WO2022254050A1

WO2022254050A1 - Method for estimating errors in the positioning of a rotary head in a precision machine, and precision machine

Info

Publication number: WO2022254050A1
Application number: PCT/ES2021/070401
Authority: WO
Inventors: Eneko Gomez-Acedo
Original assignee: Fundación Tekniker
Priority date: 2021-06-03
Filing date: 2021-06-03
Publication date: 2022-12-08

Abstract

A method for estimating errors in the positioning of a rotary head in a precision machine, comprising: bringing a reference sphere and a contact probe into contact with one another a plurality of times, the rotary head presenting different combinations of first and second angles of first and second head members; determining the position of the centre of the reference sphere for the plurality of combinations of the first and second angles of the rotary head in the contact step; recording the error values together with indicative data for each combination of the first and second angles of the rotary head when the reference sphere comes into contact with the contact probe; prior to and/or during the processing of a workpiece, the rotary head presenting a combination of a first target angle as the first angle and a second target angle as the second angle, and with the tool disposed on the head: recovering a plurality of recorded errors for which the combinations of the first and second angles are closest to the first and second target angles, respectively; and estimating the error values along the direction of each linear axis for the current combination of the first and second target angles, interpolating the error values of the plurality of digitally-recovered recorded errors that are the closest according to the respective combinations of the first and second angles.

Description

METHOD TO ESTIMATE ERRORS IN THE POSITIONING OF A

ROTATING HEAD OF A PRECISION MACHINE AND MACHINE

PRECISION

TECHNICAL FIELD

The present invention relates to the field of precision machines. More particularly, the invention relates to both methods and precision machines capable of estimating errors in the positioning of a rotary head of a precision machine. In some cases, precision machines and methods can also reduce such errors.

STATE OF THE ART

Precision machines, for example, machine tools, positioning machines, robots with serial or parallel kinematics, bin picking machines, or any other machine with any number of axes of rotation, are machines widely used to achieve a precise positioning of the tool or the end effector, which, for example, in the case of machine tools is essential for the correct manufacturing of the part within tolerances. In this sense, precision machines incorporate one or more tools or end effectors to perform different processing operations. Each selected tool must be arranged in the head of the precision machine when the operation is to be performed, for example, when a machine tool manufactures a workpiece.

Many precision machines incorporate a swivel head so that the tool can reach different parts of the workpiece and/or with different angles. To this end, the head incorporates a plurality of members, each of which can rotate with respect to the other members around an axis of rotation. Although precision machines have improved in recent decades, they still suffer from positioning errors when the spindle is rotated, thus processing workpieces incorrectly.

Differences between the expected position and the actual position of the tool negatively affect the resulting work pieces because the work pieces must be processed with high precision to be fully functional and/or meet certain quality requirements. This problem is generally exacerbated the higher the number of degrees of freedom existing in the precision machine and its head as a result of possible rotations.

A technique currently used to reduce the effect of this problem is the calibration of the rotary head at specific and discrete positions of rotation, that is, specific angle combinations, by means of a probe disposed in the head that comes into contact with a reference sphere. These specific angle combinations are most commonly used when processing a workpiece, so when these combinations are set in the spindle, the positioning error can be corrected by knowing the error during the calibration procedure. However, when uncalibrated angle combinations are established (for example, in 5-axis continuous manufacturing), no such error has been determined for them, and therefore the error cannot be corrected. This, in turn, results in the processing of a workpiece with the error of positioning of the tool presented by the swivel head for that combination of angles.

Higher accuracy in processing workpieces is necessary for most or all of the rotary head's working area because calibration of very specific angles is considered insufficient.

DESCRIPTION OF THE INVENTION

A first aspect of the invention refers to a method for estimating at least errors in the positioning of a rotary head of a precision machine, comprising: - equipping the rotary head precision machine, the rotary head comprising at least one first and second members, the first member being coupled to the precision machine such that the first member is rotatable about a first axis, and the second member being coupled to the first member such that the second member is rotatable about a second axis;

- arranging a contact probe on the second member (for example, on the tool holder);

- contacting a reference sphere with the contact probe a plurality of times, the rotary head presenting different combinations of the first and second angles, the first angle being the angle of the first member around the first axis, and the second angle being the angle of the second member about the second axis;

digitally determining the position of the center of the reference sphere for each combination of the first and second angles of the rotary head of the contact stage, that is, the center is determined for each combination of the rotating head when the contact probe has contacted the reference sphere in the contact stage;

- digitally recording the error values, each error value being a difference between a determined position and a predetermined reference position for the center of the reference sphere along the direction of one of the three linear axes, all three being recorded error values together with data indicative of each combination of the first and second angles of the rotary head when the reference sphere is contacted with the contact probe; - removing the contact probe from the second member and arranging a tool on the second member; Y

- before and/or during processing of a workpiece with the rotary head presenting a combination of a first target angle, as the first angle, and a second target angle, as the second angle, and the tool being arranged therein, said otherwise, when a workpiece is processed or to be processed by the tool and the rotary head exhibits the target first and second angle combination: digitally retrieve a plurality of recorded errors whose first and second angle combinations are closest to each other to the first and second target angles, respectively, each digitally recovered recorded error comprising its respective three error values; and digitally estimating the error values along the direction of each linear axis for the combination of the first and second target angles by interpolating the error values of the plurality of digitally retrieved recorded errors as a function of the respective combinations of the first and second angles.

