KR102591942B1 - A method and apparatus for controlling a robot using a model for stiffness and a model for cutting force - Google Patents

Abstract

One embodiment of the present invention provides a technology for controlling a robot so that error correction of a machining path for a machining object can be easily performed in a machining process using a robot. A robot control method using a rigidity model and a cutting force model according to an embodiment of the present invention performs processing on a processing object by combining a robot driven with a plurality of links and a plurality of joints, the end of the robot, and equipped with a tool. A first step of providing a processing unit that performs processing and a control unit that controls the posture of the robot when the tool works on the processing object; By using stiffness data, which is the stiffness value for each of the plurality of joints that varies depending on the posture of the robot, and information about cutting force, which is the force applied to the tool, the compliance error, which is the machining error calculated for the tool, is calculated and derived. Step 2; The reference position and reference direction, which are the position and direction of the tool, on the reference machining path, which is the machining path of the tool set in advance according to the machining target, and the actual position and direction, which are the real-time position and direction of the tool, using the compliance error. A third step of measuring the actual direction; A fourth step of dividing the reference machining path, which is a preset tool machining path according to the machining target, into a plurality of straight lines; The contour reference position is a position where the position error of the tool is corrected using two straight lines passing through the nearest reference position, which is the reference position closest to the actual position among a plurality of straight lines, and two vertical connecting lines that vertically connect the actual position. The fifth step of deriving; A sixth step of deriving a contour reference direction, which is a direction in which the direction error of the tool is corrected, using the reference direction and the actual direction at the nearest reference position; and a seventh step of controlling the position and direction of the tool by controlling the robot using the outline reference position and outline reference direction.

Description

Robot control method and device using stiffness model and cutting force model {A METHOD AND APPARATUS FOR CONTROLLING A ROBOT USING A MODEL FOR STIFFNESS AND A MODEL FOR CUTTING FORCE}

The present invention relates to a robot control method and device using a rigidity model and a cutting force model. More specifically, in a machining process using a robot, a technology for controlling a robot so that error correction of the machining path for the machining object can be easily performed. It's about.

Recently, manufacturing systems using multiple robots have been increasing, and in the production of products such as automobiles and aircraft, the demand for improved automatic manufacturing efficiency and increased processing precision using robots is continuously increasing.

When performing machining using a robot, machining errors occur due to the robot's rigidity, kinematics factors, etc., but it was not easy to precisely measure the machining errors as described above in the robot control device of the prior art. Accordingly, In technology, in order to improve the precision of processing errors, a method has been used to collect feedback on the robot's processing using multiple sensors and analyze this information to reduce the robot's processing errors. However, this method allows the user to There is a problem in that the developed control algorithm can only be applied to some open robot controllers.

In Korean Patent Publication No. 10-2015-0070370 (title of the invention: method for in-line calibration of an industrial robot, calibration system for performing such method, and industrial robot including such calibration system), at least three light rays At least one light source (7) produced using at least one light source (7) rigidly connected to the end (6) of the robotic arm (3) and adapted to determine the position of the light beam affecting the sensor in a two-dimensional plane. The optical position sensor 12 is configured to detect at least some of the rays generated by the at least one light source 7 at a predefined calibration location at the end 6 of the robotic arm 3. ) or located at a fixed location relative to the base portion 2 of the robot to affect at least one of the sensors 12 is disclosed.

Republic of Korea Patent Publication No. 10-2015-0070370

The purpose of the present invention to solve the above problems is to control the robot so that error correction of the machining path for the machining object can be easily performed in a machining process using a robot.

And, the purpose of the present invention is to enable the above-mentioned control of the robot to be performed while minimizing additional separate devices or complex calculations.

The technical problem to be achieved by the present invention is not limited to the technical problem mentioned above, and other technical problems not mentioned can be clearly understood by those skilled in the art from the description below. There will be.

The configuration of the present invention for achieving the above object includes a robot that is driven by having a plurality of links and a plurality of joints, a processing unit coupled to the end of the robot and equipped with a tool to perform processing on the processing object, and , a first step of providing a control unit that controls the posture of the robot when the tool works on the processing object; By using stiffness data, which is the stiffness value for each of the plurality of joints that varies depending on the posture of the robot, and information about cutting force, which is the force applied to the tool, the compliance error, which is the machining error calculated for the tool, is calculated and derived. Step 2; The reference position and reference direction, which are the position and direction of the tool, on the reference machining path, which is the machining path of the tool set in advance according to the machining target, and the actual position and direction, which are the real-time position and direction of the tool, using the compliance error. A third step of measuring the actual direction; A fourth step of dividing the reference machining path, which is the machining path of the tool set in advance according to the machining target, into a plurality of straight lines; A position where the position error of the tool is corrected using two straight lines passing through the nearest reference position, which is the reference position closest to the actual position among the plurality of straight lines, and two vertical connecting lines vertically connecting the actual position. The fifth step of deriving the outline reference position; A sixth step of deriving a contour reference direction, which is a direction in which a direction error of the tool is corrected, using the reference direction and the actual direction at the closest reference position; and a seventh step of controlling the position and direction of the tool by controlling the robot using the outline reference position and the outline reference direction.

