TWI706843B - Alignment method for robot tool and alignment system for using the same - Google Patents

Abstract

An alignment system for robot tool including a robotic arm adopting a first coordinate system, a tool arranged on a flange of the robotic arm, and an image device adopting a second coordinate system is disclosed, wherein an image sensing area is established by the image device. An alignment method is also disclosed and includes following steps of: controlling the robotic arm to move for leading a tool working point (TWP) of the tool enters the image sensing area; recording a current gesture of the robotic arm as well as a specific coordinate of the TWP currently upon the second coordinate system; obtaining a transformation matrix previously built for describing a relationship between the first and the second coordinate systems; and importing the specific coordinate and the current gesture to the transformation matrix for calculating an absolute position of the TWP upon the first coordinate system.

Description

Correction method and correction system of robot tool

The present invention relates to a calibration method and a calibration system of a robot, in particular to a calibration method and a calibration system for calibrating a tool configured on the robot.

At present, robotic arms have been widely used in automated production procedures. Specifically, production line personnel usually install various types of tools (such as fixtures, connection tools, welding tools, etc.) on the flange surface of the robot arm, so that the robot arm can use these tools to achieve production. Line automation process.

Since the robot arm itself only knows the position of the flange surface, but does not know the actual position of the Tool Working Point (TWP) on the tool, after installing or replacing the tool, it must be calibrated first to make the machine The controller of the arm obtains the position information of the working point of the tool.

The position accuracy of the tool working point will affect the accuracy of the automated process. For example, if the position of the tool working point is wrong, the product may be damaged during the operation of the robot arm, and the production line may even stop in severe cases. Therefore, how to effectively calibrate tools is a very important topic in related fields.

Since tools may have machining tolerances during manufacturing, geometric deviations may also occur when the tools are installed on the robot arm. Moreover, if the tool is used for a long time, it may also cause the tool to wear and cause gaps or deformation. The above reasons may affect the position accuracy of the tool working point.

In order to solve the above problems, the following correction methods are currently available on the market: (1) Manual teaching method: Manually operate the robot arm, so that the robot arm reaches the same position in the space in several different postures, thereby performing tool work Point position recording; (2) Contact correction method: the robot arm drives the tool to move, and the tool working point sequentially touches the squares of each axis in the correction device and obtains the direction offset, thereby offsetting according to the direction (3) Non-contact correction method: the robot arm drives the tool to move in the space formed by the light blocking correction device, so that the tool working point reaches the intersection of light in several different postures, thereby establishing The sample of the tool and the corresponding deviation are generated, and then the tool is corrected according to the sample and the deviation.

However, the above-mentioned manual teaching methods rely entirely on human eyes to teach in an unquantifiable manner (that is, judging whether the robot arm reaches the desired position based on human eyes and human experience), and its accuracy is completely affected by the operator’s experience and technology. So it is quite inaccurate. The above-mentioned contact-type calibration method requires that the working point of the tool touch each block, so the tool may be worn during the calibration process, and the more high-precision tool, the greater the impact.

Although the above non-contact correction method does not have the problem of tool wear, this method can only obtain the deviation between the tool working point and the flange surface of the robot arm, and cannot obtain the absolute value of the tool working point on the robot arm coordinate system. position. Moreover, when the tool is changed, it is necessary to recreate the sample and obtain the deviation again, which is quite troublesome for the production line personnel.

The main purpose of the present invention is to provide a calibration method and system for robot tools, which can effectively calibrate the tools configured on the robot arm, so as to eliminate the tolerances that exist in the production and manufacture of the tools, the deviations generated during assembly, or the tool factors. Use wear and produce errors.