The present method makes it possible to estimate the positioning errors of the rotary head and therefore of the tool arranged therein, thus making it possible to determine if the processing of the workpiece will not be correct. The errors can be estimated for combinations that have not been calibrated, in other words, for combinations of angles other than those used when contacting the reference sphere (which is used as a calibration sphere) with the contact probe, the last combinations being the calibrated ones Said estimation can be carried out by means of a numerical control device of the precision machine, for example, a numerical control, a programmable logic controller, etc. The precision machine can be a machine tool, a positioning machine, a robot with serial or parallel kinematics, a machine for extracting parts from containers, or any other precision machine arranged with any number of axes of rotation.

After contacting the reference sphere with the contact probe while the rotary head has a combination of first and second angles on rotary heads with two rotary axes, or the number of axes required on any other head with any other number of rotary axes , the position of the center of the sphere is determined digitally based on this measurement, then an additional contact of the reference sphere is made with the contact probe while the rotary head presents another combination of the first and second angles, and it is determined again the position of the center of the sphere digitally based on said measurement. The process is repeated a plurality of times to calibrate a plurality of combinations of the first and second angles. Each given position will be offset from the expected position (i.e., the default reference position for the center of the reference sphere) due to imperfect rotation of the members, causing an offset in the position of one or both members first and second, thus also displacing the tool from its expected position. The expected position is the position where the center of the reference sphere would have to be if the member or members had not been displaced; preferably, the predetermined reference position is the determined position of the center of the reference sphere when the angular positioning error of both the first and the second angle is zero, since in that case there is little or no movement of the head. By calculating the difference between the determined and expected position, the error for the actual combination of first and second angles used when contacting the reference sphere is determined. Each error comprises three error values, one for each of the three linear axes, preferably of a global Cartesian coordinate system; In other words, an error value for the X axis, an error value for the Y axis, as well as an error value for the Z axis. By recording the different error values, for example, in a memory of the device control, the precision machine can subsequently retrieve the set of three error values for any combination of the first and second angles used in contacting the reference sphere during calibration.

In the same way, the precision machine can recover those errors to estimate the three error values for any other combination than the first and second angles, for example, a combination of the first and second target angles. For this purpose, a number of errors (each with the three respective error values) is retrieved, eg two, three, four, five or even more errors. Preferably, these errors are provided for combinations of the first and second angles that most closely approximate the combination of the first and second target angles, namely, the angular distance between the combination of the first and second angles is as short as possible with respect to the combination of the first and second target angles. The recovery from certain errors is such that the interpolation of the error values to be estimated can be as accurate as possible.

In some embodiments, the digital retrieval and digital estimation steps are performed after selecting the combination of the first and second target angles. In some of these embodiments, the digital retrieval and digital estimation steps are performed during the tooling step of processing the workpiece.

The three error values can be estimated during the use of the precision machine and continuously. In other words, the error values can be estimated each time a new combination of the first and second target angles is selected, therefore the process can be in real time or near real time so that corrective action can be taken. during the operation of the precision machine, and even during the processing of the workpiece.

In some embodiments, the method further comprises moving the precision machine to compensate for the three digitally estimated error values; for this purpose, one, two or all three linear axes can be moved precision machine movables to compensate for the error. In some of these embodiments, the moving step takes place before the workpiece processing step. In some of these embodiments, the moving step takes place during the workpiece processing step. Based on the estimated error values, corrective measures are taken by moving the precision machine along its three linear axes (X, Y, Z), so as to compensate for the error as estimated. Consequently, the precision machine moves along each linear axis by the estimated amount, but with the opposite sign, that is, in the opposite direction. This corrective action is possible because the error values are estimated and can be provided in real time.

In some embodiments, the method further comprises converting digitally estimated error values from a global Cartesian coordinate system to a precision machine Cartesian coordinate system; and the movement of the precision machine is based on the error values of the global coordinate system converted to the local coordinate system of the precision machine.

Preferably, the digitally estimated error values can be recorded in the global Cartesian coordinate system so that they can be more easily used regardless of the tool used on the precision machine. Each precision machine tool can result in a Cartesian machine coordinate system that is different from the Cartesian machine coordinate system when another tool is used due to how the tools are operated. When digitally retrieving errors, their error values are converted to the Cartesian coordinate system of the machine and corrective action is taken in the form of moving the precision machine in accordance with the converted error values.

In some embodiments, the plurality of logged errors that is digitally retrieved are four logged errors (each of the four logged errors comprising the three error values) whose combinations of the first and second angles most closely approximate the first and second target angles, respectively. .

In some embodiments, each of the four recorded errors has a combination of the first and second angles such that the first and second target angles are within the values of the four recorded combinations and most closely approximate the first and second angles, each one with its respective error (each of the four recorded errors comprising the three error values). Specifically: A1 > TA, B1 > TB; A2 > TA, B2 < TB; A3 < TA, B3 > TB; and A4 < TA, B4 < TB. Where: TA and TB are the first and second objective angles, respectively; A1 and B1 are the first and second angles of the digitally recovered first error combination, respectively; A2 and B2 are the first and second angles of the digitally recovered second error combination, respectively; A3 and B3 are the first and second angles of the digitally recovered third error combination, respectively; and A4 and B4 are the first and second angles of the digitally recovered fourth error combination, respectively.