In an embodiment of the present invention, in the second step, the compliance error (Δx) is derived by the following equation, where Δx is the compliance error, Cx is the Cartesian compliance matrix, Ф is the cutting force, and C _θ is a compliance matrix for the joints of the robot, and T may be a transpose matrix.

In an embodiment of the present invention, in the fourth step, the plurality of straight lines may be formed by dividing the reference processing path by an interpolator.

In an embodiment of the present invention, in the fifth step, the two straight lines are a first connecting straight line and a second straight line formed by connecting the closest reference position with each of two reference positions adjacent to the closest reference position. May include connecting straight lines.

In an embodiment of the present invention, a first projection point that is an intersection of a first vertical connection line connected from the actual position to the first connection straight line and the first connection straight line, and a first projection point connected from the actual position to the second connection straight line. A second projection point that is the intersection of the second vertical connection line and the second connection straight line may be set.

In an embodiment of the present invention, the outline reference position and the outline reference direction may be determined according to the positions of each of the first projection point and the second projection point with respect to each of the first connection straight line and the second connection straight line. there is.

In an embodiment of the present invention, when the first projection point is located on an extension of the first connection straight line and the second projection point is located on an extension of the second connection straight line, the closest reference position is the contour reference. Can be set to location.

In an embodiment of the present invention, in the sixth step, the outline reference direction may be a reference direction at the nearest reference position.

In an embodiment of the present invention, when the first projection point is located on an extension of the first connection straight line and the second projection point is located on the second connection straight line, the second projection point is set as the outline reference position. It can be.

In an embodiment of the present invention, when the first projection point is located on the first connection straight line and the second projection point is located on an extension of the second connection straight line, the first projection point is set as the outline reference position. It can be.

In an embodiment of the present invention, when the first projection point is located on the first connection straight line and the second projection point is located on the second connection straight line, among the first projection point and the second projection point, the A projection point closer to the actual position may be set as the outline reference position.

In an embodiment of the present invention, in the sixth step, the outline reference direction is a quaternion sphere using the reference direction at the nearest reference position and the reference direction at the reference position adjacent to the nearest reference position. It can be obtained by linear interpolation (SLERP) operation.

In an embodiment of the present invention, in the sixth step, the outline reference direction may be a vector direction of an axis formed along the longitudinal direction of the tool at the outline reference position.

The configuration of the present invention for achieving the above object includes a processing unit coupled to the robot and provided with the tool; And using the information on the rigidity data and the cutting force, the compliance error of the tool is derived, and the reference is the position and direction of the tool on the reference machining path, which is a machining path of the tool preset according to the machining target. By setting the position and the reference direction, information about the actual position and the actual direction is generated using the compliance error and the reference position and the reference direction, and the outline reference position and the outline reference direction are derived, It includes a control unit that controls the position and direction of the tool by transmitting a control signal to the robot.

The effect of the present invention according to the above configuration is that error correction of the machining path for the machining object is performed using the difference between the current position of the tool and the preset reference position, thereby improving error correction performance. .

Additionally, the effect of the present invention is that error correction control for the robot can be performed while minimizing additional separate devices or complex calculations, thereby significantly improving the efficiency of the machining process using the robot.

The effects of the present invention are not limited to the effects described above, and should be understood to include all effects that can be inferred from the configuration of the invention described in the detailed description or claims of the present invention.

1 is a schematic diagram of a processing device according to an embodiment of the present invention.
Figure 2 is a configuration diagram of a part of a processing device according to an embodiment of the present invention.
Figure 3 is a schematic diagram of position correction of a tool according to an embodiment of the present invention.
Figure 4 is a graph of position correction of a tool according to an embodiment of the present invention.
Figure 5 is an image for explanation of each case for measuring the outline reference position according to an embodiment of the present invention.
Figure 6 is an image of measurement of directional contour error according to an embodiment of the present invention.

Hereinafter, the present invention will be described with reference to the attached drawings. However, the present invention may be implemented in various different forms and, therefore, is not limited to the embodiments described herein. In order to clearly explain the present invention in the drawings, parts that are not related to the description are omitted, and similar parts are given similar reference numerals throughout the specification.

