ₃₄ > = A: b: c (5)
[0077]
Further, each of the theoretical direction data of the points P31, P32, P33, and P34 is ₃₁ , B ₃₂ , B ₃₃ , B ₃₄ And the direction vector B of each working point P32, P33 ₃₂ , B ₃₃ Are the angles around the y-axis that the z-axis makes ₃₁ , Θ ₃₂ And the line segment R ₃₂ P ₃₂ And line segment R ₃₃ P ₃₃ Are the angles around the y-axis that the z-axis makes ₃₃ , Θ ₃₄ And the line segment R ₃₂ P ₃₂ And line segment R ₃₃ P ₃₃ Is the value of L ₃₁ , Value L ₃₂ And
[0078]
Here, as shown in FIG. 3B, the detection direction data (direction vector) at each of the measurement points P31 and P34 is B ₃₁ ´, B ₃₄ ′, And the direction deviation amount is the angle θ ₃₅ , Θ ₃₆ The direction deviation amount θ corresponding to each of the working points P32 and P33 by the following equations (6) and (7) according to the main point 2 ₃₇ , Θ ₃₈ Is calculated.
[0079]
θ ₃₇ = (1-a) × θ ₃₅ + A × θ ₃₆ (6)
[0080]
θ ₃₈ = C × θ ₃₅ + (1-c) × θ ₃₆ (7)
[0081]
At this time, the estimated positions of the working points P32 and P33 (hereinafter referred to as estimated working points) P estimated based on the detected position data and the detected direction data of the respective measurement points P31 and P34. ₃₂ ´, P ₃₃ 'Is the line segment R ₃₂ P ₃₂ And line segment R ₃₃ P ₃₃ Is the angle θ corresponding to the direction deviation amount about each of the correction reference points R32 and R33. ₃₇ , Θ ₃₈ Each is set as a rotated position. Also, the point R ₃₂ ´, P ₃₂ The distance between 'is the value L ₃₁ And the point R ₃₃ ´, P ₃₃ The distance between 'is the value L ₃₂ It is estimated to remain.
[0082]
Also, the estimated direction data (vector) B of each estimated working point P32 ', P33' ₃₂ ´, B ₃₃ Is the angle about the y-axis that the z-axis makes with the z-axis ₃₁ ´, θ ₃₂ 'Is calculated by the following equations (8) and (9).
[0083]
θ ₃₁ ´ = θ ₃₁ + Θ ₃₇ (8)
[0084]
θ ₃₂ ´ = θ ₃₂ + Θ ₃₈ (9)
[0085]
4-3. When the work W expands or contracts and twists occur
[0086]
As shown in FIG. 8A, there are a plurality of working points P42 and P43 between the measurement points P41 and P44, and the ratio of the distance between the points P41 to P44, ie, <P ₄₁ P ₄₂ >: <P ₄₂ P ₄₃ >: <P ₄₃ P ₄₄ > Is a: b: c (a, b, c: a> 0, b> 0, c> 0 and a + b + c = 1).
[0087]
In a state where the work W is not deformed, the correction reference points R42 and R43 of the working points P42 and P43 are line segments P so that the following equation (10) is satisfied. ₄₁ P ₄₄ Set above.
[0088]
<P ₄₁ R ₄₂ >: <R ₄₂ R ₄₃ >: <R ₄₃ P ₄₄ > = A: b: c (10)
[0089]
Further, each theoretical direction data of each point P41, P42, P43, P44 is converted into a direction vector B. ₄₁ , B ₄₂ , B ₄₃ , B ₄₄ , The direction vector B of each working point P42, P43 ₄₂ , B ₄₃ Are the angles around the y-axis that the z-axis makes ₄₁ , Θ ₄₂ And the line segment R ₄₂ P ₄₂ And line segment R ₄₃ P ₄₃ Are the angles around the y-axis that the z-axis makes ₄₃ , Θ ₄₄ And the line segment R ₄₂ P ₄₂ And line segment R ₄₃ P ₄₃ Is the value of L ₄₁ , Value L ₄₂ And
[0090]
Here, as shown in FIG. 3B, the positions of the measurement points P41 and P44 are respectively set to the point P based on the detection position data at the measurement points P41 and P44. ₄₁ ´, P ₄₄ ′, The elongation α of the workpiece W between the measurement points P41 and P44 is calculated by the following equation (11).