Therefore, each pair of angles An and Bn of a digitally recovered error (where n is the nth error recovered) is such that it meets the criterion respective previous one and is the one that most closely approximates the pair of angles TA and TB. In this way, the interpolation used for the digital estimate has the three different four-sided error values, so that the estimate strikes a compromise between simplicity (because interpolation with four points arranged in this way is computationally simple) and precision. .

In some embodiments, the interpolation to digitally estimate the error value along the direction of each linear axis is or comprises bilinear interpolation.

In some embodiments, the combination of the first and second target angles (from the stage before and/or during the processing of a workpiece) is not any combination used in the contact stage, i.e. it is not any combination of the rotary head. when the reference sphere is contacted with the contact probe.

In some embodiments, the method further comprises selecting the combination of the first and second target angles (from the stage before and/or during processing of a workpiece).

In some embodiments, the method further comprises the step of processing the workpiece (the rotary head presenting the combination of the first target angle, as the first angle, and the second target angle, as the second angle, and the tool arranged therein). ).

In some embodiments, the tool is a milling machine or a drilling machine.

A second aspect of the invention relates to a precision machine (for example, a machine tool, a positioning machine, a robot with serial or parallel kinematics, a part extraction machine of containers, or any other precision machine arranged with any number of axes of rotation) comprising:

- a rotary head comprising at least a first and a second member, the first member being coupled to the precision machine such that the first member rotates about a first axis, and the second member being coupled to the first member such that the second member can rotate about a second axis;

- a contact probe adapted for releasable engagement with the second member; - a tool adapted for separable engagement with the second member; Y

- a control device, for example a numerical control device, configured to at least: change the first and second angles of the rotary head, the first angle being the angle of the first member around the first axis, the second angle being the angle of the second member around the second axis; moving the precision machine when the contact probe is attached to the second member to contact a reference sphere a plurality of times, the rotary head presenting different combinations of the first and second angles; determining a position of a center of the reference sphere for each combination of the first and second angles of the rotary head when the reference sphere is contacted with the contact probe; recording error values, each error value being a difference between a given position and a predetermined reference position for the center of the reference sphere, the error values being recorded together with data indicative of each combination of the first and second angles of the rotary head when the reference sphere is contacted with the contact probe; processing a workpiece by moving the precision machine both when the tool is attached to the second member and when the rotary head exhibits a combination of a first target angle, as the first angle, and a second target angle, as the second angle; retrieving a plurality of logged errors whose combinations of the first and second angles most closely approximate the first and second target angles, respectively, in processing the workpiece, each retrieved logged error comprising its respective three error values; and estimating the error values along each linear axis for the combination of the first and second target angles by interpolating the error values of the plurality of recorded errors retrieved based on the respective combinations of the first and second angles when processing the workpiece. worked.

The precision machine is capable of estimating the error present in the positioning of the rotating head and, therefore, in the positioning of a tool arranged in it, depending on the use of the contact probe in the reference sphere and the device of control. The control device, which can be, for example, a programmable logic controller, determines the error present in the positioning when it commands the movement of the precision machine (that is, it commands the movement of one, two or all three axes). moving lines of the same) so that the contact probe enters into contact with the reference sphere for various angular configurations of the rotating head.

Based on the error values provided in this way, the control device is then able to estimate the error for other angular configurations of the rotary head even if they have not been selected when the contact probe measures the reference sphere. The error estimate for each such angular configuration is achieved by interpolation of the error values provided during the contact phase. In some embodiments, the precision machine further comprises the reference sphere.

In some embodiments, the control device is further configured to adjust the movement of the precision machine before and/or during processing of the workpiece to compensate for the estimated error value.

The control device commands the precision machine to move on one, two, or all three of its movable linear axes to compensate for the error.

In some embodiments, the control device retrieves the plurality of recorded errors and estimates the error values before and/or during processing of the workpiece.

In some embodiments, the control device is further configured to convert the estimated error values from a global Cartesian coordinate system to a precision machine Cartesian coordinate system; and the adjustment of the movement of the precision machine is based on the error values converted to the Cartesian coordinate system of the precision machine.

In some embodiments, the plurality of logged errors that are recovered is four logged errors. In some of these embodiments, each of the four recorded errors has a combination of the first and second angles, such that the combination is in a different quadrant of four quadrants surrounding the combination of the first and second target angles.

In some embodiments, the interpolation to estimate the error value along the direction of each linear axis is or comprises bilinear interpolation.

In some embodiments, the control device is configured to perform error estimation automatically. Furthermore, in some of these embodiments, the control device is configured to automatically compensate for the error. In some embodiments, the tool is a milling machine or a drilling machine.