Throughout the specification, when a part is said to be "connected (connected, contacted, combined)" with another part, this means not only "directly connected" but also "indirectly connected" with another member in between. "Includes cases where it is. Additionally, when a part is said to “include” a certain component, this does not mean that other components are excluded, but that other components can be added, unless specifically stated to the contrary.

The terms used in this specification are merely used to describe specific embodiments and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, terms such as “comprise” or “have” are intended to indicate the presence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification, but are not intended to indicate the presence of one or more other features. It should be understood that this does not exclude in advance the possibility of the existence or addition of elements, numbers, steps, operations, components, parts, or combinations thereof.

Hereinafter, the present invention will be described in detail with reference to the attached drawings.

Figure 1 is a schematic diagram of a processing device according to an embodiment of the present invention, and Figure 2 is a configuration diagram of a part of the processing device according to an embodiment of the present invention. Additionally, FIG. 3 is a schematic diagram of position correction of the tool 111 according to an embodiment of the present invention, and FIG. 4 is a graph of position correction of the tool 111 according to an embodiment of the present invention.

And, Figure 5 is an image for explaining each case for measuring the outline reference position according to an embodiment of the present invention, and Figure 6 is an image for measuring the directional outline error according to an embodiment of the present invention.

First, in the first step, the robot 120, which is driven by having a plurality of links and a plurality of joints, is coupled to the end of the robot 120 and is provided with a tool 111 to perform processing on the processing object 10. A processing unit 110 that performs processing and a control unit 130 that controls the posture of the robot 120 when the tool 111 works on the processing object 10 may be provided.

And, the control unit 130 uses the information on the rigidity data and cutting force to derive the compliance error (Δx) of the tool and performs the reference machining, which is the machining path of the tool 111 set in advance according to the machining object 10. By setting the reference position and direction, which are the position and direction of the tool 111 on the path, information about the actual position (P _act ) and actual direction (O _act ) is obtained using the compliance error (Δx) and the reference position and reference direction. By generating a contour reference position (Pc) and a contour reference direction (Oc) and transmitting a control signal to the robot 120, the position and direction of the tool can be controlled.

The method of controlling the robot 120 of the present invention using the robot 120 control device of the present invention will be described in more detail in each step below.

In the second step after the first step as described above, using stiffness data, which is the stiffness value for each of a plurality of joints that varies depending on the posture of the robot 120, and information about the cutting force, which is the force applied to the tool 111, the tool It can be derived by calculating the compliance error (Δx), which is the processing error calculated for (111).

In the stiffness data, data on the posture of the robot 120 when performing each processing process is collected and stored, and using such stiffness data, the stiffness matrix (Kx) as shown below can be derived.

The control unit 130 stores in advance a reference processing path, which is the movement path of the tool 111 for the processing object 10. When the tool 111 moves along this reference processing path, the tool 111 The posture of the robot 120 at each location on the movement path may be stored in the control unit 130, and using this, the control unit 130 can predict the posture deformation of the robot 120 using the reference processing path. You can.

Here, the posture deformation of the robot 120 may mean that the posture of the robot 120 is changed and determined by reflecting the 3D rotation angle of each of the plurality of joints and the 3D position of each of the plurality of links.

The control unit 130 may store cutting force modeling data, which is data of cutting force on each of the plurality of machining parts 110 formed on the machining object 10. When the user inputs information about the type of processing object 10 into the control unit 130, the control unit 130 searches for information related to the processing object 10 in the cutting force modeling data and uses the matching data. The cutting force for each of the plurality of processing parts 110 can be derived from the processing object 10 and used.

When the cutting force is derived in this way, the robot 120 can be controlled even in an offline state, thereby improving the stability of machining using the robot 120.

Alternatively, the processing unit 110 may be provided with a force sensor 121 that measures cutting force. Here, the force sensor 121 may be formed outside or inside the processing unit 110, and the force sensor 121 measures the pressure and torque applied to the spindle connected to the tool 111 within the processing unit 110. can do.

Here, the pressure and torque applied to the spindle can be measured in a three-dimensional direction, and the control unit 130 can derive the cutting force using the pressure and torque values for the spindle, and for this purpose, the control unit 130 The tool 111 may be equipped with a dynamometer for measuring cutting force.

Compliance error (Δx) may occur due to insufficient structural rigidity of the robot manipulator. Key factors in compliance can be parts installed in the robot's joints, such as bearings, motor shafts, gears, and gearboxes. Since the stiffness of the link is much greater than that of the joint, the link can be assumed to be rigid and all compliance can be focused only on the joint portion. Since the compliance error (Δx) cannot be measured with a motor encoder, a compliance model can be used to estimate the compliance error (Δx) during processing of the robot 120.