In order to achieve the above object, the calibration system of the present invention mainly includes a robotic arm using a first coordinate system, a tool arranged on a flange surface of the robotic arm, and an imaging device using a second coordinate system, wherein the image The device creates an image sensing area. The calibration method of the present invention includes the following steps: controlling the movement of the robotic arm so that a working point of the tool enters the image sensing area; recording the current posture of the robotic arm and a specific position of the working point on the second coordinate system Coordinates; obtain a conversion matrix that records the correspondence between the first coordinate system and the second coordinate system; and nest the specific coordinates and the current posture into the conversion matrix to obtain the working point in the first An absolute position on the coordinate system.

In contrast to the prior related art, the calibration system and calibration method of the present invention do not require manual teaching of the robotic arm, and use non-contact light sensing means, which can provide higher-precision calibration results without being affected by calibration procedures. This leads to wear of the tool itself.

In addition, with the calibration system and the calibration method of the present invention, the absolute position of one or more working points on the tool in the coordinate system of the robot arm can be directly obtained. Compared with the non-contact correction method in the prior related art, there is no need to create a sample of the tool, and compared with the deviation value used in the prior related art, the correction accuracy can be further improved.

1: Calibration system

2: robotic arm

20: Controller

21: Flange surface

22: Tools

221: Tool work point

3: Imaging device

30: image processing unit

31: Light source device

32: photosensitive device

33: Image sensing area

4: Two-dimensional image of tool working point

41, 42, 4n: Tool two-dimensional image

5: Two-dimensional image of flange surface

S10~S16: Calibration steps

S20~S32: Steps to obtain absolute position

S40~S50: Steps to build transformation matrix

S60~S82: Steps to construct the coordinate system of the imaging device

S90~S98: 3D image creation steps

A: First position

B: second position

C: third position

T _I ^B : Conversion matrix

{B}: Robotic arm coordinate system

{I}: Coordinate system of image device

Fig. 1 is a first specific embodiment of the schematic diagram of the correction system of the present invention.

Fig. 2 is a first specific embodiment of a flowchart of the correction method of the present invention.

FIG. 3 is a second specific embodiment of the flowchart of the calibration method of the present invention.

Fig. 4 is a first specific embodiment of the correction diagram of the present invention.

Fig. 5 is a first specific embodiment of the establishment flow chart of the present invention.

FIG. 6 is a first specific embodiment of the schematic diagram of establishing the coordinate system of the imaging device of the present invention.

Fig. 7 is a second specific embodiment of the establishment flow chart of the present invention.

Fig. 8 is a first specific embodiment of a schematic diagram of a conversion matrix of the present invention.

FIG. 9 is a first embodiment of the image information schematic diagram of the present invention.

FIG. 10 is a second embodiment of the image information schematic diagram of the present invention.

FIG. 11 is a first specific embodiment of the flow chart of creating a three-dimensional image of the present invention.

For a preferred embodiment of the present invention, with the drawings, the detailed description is as follows.

First, please refer to FIG. 1, which is a first specific embodiment of a schematic diagram of the calibration system of the present invention. Figure 1 discloses the calibration system of the robot tool of the present invention (hereinafter referred to as the calibration system 1 in the specification). As shown in Figure 1, the calibration system 1 of the present invention mainly includes a controller 20, a robot arm 2, a tool 22 And imaging device 3.

One end of the robotic arm 2 is fixedly arranged on the plane of the machine platform, and the other end extends outward and has a flange 21. The flange surface 21 is a well-known technology in the robotics field, and will not be repeated here. In the present invention, the robot arm 2 uses an independent robot arm coordinate system. In the embodiment of FIG. 1, the robot arm coordinate system is represented by {B}, and the three-axis coordinates of the robot arm coordinate system {B} are represented by X ^B , Y ^B , and Z ^{B respectively} .

The tool 22 is arranged on the flange surface 21 of the robotic arm 2 and based on the usage of the tool 22, a Tool Working Point (TWP) 221 can be set thereon. Specifically, the tool working point 221 refers to a specific point on the tool 22 that is mainly used to perform operations (such as the welding point of the welding tool, the clamping point of the fixture, etc.), and can be freely controlled by the user of the robot arm 2 set up. For ease of understanding, a single tool working point 221 will be taken as an example in the description below, but a single tool working point 221 is not limited to a single tool 22.