A third aspect of the invention relates to a computer program product having instructions that, when executed by a control device of a precision machine (for example, a machine tool, a positioning machine, a robot with kinematics in series or parallel, etc.), comprising a rotary head with a first member coupled to the precision machine such that the first member can rotate about a first axis, and a second member coupled to the first member such that the second member can rotate about a second axis, cause the control device to command: - contacting a reference sphere with a contact probe arranged in the rotary head a plurality of times, the rotary head presenting different combinations of the first and second angles, the first angle being the angle of the first member around the first axis, the second being angle the angle of the second member around the second axis;

- determining the position of the center of the reference sphere for each combination of the first and second angles of the rotary head when the reference sphere is contacted with the contact probe;

- recording error values, each error value being a difference between a determined position and a predetermined reference position for the center of the reference sphere along one of the three linear axes, the three error values being recorded together with data indicative of each combination of the first and second angles of the rotary head when the reference sphere is contacted with the contact probe; - retrieving a plurality of recorded errors whose combinations of the first and second angles are closest to the first and second target angles, respectively, of a combination used or to be used by the rotary head to process a workpiece, each error comprising registered digitally retrieved their respective three error values; Y

- estimating the error values along the direction of each linear axis for the combination of the first and second target angles by interpolating the error values of the plurality of retrieved recorded errors based on the respective combinations of the first and second angles. When the computer program product is executed on one or more processors of the computing device, the precision machine estimates the positioning errors of the rotary head that exist due to the rotations thereof, and preferably also moves the precision machine itself based on the estimated error values to correct them. In some embodiments, the instructions cause the control device to command recovery and estimation after commanding the selection of the combination of the first and second target angles. In some of these embodiments, the instructions cause the control device to command retrieval and estimation before and/or when commanding processing of a workpiece with the precision machine while a tool is being arranged in the rotary head.

In some embodiments, the instructions further cause the control device to command the precision machine to move to compensate for the estimated error values. In some embodiments, the instructions further cause the control device to command to convert the estimated error values from a global Cartesian coordinate system to a precision machine Cartesian coordinate system; and the commanded movement of the precision machine is based on the error values converted to the Cartesian coordinate system of the precision machine.

In some embodiments, the combination of the first and second target angles (used or to be used by the rotary head for processing a work piece) is not any combination used when contacting the reference sphere with the contact probe.

In some embodiments, the instructions further cause the control device to command to select the combination of the first and second target angles. In some of these embodiments, the first and second target angle pair of the selected combination does not match the first and second angle pair of any combination used when contacting the reference sphere with the contact probe.

In some embodiments, each of the four recorded errors exhibits a combination of the first and second angles, such that the combination is in a different quadrant of four quadrants surrounding the combination of the first and second target angles. In some embodiments, the computer program product is embedded on a non-transient computer-readable medium. A fourth aspect of the invention relates to a data stream that is representative of a computer program product according to the third aspect of the invention.

The same advantages as those described for the first aspect of the invention can also be applied to the second, third and/or fourth aspects of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

To complete the description and in order to provide a better understanding of the invention, a set of drawings is provided. Said drawings form an integral part of the description and illustrate embodiments of the invention, which are not to be construed as a restriction of the scope of the invention, but only as examples of how the invention can be carried out. The drawings comprise the following figures:

FIG. 1 schematically shows a cross section of a rotary head of a precision machine, in which both axes of rotation and a positioning error have been represented.

Fig. 2 schematically shows a cross section of a rotary head of a precision machine with a contact probe arranged therein to contact a reference sphere according to the embodiments.

Figure 3 schematically shows the error values recorded according to the embodiments after contacting a reference sphere, as illustrated in Figure 2.

Fig. 4 illustrates the estimation of the values of an error according to the embodiments.

Figures 5A to 5D illustrate the calculation of a bilinear interpolation in accordance with the embodiments.

Figures 6 and 7 schematically show the coordinate systems of a precision machine when some tools are respectively arranged therein.

DESCRIPTION OF THE FORM OF EMBODIMENT OF THE INVENTION Figure 1 schematically shows a cross section of a rotary head 20 of a precision machine 10, such as a machine tool.

The rotary head 20 comprises a first member 22, which is coupled to the precision machine 10, and a second member 24 coupled to the first member 22. The precision machine 10 comprises a control device 100, for example a numerical control, which operates the rotary head 20.

The first member 22 is coupled to the machine 10 such that the first member 22 is rotatable with respect to the machine 10 about a first axis 31 defined between both the first member 22 and the precision machine 10 (shown with dashed lines). for illustrative purposes only). The possible rotation of the first member 22 is illustrated by means of the arrow 30. When the first member 22 rotates, a first angle indicative of the angular displacement of the first member 22 with respect to the first axis 31 is formed; the first angle corresponds to a rotation around the Z' axis (numbered 31 in the drawing) contained in the drawing plane (in this case, the axis is the intersection between the XZ and ZY planes). Said first angle, indicated with the letter A in the description of the drawings, is indicative of how much the first member 22 rotates with respect to an initial angular position thereof, said initial angular position corresponding to an angle of 0°, that is, A = 0°.

The second member 24 is coupled to the first member 22 such that the second member 24 is rotatable relative to the first member 22 about a second axis 41 which forms 45° in the Y'Z' plane between the second member 24 and the first. member 22 (shown with dashed lines for illustrative purposes only). The possible rotation of the second member 24 is illustrated by arrow 40. A tool 50 (for example, a milling machine, a drilling machine, or a different tool used in a precision machine as known in the art) of the precision machine 10 can be arranged at one end of the second member 24 so that it can process a workpiece 55. When the second member 24 rotates, a second angle indicative of the angular displacement of the second member 24 with respect to the second axis 41 is formed. Said second angle, indicated with the letter B in the description of the drawings, it is indicative of how much the second member 24 rotates with respect to an initial angular position thereof, said initial angular position corresponding to an angle of 0°, that is, B = 0°.