The joint torque (τ) of the robot induced by an external wrench can be derived by [Equation 1] below.

[Equation 1]

Here, τ is the joint torque, J is the Jacobian matrix, and Ф may be an external force. And, as shown below, the external force may be a cutting force.

In addition, the partial derivative for [Equation 1] as described above can be expressed as [Equation 2], and [Equation 2] is rearranged in terms of the Cartesian stiffness matrix (Kx) as in [Equation 3] It can be.

[Equation 2]

[Equation 3]

Here, K _θ is the joint stiffness matrix, may be a complementary stiffness matrix. In general cases, the effect of the complementary stiffness matrix can be ignored because it is relatively small compared to the joint stiffness matrix. Therefore, the mathematical equation for the compliance model can be derived as [Equation 4].

And, the control unit 130 can derive the compliance error (Δx) using [Equation 4] below. Here, Δx is the compliance error, Cx is the Cartesian compliance matrix, Ф is the cutting force, C _θ is the compliance matrix for the joints of the robot 120, J is the Jacobian matrix, and T is the transpose matrix (hereinafter, the same ) can be. Hereinafter, the compliance matrix for the joints of the robot 120 may be referred to as a joint compliance matrix.

[Equation 4]

The Cartesian compliance matrix can be calculated by multiplying the Jacobian matrix and the joint compliance matrix. The joint compliance matrix may be a diagonal matrix in which each element is a joint compliance value, that is, C _θ =diag(1/K _θ1 , ..., 1/K _θi , ..., 1/K _θ6 ). Here, K _θi represents the stiffness value for each joint of the robot 120, and the stiffness value of the joint can be identified using the force/displacement relationship of each joint.

During machining, the actual posture (actual position and direction) of the tool 111 may deviate from the desired trajectory, resulting in contour errors, mainly due to deflection due to cutting force, i.e., compliance error and inaccuracy of kinematic parameters. In three-dimensional (3-D) robot processing, the contour error is the position contour error (ε _p ), which is the difference between the reference position of the tool 111 and the actual position (P _act ), and the reference direction of the tool 111 and the actual position. It can be composed of the direction contour error (ε _o ), which is the difference in direction (O _act ).

As shown in Figures 3 and 4, using the reference pose and the actual pose, the contour reference pose including the contour reference position (Pc) and the contour reference direction (Oc) ) can be derived, and the posture error (ε contour ) due to the positional contour error (ε _p ) and directional _contour error (ε _o ) in processing of the robot 120 can be derived.

The position contour error (ε _p ) can be defined as the orthogonal distance from the actual position (P _act ) to the desired reference trajectory in orthogonal coordinates, and the position contour error (ε _p ) is the difference between the actual position (P _act ) and the actual position (P _act ), it can be calculated using the nearest reference position (P _n ), which is the closest reference position to the reference trajectory. And, the directional contour error (ε _o ) can be calculated using the desired direction (direction of the contour reference posture) and the actual direction. These matters will be explained in detail in the remaining steps below.

In the third step after the second step as described above, the reference position and direction, which are the position and direction of the tool 111, on the reference machining path, which is the machining path of the tool 111 set in advance according to the machining object 10, and , the actual position and direction, which are the real-time position and direction of the tool 111, can be measured using the compliance error (Δx).

The control unit 130 can derive the real-time position and direction of the tool 111, which are the real-time position and direction, using the reference position and direction of the reference machining path and the compliance error (Δx). Specifically, the reference position on the reference processing path is expressed as a 3-dimensional position, the reference direction is set to a 3-dimensional direction, and the compliance error (Δx) may include both 3-dimensional position and direction. And, by calculating the error due to the compliance error (Δx) in the reference position and reference direction, the real-time, that is, the actual position and direction of the tool 111 in the current state in terms of the three-dimensional position and direction are calculated. It can be calculated.

In the fourth step after the third step as described above, the reference machining path, which is the machining path of the tool 111 set in advance according to the machining object 10, can be divided into a plurality of straight lines. Here, a plurality of straight lines may be formed by dividing the reference processing path by an interpolator.

The control unit 130, which controls the robot 120, can provide a reference posture set consisting of a reference position and reference direction at each sampling cycle as shown in [Equation 5] below to plan the machining path of the robot 120. there is.

[Equation 5]

Here, Pr and Or may be the reference position and direction of Cartesian coordinates. Additionally, Xr, Yr, and Zr may be reference positions of the X, Y, and Z axes, respectively. And, Rx _r , Ry _r , and Rz _r may be reference directions of the Rx, Ry, and Rz axes, respectively, expressed as Euler angles.