[0091]
α = <P ₄₁ 'P ₄₄ '> / <P ₄₁ P ₄₄ ＞ (11)
[0092]
In addition, the estimated working point P of each of the working points P42 and P43 according to the main point 3 described above. ₄₂ ´, P ₄₃ ', A correction reference point (hereinafter referred to as an estimated correction reference point) R ₄₂ ´, R ₄₃ 'Is a line segment P satisfying the following equation (12). ₄₁ 'P ₄₄ 'Is set on
[0093]
<P ₄₁ 'R ₄₂ '>: <R ₄₂ 'R ₄₃ '>: <R ₄₃ 'P ₄₄ '> = A: b: c (12)
[0094]
Furthermore, the direction deviation amount at each of the measurement points P41 and P44 is the angle θ. ₄₅ , Θ ₄₆ , The direction deviation amount θ corresponding to each of the estimated working points P42 and P43 is calculated according to the following equations (13) and (14) according to the main point 2. ₄₇ , Θ ₄₈ Is calculated.
[0095]
θ ₄₇ = (1-a) × θ ₄₅ + A × θ ₄₆ (13)
[0096]
θ ₄₈ = C × θ ₄₅ + (1-c) × θ ₄₆ (14)
[0097]
At this time, the line segment R ₄₂ 'P ₄₂ 'And line segment R ₄₃ 'P ₄₃ ′ Each make an angle θ with the z-axis ₄₉ , Θ ₅₀ Is calculated by the following equations (15) and (16).
[0098]
θ ₄₉ = Θ ₄₃ + Θ ₄₇ (15)
[0099]
θ ₅₀ = Θ ₄₄ + Θ ₄₈ (16)
[0100]
The estimated distance between each point R42 and P42 is α × L ₄₁ And the estimated distance between the points R43 and P43 is α × L ₄₂ It is estimated that
[0101]
Also, each direction vector B indicating the direction data of each estimated working point P42, P43 ₄₂ ´, B ₄₃ ´ is the angle θ with the z-axis ₄₁ ´, θ ₄₂ 'Is calculated by the following equations (17) and (18).
[0102]
θ ₄₁ ´ = θ ₄₁ + Θ ₄₇ (17)
[0103]
θ ₄₂ ´ = θ ₄₂ + Θ ₄₈ (18)
[0104]
As described above, according to the embodiment of the present invention, for example, among the working points on one cross section of the workpiece W, for example, the working points at both ends are formed as measurement points in advance in the preceding work process, and the teaching data The correction unit 34 estimates the position data and direction data of each working point other than the measurement point based on the position data and direction data of each measurement point detected by the vision sensor 40, and detects the detected position data of each measurement point. Since the teaching data is corrected using the detection direction data and the estimated position data and the estimated direction data of each work point, the teaching data by the offline teaching method can be corrected efficiently by saving labor in the preceding work process. In addition, the deviation amount of the position data and the direction data at each working point is rationally estimated using the deviation amount from the theoretical position data and the theoretical direction data of the detection position data and the detection direction data of each measurement point, and Since the teaching data is corrected based on the estimation result, it is possible to correct the teaching data with appropriate accuracy and control the robot.
[0105]
Further, the teaching data correction method of the present embodiment can estimate the position and direction of each work point even when the deformation amount of the work W cannot be predicted in advance due to a change in the temperature condition of the work place or the like. is there. In this point, the method disclosed in Japanese Patent Application Laid-Open No. H11-114771 differs from the method in which only the expected deformation amount of the work W is dealt with.
[0106]
Further, even when each work point is a point on a curved surface such as an aircraft wing, its position and direction can be estimated. In this point, the method disclosed in Japanese Patent Application Laid-Open No. 5-326364 is different from the method in which each work point on a uniform plane such as the surface of a base is targeted.
[0107]
As described above, the present invention has been described based on the embodiments. However, the present invention is not limited to the embodiments, and various modifications are possible.