When the first member 22 rotates with respect to the precision machine 10 about the first axis 31, the point of intersection 46 between the axes 31 and 41 is displaced due to misalignment of the rotary head 20. Consequently, the point 46 of the axis is moved such that it results in point 47 of the moved axis. This, in turn, causes the displacement of the first axis 31, thus producing the first offset axis 32. The displacement of the point 46 of the axis, and therefore of the first axis 31, results in a positioning error of the rotary head 20 and, consequently, of the tool 50 arranged therein, something that affects how the tool 50 processes the workpiece 55.

Additionally or alternatively, when the second member 24 rotates with respect to the first member 22 about the second axis 41, the point of intersection 46 between the axes 31 and 41 can be displaced due to misalignment of the rotary head 20. Consequently, the point 46 of the axis can be moved. This, in turn, can also cause displacement of the second axis 41, thus producing the displaced second axis 42. This additional or reciprocating displacement results in a positioning error (or additional positioning error) of the rotary head 20 and consequently of the tool 50 arranged therein, thus affecting how the tool 50 processes the workpiece 55.

A global Cartesian coordinate system with X, Y, and Z axes has been depicted for clarity. In this sense, both the rotary head 20 and the precision machine 10 can have their own Cartesian coordinate system with axes X', Y' and Z'. Figure 2 schematically shows a cross section of a rotary head 20 of a precision machine 10.

The rotary head 20 has a contact probe 51 arranged therein. The control device 100 commands the movable linear axes of the precision machine 10 to move so that the probe 51 comes into contact with a reference sphere 60, which may be part of the precision machine 10, for the purposes of calibration. Each movable linear axis moves along one of the X, Y and Z' axes.

For the calibration to be effective, the contact probe 51 must come into contact multiple times with the reference sphere 60, presenting the rotating head 20 different combinations of the first and second angles A and B. In this sense, each time it is configured a combination of the first and second angles A and B on the rotary head 20, the contact probe 51 contacts the reference sphere 60 one or more times (depending on whether the contact probe 51 makes a continuous measurement on the surface of the sphere 60 or measured several times on different parts of the surface of the sphere 60) and based on the measurement of the contact probe 51, a control device 100 will numerically determine the position of the center of the reference sphere 60 for said combination of the first and second angles A and B of the rotary head 20, as shown know in the art; then the same procedure is repeated one or more times with a different combination of the first and second angles A and B, and the position of the center of the reference sphere 60 will be determined numerically according to the respective measurement.

As can be seen from the explanation with reference to figure 1, there is a different displacement for each pair of the first and second angles.

A and B due to misalignments of the head 20. Based on the measurements of the contact probe 51 upon contact with the reference sphere 60, the center of the reference sphere 60 can be determined digitally (with the control device 100) each time, as is known in the art. An error value along the direction of each linear axis can then be computed for each of these determined centers by calculating a difference between each given determined center and a predetermined reference position for the center of the reference sphere 60 The predetermined reference position is preferably the center of the reference sphere 60 determined after contacting the reference sphere 60 with the probe 51 and the first and second angles A and B are both, for example, 0°. Therefore, all error values refer to said predetermined reference position and are calculated as:

Where: εp_x, ep_y, and ep_z are the error values for the X, Y, and Z coordinates in a Cartesian coordinate system, preferably a global Cartesian coordinate system, with the first angle A equal to angle a (i.e., alpha) and the second angle B equal to b (beta); P _center.x, P _center.y, and P _center.z are the X, Y, and Z coordinates (in the same Cartesian coordinate system) of the determined positions of the center of the reference sphere 60 for the first and second angles indicated A and B. In this case, the default reference position is the determined position of the center when the first and second angles A and B are both 0°, but other combinations of the first and second angles are also possible, including combinations in where A and B do not have the same value. The control device 100 can digitally record the three coordinates for the same error, together in the form of a 3D vector, or separately, for example, as the example illustrated in Figure 3; Regardless of how the errors are digitally recorded, the errors are always recorded with reference to the values of the first and second angles A and B and include the three respective error values, one along each of the three axes.

Based on the digitally recorded errors, the control device 100 can estimate the error for other combinations of the first and second angles A and B, that is, pairs of the first and second angles A and B not established in the rotary head 20 during calibration. . To this end, the control device 100 interpolates the three error values for each of said other combinations from the error values of the calibrated combinations of the first and second angles A and B. Preferably, control device 100 moves one or more of the three movable linear axes of precision machine 10 based on interpolated error values to compensate for them when processing a workpiece. Figure 3 schematically shows the error values εp_x, ep_y, ep_z recorded according to the embodiments.