In order to derive the positional contour error (ε _p ) and the directional contour error (ε _o ), the reference posture as above must be found, and the reference trajectory by the reference processing path is divided into a plurality of straight lines divided by an interpolator, and as a result, A reference location can be created for each sampling period. And, a reference direction, which is the direction of the tool 111 at each reference position, can be created.

In the fifth step after the fourth step as described above, each of the two straight lines passing through the nearest reference position (P _n ), which is the reference position closest to the actual position (P _act ) among the plurality of straight lines, and the actual position (P _act ) By using two vertical connecting lines, the contour reference position (Pc), which is a position where the position error of the tool 111 is corrected, can be derived. Here, the two straight lines may include a first connection straight line and a second connection straight line formed by connecting each of the two reference positions adjacent to the nearest reference position (P _n ) and the nearest reference position (P _n ). .

The position contour error (ε _p ) can be calculated using the distance between the contour reference position (Pc) found on the line segment connecting the actual position (P _act ) and the interpolated reference position. To this end, first, set the search range and set a plurality of reference positions on the reference machining path within the search range, and then select the closest reference position (P _{act) that is closest to the actual position (P act} ) of the tool 111 among the plurality of reference positions. It can be derived by calculating P _n ).

Here, experimentally, it may be desirable to have 20 or more reference positions within the search range, but it is not necessarily limited thereto.

After deriving the nearest reference position (P _n ) as above, two reference positions adjacent to the nearest reference position (P _n ), that is, the closest reference position (P _n ) with respect to the machining direction on the reference machining path, are Two reference positions located in the front and back directions, a first adjacent reference position (P _n-1 ) and a second adjacent reference position (P _n+1 ), may be set.

Here, a first connecting straight line is formed by connecting the first adjacent reference position (P _n-1 ) and the nearest reference position (P _n ), and the second adjacent reference position (P _n+1 ) and the nearest reference position (P n) are formed. A second connecting straight line can be formed by connecting P _n ).

And, the first projection point (H ₁ ), which is the intersection of the first vertical connecting line and _the first connecting straight line, which is connected from the actual position (P act) to the first connecting straight line, and the actual position (P _act ) to the second connecting straight line. A second projection point (H ₂ ), which is the intersection of the second vertical connection line and the second connection straight line, may be set.

Using the closest reference position (P _n ), first projection point (H ₁ ), and second projection point (H ₁ ) as described above, the contour reference position (Pc) can be derived according to each case, and thus This will be explained in detail below.

Figure 5(a) is for the first case, Figure 5(b) is for the second case, and Figure 5(c) is for the third case. And, (d) in Figure 5 is for the fourth case.

As shown in Figure 5, the outline reference position (Pc) and the outline reference direction (Oc) can be determined depending on the positions of the first projection point and the second projection point for each of the first connection straight line and the second connection straight line. . Specifically, four cases can be established _, and the actual position ₍ The contour reference position (Pc) for P _act ) can be determined.

And, as shown in [Equation 6-1] and [Equation 6-2] below, calculation of the first projection point (H ₁ ), the second projection point (H ₁ ), and the outline reference position (Pc) r1 and r2 can be defined to determine the weight for .

[Equation 6-1]

=

[Equation 6-2]

=

Here, Pc may represent the outline reference position, P _n-1 may represent the first adjacent reference position, and P _n+1 may represent the second adjacent reference position. And, H ₁ may represent the first projection point, and H ₂ may represent the second projection point.

As shown in (a) of Figure 5, in the first case, the first projection point (H ₁ ) is located on the extension line of the first connection straight line and the second projection point (H ₂ ) is located on the extension line of the second connection straight line. When located, the closest reference position (P _n ) may be the outline reference position (Pc).

In the first case, since r ₁ >1 and r ₂ >1, the closest reference position (P _n ) closest to the actual position (P _act ) may be the contour reference position (Pc).

As shown in Figure 5 (b), in the second case, the first projection point (H ₁ ) is located on the extension of the first connection straight line and the second projection point (H ₂ ) is located on the second connection straight line. In this case, the second projection point (H ₂ ) may be the outline reference position (Pc).

And, as shown in Figure 5 (c), in the third case, the first projection point (H ₁ ) is located on the first connection straight line and the second projection point (H ₂ ) is located on the extension of the second connection straight line. When located, the first projection point (H ₁ ) may be the outline reference position (Pc).

In the second and third cases, one of the first projection point (H ₁ ) and the second projection point (H ₂ ) is located inside one connecting straight line, and the other projection point is located inside the other connecting straight line. It can be located outside. And r ₁ and r ₂ , and the projection point located inside the connecting straight line, that is, on the connecting straight line, may be the outline reference position (Pc).