[0108]
For example, in the embodiment, the actual position and direction of each work point are determined based on a point obtained by internally dividing a line segment connecting each measurement point by a ratio of a distance between a series of points formed by each measurement point and each work point. Although it is assumed that the actual position and direction of each work point is estimated based on a point internally dividing the line segment connecting each measurement point based on the ratio of the distance from each measurement point to each work point, Good.
[0109]
Further, in the embodiment, the work points between the two measurement points are targeted, but the work points outside the two measurement points may be corrected by providing a correction reference point at a position outside each measurement point. It is possible to do.
[0110]
【The invention's effect】
As described above, according to the present invention, when correcting teaching data generated based on design data of a work so as to eliminate an error caused by deformation of the work, the position of two measurement points provided on the work is corrected. A direction is detected, and based on the detected position data and direction data of each measurement point, an actual position and direction of a work point other than each measurement point are estimated, and the estimated position data of each work point is estimated. Since the teaching data is corrected based on the direction data and the direction data, it is possible to reduce the labor required for the preceding work for correction and to improve the working efficiency.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a schematic configuration of a robot control system to which a teaching data correction method according to an embodiment of the present invention is applied.
FIG. 2 is a block diagram showing details of a robot controller.
FIG. 3 is a schematic diagram for explaining the basic concept of the teaching data correction method of the present invention.
FIG. 4 is a schematic diagram for explaining a procedure of the teaching data correction method.
FIG. 5 is a schematic diagram for explaining a specific procedure of a work point estimation process in the teaching data correction method.
FIG. 6 is a schematic diagram for explaining an example of a specific procedure of a work point estimation process in the teaching data correction method.
FIG. 7 is a schematic diagram for explaining another example of the specific procedure of the work point estimation process in the teaching data correction method.
FIG. 8 is a schematic diagram for explaining still another example of the specific procedure of the work point estimation process in the teaching data correction method.
[Explanation of symbols]
10 CAD equipment
20 CAM device
30 robots
31 Robot Controller
34 Teaching data correction unit
40 Vision Sensor
W Work

Claims (13)

A teaching data correction method for correcting teaching data for causing a robot to perform a predetermined operation, which is created based on work design data, comprising: a position of two measurement points provided on the work; and a work direction at each point. And based on the detected position data of each measurement point and the work direction data at each point , estimate the actual position data of each work point and the actual work direction data at each point. And correcting the teaching data based on the position data of each working point and the working direction data at each point . Calculate the deviation of the detected position data of each measurement point and the work direction data at each point from the position data and work direction data created based on the design data of the work, and calculate the position deviation of each work point. and as the working direction deviation amount at the respective points are changed depending on the positional relationship between the respective measurement points, the respective deviation amount of said calculated work direction deviation at the position deviation amount and the respective points of each working point estimated, it estimates the working direction data in the position data and the respective points of each working point by the positional deviation amount and the working direction deviation, in the working direction data in the position data and the respective points of each working point which is the estimated 2. The teaching data correction method according to claim 1, wherein the teaching data is corrected based on the teaching data. The working direction data in the position data and the respective points of each measurement point is detected, it calculates the deviation amount from the position data and the working direction data is created based on the work of the design data, connecting the respective measurement points line A correction reference point corresponding to each work point is set in accordance with the positional relationship between each work point and the measurement point, and each of the calculated deviation amounts changes proportionally on the line segment. Calculating each deviation amount at the correction reference point, estimating the position data of each work point and the work direction data at each point based on the calculated deviation amounts, and calculating the estimated position of each work point. 2. The teaching data correction method according to claim 1, wherein the teaching data is corrected based on the data and the work direction data at each point . The working direction data in the position data and the respective points of each measurement point is detected, it calculates the deviation amount from the position data and the working direction data is created based on the work of the design data, connecting the respective measurement points line A correction reference point corresponding to each work point is set in accordance with the ratio of the distance of each work point from the measurement point, and the calculated deviation amounts change proportionally on the line segment. Assuming that each deviation amount at the correction reference point is calculated, the position data of each work point and the work direction data at each point are estimated based on the calculated deviation amounts, and the estimated work point 2. The teaching data correction method according to claim 1, wherein the teaching data is corrected based on the position data and the work direction data at each point . A teaching data correction program for correcting teaching data for causing a robot to perform a predetermined operation, which is created based on work design data,
On the computer,