The error values εp_x, ep_y, and ep_z are recorded by the control device (for example, the control device 100 of Figures 1 and 2) in three different tables or graphs 70, 71, 72, each table or graph corresponding graph to a different Cartesian coordinate, that is, the X, Y, and Z coordinates. For example, the first table or graph 70 corresponds to the X coordinate, the second table or graph 71 corresponds to the Y coordinate, and the third table or Graph 72 corresponds to the Z coordinate. For the sake of better illustration only, the error values εp_x, ep_y, and ep_sze have been represented by a square in the first table or graph 70, a triangle in the second table or graph 71, and a circle in the third table or graph 72, but it is clear that each of these error values εp_x, ep_y, and ep_z is a numerical value that can be positive or negative.

Each table or graph 70, 71, 72 comprises the same number of error values, which in this example have been calculated following a calibration procedure for all pairs in which the first angle A was 0°, 30°, 60 °, 90°, -30°, -60° and -90°, and the second angle B was 0°, 30°, 60°, 90°, -30°, -60° and -90°, thus totaling 49 different combinations, whereby the combination of A = 0° and B = 0° is, for example, for use in determining a predetermined reference position for the center of a sphere of reference. In other embodiments, other values have been calibrated for the first and second angles A and B.

Figure 4 illustrates how error values of angular positions other than those used during initial calibration can be estimated according to the embodiments.

An A-B graph 80 has been represented in which it is illustratively shown that the three error values must be estimated for the 95 er combination. Combination 95 corresponds to a combination er (which can be εp_x, ep_y, and ep_z of a first target angle A and a second target angle B to which the rotary head of a precision machine must be set or has been set to process a workpiece with a tool.In this case, the first target angle A is Aii and the second target angle B is Bjj.

In this example, the control device of a precision machine has digitally retrieved the four combinations 91-94 of the first and second angles A and B that most closely approximate the combination er for which their three error values have been digitally recorded. previously. The four combinations 91-94 are called ε(i,j), ε(i,j+1), ε(i+1,j) and ε(i+1,j+1), each with a prime angle A equal to either A(i) or A(i+1), and each with a second angle B equal to either B(j) or B(j+1). As in this example, it is preferred that each of the four retrieved combinations 91-94 be in a different one than the four quadrants surrounding combination er.

The control device interpolates the three error values corresponding to each of the four digitally recovered combinations, such that the error values along the same linear axis they are interpolated independently of the error values along the other two linear axes. Preferably, the interpolation is performed taking into account the angular distance of the combination with the error er to each of the four digitally recovered combinations, thus, in this example, the error values of the combination 91 will have the greatest contribution in the estimate of the error value of combination er, then the error value of combination 93, then the error value of combination 92, and finally the error value of combination 94 will be the one with the lower contribution due to the greater distance to the combination er. Each of the three error coordinates, ie, the three error values, is interpolated in this manner to provide the three error value coordinates for the combination er.

In other embodiments, fewer combinations (eg, two, three) and their three error values are digitally retrieved for interpolation, or more combinations (eg, five, six, etc.) and their three error values are digitally retrieved. for interpolation.

Figures 5A to 5D illustrate calculation estimation of a bilinear interpolation according to the embodiments.

Bilinear interpolation must be calculated to estimate the error value of a combination of the first and second target angles as angles A and B, of a rotary head. The interpolation should be based on four combinations 91-94 that have been calibrated, i.e. the respective combinations of angles A and B that were set on the rotary head when a contact probe was brought into contact with a reference sphere for Determine numerically the center of the sphere. As shown in Figure 5A, the calculation of the bilinear interpolation can be performed by first calculating e1 , then ε2 and finally er for each of the three error values; in this sense, for calculations, triangles can be defined at the top, at the bottom and to the right. The same triangles are shown in Figures 5B, 5C and 5D for clarity only.

The following calculations should be performed for each of the three error values of the combinations 91-95 according to the illustrations in Figures 5A-5D.

First, e1 must be calculated (see Figures 5A and 5B):

Then, e2 must be calculated (see figures 5A and 5C):

Then, the partial calculation of er is performed (see figures 5A and 5D):

This means that the X, Y, and Z coordinates of the error for combination 95 (of the first and second target angles) are respectively calculated as:

Fig. 6 schematically shows the coordinate system of a precision machine 10 when some tools are arranged on the second member, as the tool thereof.

In some machines, the XΎ'Z' coordinate system of the precision machine 10 is such that the tool coincides with the -Y' axis (ie, negative Y). However, for the combination of A = 180° and B = 0°, both the coordinate system CΎ'Z' of the machine 10 and the global coordinate system XYZ coincide.

The transformations of the error corrections may have to be done depending on the configuration of the 10 precision machine and/or the configuration of the tool, for example, depending on the models of the tools used, which have different hardware configurations/ software. By way of example only, the transformations described with reference to this figure and to figure 7 can be applied to error corrections when the precision machine 10 is using a milling configuration or a drilling configuration.

In these cases, to transform an error correction vector V from the global coordinate system to the machine coordinate system 10, the necessary transformation to be applied is: V' = T * V; where T is:

Furthermore, to transform the error correction vector V from the machine coordinate system 10 to the global coordinate system, the necessary transformation to be applied is: V = T * V'; where T is:

In the two previous transformations and in the transformations described in relation to figure 7, cA = cos(A), sA = sin(A), cB = cos(B), sB = sin(B) and K = cos(45°).