As shown in (d) of Figure 5, in the fourth case, when the first projection point (H ₁ ) is located on the first connection straight line and the second projection point (H ₂ ) is located on the second connection straight line , Among the first projection point (H ₁ ) and the second projection point (H ₂ ), the projection point that is closer to the actual position (P _act ) may be the outline reference position (Pc).

In the fourth case, the first projection point (H ₁ ) and the second projection point (H ₂ ) are each located inside the first connection straight line and the second connection straight line, and 0≤r ₁ ≤1 and 0≤r ₂ ≤ 1, and the projection point closest to the actual position (P _act ) may be the contour reference position (Pc).

And, in the second to fourth cases as described above, the outline reference position (Pc) on the connecting straight line is the nearest reference by [Equation 7-1] or [Equation 7-2] below. It can be calculated using linear interpolation of the first adjacent reference position (P _n-1 ) and the second adjacent reference position (P _n+1 ), which are reference positions adjacent to the _position (P n).

[Equation 7-1]

[Equation 7-2]

Here, Pc may represent the outline reference position, P _n-1 may represent the first adjacent reference position, and P _n+1 may represent the second adjacent reference position.

And, the position contour error (ε _p ) can be calculated between the reference position and the actual position (P _act ) as shown in [Equation 8] below.

[Equation 8]

Here, ε _p represents the position contour error, Pc represents the outline reference position, and Pa may represent the actual position (P _act ).

In the sixth step after _the fifth step as described above, the outline reference _direction ( Oc) can be derived.

Because general robot 120 control is intuitive, Euler angles can be used to express the direction of the tool 111. However, Euler angles may not be suitable for handling orientation correction because changes in one axis affect the other axes due to sequential rotation.

Therefore, quaternions can be convenient for direction calculations because they simultaneously process rotation axes and are more efficient than rotation matrices. The directional contour error (ε _o ) can be calculated using the reference direction and the actual direction (O _act ), and the outline reference direction (Oc) can be calculated using different methods depending on each case in the third step described above. can be derived.

Specifically, in the case of the first case described above, the outline reference direction (Oc) may be a reference direction at the nearest reference position (P _n ). And, in the case of the second to fourth cases described above, the outline reference direction (Oc) is the reference direction at the nearest reference position (P _n ) and the reference position adjacent to the nearest reference position (P _n ) (the first It can be obtained by quaternion spherical linear interpolation (SLERP) operation using the reference direction at the adjacent reference position (P _n-1 ) and the second adjacent reference position (P _n+1 ).

In the second to fourth cases, the outline reference direction (Oc) is the nearest reference direction (O _n ), which is the reference direction at the nearest reference position (P _n ), and the above-mentioned first adjacent reference position (P _{n- 1} ) and the second adjacent reference position (P _n+1 ). Quaternion spherical linearity of the first adjacent reference direction (O _n-1 ) and the second adjacent reference direction (O _n+1 ), respectively. It can be obtained through interpolation (SLERP) operation. Here, the quaternion spherical linear interpolation (SLERP) operation can be calculated using [Equation 9-1] and [Equation 9-2] below.

[Equation 9-1]

[Equation 9-2]

Here, as described above, q _c represents the quaternion of the contour reference direction (Oc), q _n represents the quaternion of the nearest reference direction (O _n ), and q _n-1 represents the first adjacent reference direction (O _{n -1} ) can represent a quaternion. And, q _n+1 may represent a quaternion in the second adjacent reference direction (O _n+1 ).

In addition, the three-dimensional coordinates for the direction outline error (ε o ) can be derived _using the difference between the three-dimensional coordinates of the outline reference direction (Oc) obtained as above and the three-dimensional coordinates of the actual direction (O _act ). . This will be explained in detail in the fifth step description below.

In addition to one method of deriving the outline reference direction Oc as described above, the outline reference direction Oc may be a vector direction of an axis formed along the longitudinal direction of the tool 111 at the outline reference position Pc. Here, the axis vector formed along the longitudinal direction of the tool 111 may be a z-axis vector in the tool 111.

Specifically, in order to calculate the directional contour error (ε _o ), the z-axis vector can be extracted from the actual direction (O _act ) at the actual position (P _act ) and the reference direction at the outline reference position (Pc). And, as shown in [Equation 10] below, the directional contour error (ε _o ) is calculated using the angle difference between the z-axis vector in each reference direction at the actual direction (O _act ) and the outline reference position (Pc). It can be.

[Equation 10]

Here, ε _o represents the directional contour error, ν' _c represents the z-axis vector in the reference direction at the outline reference position (Pc), and ν _a may represent the z-axis vector in the actual direction (O _act ).