A position / direction detection procedure for detecting the position of two measurement points provided on the work and the work direction at each point ;
Based on the detected position data of each measurement point and the work direction data at each point, a position / direction estimation procedure for estimating the actual position data of each work point and the actual work direction data at each point ,
A correction procedure for correcting teaching data based on the estimated position data of each work point and the work direction data at each point . Position and direction estimation steps, the working direction data in the position data and the respective points of each measurement point detected by the position and direction detection procedure, each of the position data and the working direction data is created based on the work of the design data The deviation amount is calculated, and assuming that the position deviation amount of each working point and the working direction deviation amount at each point change according to the positional relationship with each measurement point, the position deviation amount of each working point and the 6. The teaching data according to claim 5, wherein the work direction deviation is estimated, and the position data of each work point and the work direction data at each point are estimated based on the estimated position deviation and work direction deviation. Correction program. Position and direction estimation steps, the working direction data in the position data and the respective points of each measurement point detected by the position and direction detection procedure, each of the position data and the working direction data is created based on the work of the design data Calculate the deviation amount, and set a correction reference point corresponding to each work point on a line connecting the measurement points in accordance with the positional relationship between each work point and the measurement point. wherein calculating each deviation in the correction reference point as to vary proportionally on the line, based on each deviation issued the calculated, estimated working direction data in the position data and the respective points of each working point The teaching data correction program according to claim 5, wherein Position and direction estimation steps, the working direction data in the position data and the respective points of each measurement point detected by the position and direction detection procedure, each of the position data and the working direction data is created based on the work of the design data Calculate the deviation amount, set a correction reference point corresponding to each work point on a line connecting the measurement points according to the ratio of the distance of each work point from the measurement point, and each deviation amount is Calculate the respective deviation amounts at the correction reference point as those that change proportionally on the line segment, and estimate the position data of each work point and the work direction data at each point based on the calculated deviation amounts. The teaching data correction program according to claim 5, wherein A recording medium storing the teaching data correction program according to claim 5, 6, 7, or 8. A teaching data correction device that is created based on work design data and that corrects teaching data for causing a robot to perform a predetermined operation,
Detecting means for detecting the positions of the two measurement points provided on the workpiece and the work direction at each point; and detecting the actual position of each work point based on the detected position data of each measurement point and the work direction data at each point. And a corrector for estimating the actual work direction at each point and the teaching data based on the estimated position data of each work point and the work direction data at each point. A teaching data correction device that is a feature. The correction means calculates each deviation amount of the detected position data of each measurement point and the work direction data at each point from the position data and work direction data created based on the design data of the work. Assuming that the position deviation amount and the work direction deviation amount at each point change according to the positional relationship with each measurement point, the position deviation amount of each work point and the work direction deviation amount at each point are estimated. The position data of each work point and the work direction data at each point are estimated based on the position deviation amount and the work direction deviation amount, and the teaching data is obtained based on the estimated position data of each work point and the work direction data at each point. 11. The teaching data correcting apparatus according to claim 10, wherein the teaching data is corrected. The correction means calculates each deviation amount of the detected position data of each measurement point and the work direction data at each point from the position data and work direction data created based on the design data of the work. A correction reference point corresponding to each work point is set on a line connecting the points in accordance with the positional relationship between each work point and the measurement point, and each deviation amount changes proportionally on the line segment. Assuming that each deviation amount at the correction reference point is calculated, the position data of each work point and the work direction data at each point are estimated based on the calculated deviation amounts, and the estimated work point The teaching data correction device according to claim 10, wherein the teaching data is corrected based on the position data and the work direction data at each point . The correction means calculates each deviation amount of the detected position data of each measurement point and the work direction data at each point from the position data and work direction data created based on the design data of the work. A correction reference point corresponding to each work point is set on a line connecting the points in accordance with a ratio of a distance of each work point from the measurement point, and each deviation amount changes proportionally on the line segment. Calculating each deviation amount at the correction reference point, estimating the position data of each work point and the work direction data at each point based on the calculated deviation amounts, and calculating the estimated work point The teaching data correcting apparatus according to claim 10, wherein the teaching data is corrected based on the position data and the work direction data at each point .

Images