Fig. 7 schematically shows the coordinate system of a precision machine 10 when some tools are arranged on the second member, as the tool thereof. On some other machines, the X'Y'Z coordinate system of the precision machine 10 is such that the tool coincides with the -Z axis (ie, negative Z). However, for the combination of A=0° and B=0°, both the X'Y'Z coordinate system of the machine 10 and the global XYZ coordinate system coincide.

In these cases, to transform an error correction vector V from the global coordinate system to the machine coordinate system 10, the necessary transformation to be applied is: V' = T ^* V; where T is:

Furthermore, to transform the error correction vector V from the machine coordinate system 10 to the global coordinate system, the necessary transformation to be applied is: V = T ^* V'; where T is:

In this text, the terms first, second, third, etc. have been used herein to describe various devices, items, or parameters, it is understood that devices, items, or parameters should not be limited by these terms, as the terms are only used to distinguish one device, element or parameter from another. For example, the first member could also be called the second member, and the second member could be called the first member without departing from the scope of this disclosure.

In this text, the term "comprising" and its derivatives (such as "comprising", etc.) should not be understood in an exclusive sense, that is, these terms should not be interpreted as excluding the possibility that what is described and defines could include additional elements or stages, etc.

On the other hand, the present invention is obviously not limited to the specific embodiments described herein, but also covers any variations that can be considered by any person skilled in the art (for example, regarding the choice of materials, dimensions, components, configuration, etc.), within the general scope of the invention, as defined in the claims.

Claims

1. A method for estimating errors in the positioning of a rotary head (20) of a precision machine (10), comprising:

- equipping the precision machine (10) with the rotating head (20), the rotating head (20) comprising at least a first and a second member (22, 24), the first member (22) being coupled to the machine precision (10) so that the first member can rotate (30) about a first axis (31), and the second member (24) being coupled to the first member (22) so that the second member (24) can rotate (40) about a second axis (41);

- arranging a contact probe (51) in the second member (24);

- contacting a reference sphere (60) with the contact probe (51) a plurality of times, the rotary head (20) presenting different combinations (91-94) of the first and second angles (A, B), the first being angle (A) being the angle of the first member (22) around the first axis (31), and the second angle (B) being the angle of the second member (24) around the second axis (41);

- digitally determining a position of a center of the reference sphere (60) for each combination (91-94) of the first and second angles (A, B) of the rotating head (20) of the contact stage;

- digitally recording error values ( εp_x, ep_y, εp_z), each error value ( εp_x, ep_y, εp_z) being a difference between a determined position and a predetermined reference position for the center of the reference sphere (60) a along the direction of one of the three linear axes (X, Y, Z), recording the three error values together with data indicative of each combination (91-94) of the first and second angles (A, B) of the rotating head (20) when the reference sphere (60) is contacted with the probe. contact (51); - removing the contact probe (51) from the second member (24) and arranging a tool (50) in the second member (24); and - before and/or during processing of a workpiece (55), the rotary head (20) presenting a combination (95) of a first target angle, as first angle (A), and a second target angle, as second angle (B), and the tool (50) being arranged therein: digitally retrieving a plurality of recorded errors whose combinations (91-94) of the first and second angles (A, B) are closest to the first and second angles target, respectively, each recorded error comprising its three respective error values (er_c, sp_y, er_z); and digitally estimate the error values ( εp_x, ep_y, εp_z) along the direction of each linear axis (X, Y, Z) for the combination (95) of the first and second target angles by interpolating the error values (er_c , sp_y, er_z) of the plurality of recorded errors digitally recovered as a function of the respective combinations (91-95) of the first and second angles (A, B).

The method of claim 1, further comprising moving the precision machine (10) to compensate for digitally estimated error values (εp_x, ep_y, εp_z).

The method of claim 2, further comprising converting the digitally estimated error values ( εp_x, ep_y, εp_z ) from global Cartesian coordinates into precision machine (10) Cartesian coordinates; and where the movement of the precision machine (10) is based on the error values (εp_x, ep_y, εp_z) converted into the Cartesian coordinates of the precision machine (10).

The method of any one of the preceding claims, wherein the digital retrieval and digital estimation steps are performed during the step of processing the workpiece (55) with the tool (50).

The method of any one of the preceding claims, wherein each error value ( εp_x, ep_y, εp_z) corresponds to a Cartesian coordinate different from three Cartesian coordinates (X, Y, Z).

The method of any one of the preceding claims, wherein the plurality of logged errors that are digitally recovered are four logged errors; each of the four recorded errors presents a combination (91-94) of a first and a second angle (A,B) such that the combination is in a different quadrant of four quadrants surrounding the combination (95) of the first and second target angles.

The method of any one of the preceding claims, wherein the interpolation to digitally estimate the error values (εp_x, ep_y, εp_z) along the direction of each linear axis (X, Y, Z) is or comprises a bilinear interpolation.