And, by correcting the angle difference using the directional outline error (ε _o ) as described above, the outline reference direction (Oc) can be derived.

In the seventh step after the sixth step as described above, the position and direction of the tool 111 can be controlled by controlling the robot 120 using the outline reference position (Pc) and the outline reference direction (Oc). As above, for the tool 111, the _positional contour error (ε p ) can be derived using the actual position (P _act ) and the contour reference position (Pc), and the actual direction (O _act ) and the contour reference direction The directional contour error (ε _o ) can be derived using (Oc). Additionally, a control signal can be transmitted to the robot 120 to change the position and direction of the tool 111 so that the positional contour error (ε _p ) and the directional contour error (ε _o ) are corrected.

The above control signal is generated by the control unit 130, the joint axes of the robot 120 are individually controlled, and the controller for each individual axis may be configured in a cascade control structure connected to the position, speed, and current loop. . In this control unit 130, the positional contour error (ε _p ) and the directional contour error (ε _o ) can be derived using the contour control algorithm described above.

Since the conventional robot controller does not allow modification of the internal control structure, an additional PI (proportional-integral) position control loop is designed in Cartesian coordinates to correct the contour error, thereby enabling the control unit 130 of the present invention. can be formed.

Therefore, the above [Equation 5] can be modified to derive the following [Equation 11].

[Equation 11]

Here, P _r,m may be a modified reference position, and O _r,m may be a modified reference direction. And, K _PI may be the PI control gain for each axis, and may represent the position profile error (ε _p ) and direction profile error (ε _o ) for each axis.

Accordingly, the position correction amount of the tool 111 may be expressed as [Equation 12] below.

[Equation 12]

Here, is the position contour error (ε _p ) for each axis, Pc is the contour reference position, and Pa is the actual position (P _act ).

To compensate for the directional contour error (ε _o ), the z-axis vector can be used as above, and in this case, as shown in [Equation 13], the actual direction (O _act ) and the outline reference position (Pc ) A unit direction vector (n) orthogonal to each z-axis vector in the reference direction may be generated. Next, error correction for the unit direction vector (n) can be performed by the amount of the direction contour error (ε _o ).

[Equation 13]

Here, n is a unit direction vector, ν _a represents a z-axis vector in the actual direction (O _act ), and ν _c may represent a z-axis vector in the reference direction at the outline reference position (Pc).

As shown in Figure 6, quaternion spherical linear interpolation (SLERP) can be used as above to compensate for the direction contour error (ε _o ), and the quaternion is a unit direction vector (n) and a rotation amount (ε _o) . ) and can be converted to Euler angles. By [Equation 14] below, q->E can represent the conversion from quaternion to Euler angle.

[Equation 14]

Here, n is the unit direction vector, and is the direction contour error (ε _o ) for each axis.

A robot 120 control device of the present invention formed as described above; And processing, including a display device that displays values for the shape and posture changes of the robot 120, the reference processing path, and the amount of correction in each of the outline reference position (Pc) and outline reference direction (Oc) for the tool 111. A system can be formed.

On the display device, the three-dimensional shape of the robot 120 and some or all of the above-mentioned numerical values may be displayed, and the user can visually check the operating status of the robot 120 and the tool 111 while looking at the display screen. You can check this.

When using the robot 120 control method and device of the present invention as described above, the difference between the actual position (P _act ), which is the current position of the tool 111, and the preset reference position is used to control the processing target 10. Since error correction of the processing path is performed, error correction performance can be improved.

In addition, error correction control for the robot 120 can be performed while minimizing additional devices or complex calculations, thereby significantly improving the efficiency of the machining process using the robot 120.

The description of the present invention described above is for illustrative purposes, and those skilled in the art will understand that the present invention can be easily modified into other specific forms without changing the technical idea or essential features of the present invention. will be. Therefore, the embodiments described above should be understood in all respects as illustrative and not restrictive. For example, each component described as unitary may be implemented in a distributed manner, and similarly, components described as distributed may also be implemented in a combined form.

The scope of the present invention is indicated by the patent claims described below, and all changes or modified forms derived from the meaning and scope of the claims and their equivalent concepts should be construed as being included in the scope of the present invention.

10: Processing object 110: Processing part
111: Tool 120: Robot
121: Force sensor 130: Control unit

Claims (15)

A robot that is driven by having a plurality of links and a plurality of joints, a processing unit coupled to an end of the robot and equipped with a tool to perform processing on the processing object, and when the tool works on the processing object, the robot A first step of providing a control unit to control the posture of;
By using stiffness data, which is the stiffness value for each of the plurality of joints that varies depending on the posture of the robot, and information about cutting force, which is the force applied to the tool, the compliance error, which is the machining error calculated for the tool, is calculated and derived. Step 2;
The reference position and reference direction, which are the position and direction of the tool, on the reference machining path, which is the machining path of the tool set in advance according to the machining target, and the actual position and direction, which are the real-time position and direction of the tool, using the compliance error. A third step of measuring the actual direction;
A fourth step of dividing the reference machining path, which is the machining path of the tool set in advance according to the machining target, into a plurality of straight lines;
A position where the position error of the tool is corrected using two straight lines passing through the nearest reference position, which is the reference position closest to the actual position among the plurality of straight lines, and two vertical connecting lines vertically connecting the actual position. The fifth step of deriving the outline reference position;
A sixth step of deriving a contour reference direction, which is a direction in which a direction error of the tool is corrected, using the reference direction and the actual direction at the closest reference position; and
A seventh step of controlling the position and direction of the tool by controlling the robot using the outline reference position and the outline reference direction. A robot control method using a rigidity model and a cutting force model, comprising a.
In claim 1,
In the second step, the compliance error (Δx) is derived by the following equation, where Δx is the compliance error, Cx is the Cartesian compliance matrix, Ф is the cutting force, and C _θ is the compliance for the joints of the robot. A robot control method using a stiffness model and a cutting force model, wherein J is a Jacobian matrix, and T is a transpose matrix.

In claim 1,
In the fourth step, the plurality of straight lines are formed by dividing the reference processing path by an interpolator.
In claim 1,
In the fifth step, the two straight lines include a first connection straight line and a second connection straight line formed by connecting each of two reference positions adjacent to the nearest reference position and the nearest reference position. A robot control method using a rigidity model and cutting force model.
In claim 4,
A first projection point that is an intersection of a first vertical connection line connected from the actual position to the first connection straight line, and a second vertical connection line connected from the actual position to the second connection straight line and the first projection point. A robot control method using a rigidity model and a cutting force model, characterized in that the second projection point, which is the intersection of two connecting straight lines, is set.
In claim 5,
Stiffness model and cutting force, wherein the outline reference position and the outline reference direction are determined according to the positions of the first projection point and the second projection point with respect to each of the first connection straight line and the second connection straight line. Robot control method using models.
In claim 6,
When the first projection point is located on an extension of the first connecting straight line and the second projection point is located on an extending line of the second connecting straight line, the closest reference position is set as the outline reference position. Robot control method using rigidity model and cutting force model.
In claim 7,
In the sixth step, the outline reference direction is a reference direction at the nearest reference position. A robot control method using a rigidity model and a cutting force model.
In claim 6,
When the first projection point is located on an extension of the first connection straight line and the second projection point is located on the second connection straight line, a rigidity model characterized in that the second projection point is set as the outline reference position; Robot control method using cutting force model.
In claim 6,
When the first projection point is located on the first connection straight line and the second projection point is located on an extension of the second connection straight line, a rigidity model characterized in that the first projection point is set to the outline reference position; Robot control method using cutting force model.
In claim 6,
When the first projection point is located on the first connection straight line and the second projection point is located on the second connection straight line, the projection point that is closer to the actual location among the first projection point and the second projection point is the A robot control method using a rigidity model and a cutting force model, which are set to the contour reference position.
According to any one selected from claims 9 to 11,
In the sixth step, the outline reference direction is obtained through a quaternion spherical linear interpolation (SLERP) operation using the reference direction at the nearest reference position and the reference direction at the reference position adjacent to the nearest reference position. A robot control method using a rigidity model and a cutting force model, characterized in that:
According to any one selected from claims 7 and 9 to 11,
In the sixth step, the outline reference direction is a vector direction of an axis formed along the longitudinal direction of the tool at the outline reference position. A robot control method using a rigidity model and a cutting force model.
In the robot control device using a rigidity model and a cutting force model that performs the robot control method using the rigidity model and cutting force model of claim 1,
a processing unit coupled to the robot and equipped with the tool; and
Using the information on the rigidity data and the cutting force, the compliance error of the tool is derived, and the reference position, which is the position and direction of the tool, on the reference machining path, which is the machining path of the tool preset according to the machining target By setting the reference direction, information on the actual position and the actual direction is generated using the compliance error and the reference position and the reference direction, and the outline reference position and the outline reference direction are derived, A robot control device using a rigidity model and a cutting force model, comprising a control unit that controls the position and direction of the tool by transmitting a control signal to the robot.
Robot control device using the rigidity model and cutting force model according to claim 14; and
A machining system comprising a display device that displays values for changes in the shape and posture of the robot, the reference machining path, and the amount of correction in each of the outline reference position and the outline reference direction for the tool.