8. A precision machine (10) comprising: - a rotating head (20) comprising at least a first and a second member (22, 24), the first member (22) being coupled to the precision machine (10 ) so that the first member (22) can rotate (30) around a first axis (31), and the second member (24) being coupled to the first member (22) so that the second member (24) can rotate (40) around a second axis (41);

- a contact probe (51) adapted for detachable engagement with the second member (24);

- a tool (50) adapted for releasable engagement with the second member (24); and - a control device (100) configured at least for:

^■ changing the first and second angles (A, B) of the rotating head (20), the first angle (A) being the angle of the first member (22) around the first axis (31), the second angle (B) being the angle of the second member (24) about the second axis (41); ■ move the precision machine (10) when the contact probe (51) is attached to the second member (24) to contact a reference sphere (60) a plurality of times, presented with the rotary head (20) different combinations (91-94) of the first and second angles (A, B); ■ determine a position of the center of the reference sphere (60) for each combination (91-94) of the first and second angles (A, B) of the rotating head (20) when the reference sphere (60) is contacted with the contact probe (51);

^■ record error values (εp_x, ep_y, εp_z ), each error value (εp_x, ep_y, εp_z ) being a difference between a given position and a predetermined reference position for the center of the reference sphere (60) along along the direction of one of the three linear axes (X, Y, Z), recording the three error values together with data indicative of each combination (91-94) of the first and second angles (A, B) of the rotary head (20) when the reference sphere (60) is contacted with the contact probe (51);

^■ processing a workpiece (55) by moving the precision machine (10) both when the tool (50) is coupled to the second member (24) and when the rotary head (20) presents a combination (95) of a first angle target, as the first angle (A), and a second target angle, as the second angle (B);

^■ retrieve a plurality of recorded errors whose combinations (91-94) of the first and second angles (A,B) most closely approximate the first and second target angles, respectively, when the workpiece (55) is processed, including each error logged your three error values

(εp_x, ep_y, εp_z) respective; Y

^■ estimate the error values (εp_x, ep_y, εp_z) along the direction of each linear axis (X, Y, Z) for the combination (95) of the first and second target angles by interpolating the error values (εp_x, ep_y, εp_z) of the plurality of recorded errors recovered based on the respective combinations (91-95) of the first and second angles (A, B) when processing the workpiece (55).

The precision machine (10) of claim 8, wherein the control device (100) is further configured to adjust the movement of the precision machine (10) before and/or during processing of the workpiece. (55) to compensate for the estimated error values (er_c, sp_y, er_z).

The precision machine (10) of claim 9, wherein the control device (100) is further configured to convert the error values (εp_x, ep_y, εp_z) estimated from global Cartesian coordinates into Cartesian coordinates. precision machine (10); and wherein the adjustment of the movement of the precision machine (10) is based on the error values (εp_x, ep_y, εp_z) converted into the Cartesian coordinates of the precision machine (10).

11. The precision machine (10) of any one of claims 8-

10, wherein the control device (100) retrieves the plurality of recorded errors and estimates the error values (εp_x, ep_y, εp_z) before and/or during processing of the workpiece (55).

12. The precision machine (10) of any one of claims 8-

11, wherein each error value comprises three values (er_c, sp_y, er_z), each of the three values corresponding to a Cartesian coordinate different from three Cartesian coordinates (X, Y, Z).

The precision machine (10) of any one of claims 8-12, wherein the plurality of logged errors that are recovered are four logged errors; each of the four recorded errors presents a combination (91-94) of a first and a second angle (A,B) such that the combination is in a different quadrant of four quadrants surrounding the combination (95) of the first and second target angles.

The precision machine (10) of any one of claims 8-13, wherein the interpolation to estimate the error values (εp_x, ep_y, εp_z) along the direction of each linear axis (X, Y , Z) is or comprises a bilinear interpolation.

15. A computer program product having instructions that, when executed by a control device (100) of a precision machine

(10) comprising a rotary head (20) with a first member (22) coupled to the precision machine (10) such that the first member (22) can rotate (30) around a first axis (31), and a second member (24) coupled to the first member (22) so that the second member (24) can rotate (40) about a second axis (41), cause the control device (100) to command that:

- a reference sphere (60) is contacted with a contact probe (51) arranged on the rotary head (20) a plurality of times, the rotary head (20) presenting different combinations (91-94) of the first and second angles (A, B), the first angle (A) being the angle of the first member (22) about the first axis (31), the second angle (B) being the angle of the second member (24) about the second axis (41);

- a position of a center of the reference sphere (60) is determined for each combination (91-94) of the first and second angles (A, B) of the rotary head (20) when the reference sphere (60) is contacted with the contact probe (51);

- the error values ( εp_x, ep_y, εp_z ) are recorded, each error value (εp_x, ep_y, εp_z ) being a difference between a given position and a predetermined reference position for the center of the reference sphere (60) along the direction of one of the three linear axes (X, Y, Z), recording the three error values together with data indicative of each combination (91-94) of the first and second angles (A, B) of the rotating head (20) when the reference sphere (60) is contacted with the contact probe (51); - a plurality of registered errors is recovered whose combinations (91-

94) of the first and second angles (A, B) are closest to the first and second target angles, respectively, of a combination (95) used or to be used by the rotary head (20); Y

- the error values (εp_x, ep_y, εp_z) are estimated along the direction of each linear axis (X, Y, Z) for the combination (95) of the first and second target angles by interpolating the error values (εp_x , ep_y, εp_z ) of the plurality of recorded errors recovered as a function of the respective combinations (91-95) of the first and second angles (A, B).