One of the main purposes of the present invention is to find the tool work on the tool 22 through the calibration procedure when the robot arm 2 is replaced with a new tool 22, or when the tool 22 has been used for a period of time and the accuracy has decreased. The absolute position of point 221 on the robot arm coordinate system {B}. After the controller 20 of the robotic arm 2 obtains the correct position of the tool working point 221 of the tool 22, it can effectively eliminate the tolerances in the production and manufacture of the tool 22, the deviation generated when the tool 22 is assembled to the flange surface 21, or The tool 22 has errors due to wear.

The controller 20 is electrically connected to the robotic arm 2 for controlling various motions of the robotic arm 2, such as displacement and rotation. Moreover, when a tool 22 is provided on the robotic arm 2, the controller 20 can control the tool 22 at the same time to implement operations such as welding and clamping.

The imaging device 3 is arranged beside the robotic arm 2 and an image sensing area 33 is established within the moving range of the robotic arm 2. In the present invention, the imaging device 3 uses one imaging device coordinate system. In the embodiment of FIG. 1, the coordinate system of the imaging device is represented by {I}, and the three-axis coordinates of the coordinate system {I} of the imaging device are represented by X ^I , Y ^I , and Z ^{I respectively} .

It is worth mentioning that the robot arm coordinate system {B} can be determined after the robot arm 2 is manufactured, and the imaging device coordinate system {I} is after the robot arm 2 and the imaging device 3 are set up. , And then use the robot arm coordinate system {B} as the basic coordinate system to convert the established one (will be detailed in the following paragraphs). One of the technical features of the present invention is that a robot arm coordinate system {B} and imaging device coordinate system {I} can be defined by the determined robot arm coordinate system {B} and imaging device coordinate system {I} The conversion matrix of the correspondence between. Therefore, in the calibration procedure of the tool 22, the controller 20 can use the transformation matrix to calculate the absolute position of the tool working point 221 on the robot arm coordinate system {B}. In an embodiment, the absolute position of the tool working point 221 on the robot arm coordinate system can be represented by coordinates (X _t ^B , Y _t ^B , Z _t ^B ).

As shown in FIG. 1, the imaging device 3 mainly includes a light source device 31 and a photosensitive device 32 arranged in parallel, and an image processing unit 30 electrically connected to the photosensitive device 32. The light source device 31 emits a light source toward the photosensitive device 32, and the photosensitive device 32 captures the light source emitted by the light source device 31, so that the imaging device 3 can establish the image sensing area 33 by the light source. In an embodiment, the photosensitive device 32 may be, for example, a camera, which is not limited herein.

,

,

)來表示。 In the present invention, when the tool working point 221 of the tool 22 on the robot arm 2 enters the image sensing area 33, part of the light source sensed by the photosensitive device 32 will be cut off by the tool working point 221, whereby the image processing unit 30 can The current position of the tool working point 221 is determined by truncating the information, and the coordinate information of this position in the image device coordinate system {I} is obtained. For example, the current position of the tool working point 221 in the coordinate system of the imaging device can be coordinate (

,

,

)To represent.

In one embodiment, the light source device 31 can emit visible light or invisible light, such as X-ray, laser, infrared light, ultraviolet light, etc., which is not limited herein. It is worth mentioning that, as long as each device in the correction system 1 has a sufficient security level (for example, a sufficient waterproof level), and the refraction phenomenon of the light source under different media is eliminated, the correction system of the present invention 1 and the calibration method can also be used in other media (such as water, oil or other solutions) besides air.

In the calibration system 1 of the present invention, when different types of tools 22 are installed, the same type of tools 22 are replaced, or the existing tools 22 are deformed and worn due to long-term use, the controller 20 controls the robot arm 2 to execute Related calibration procedures.

Specifically, when the tool 22 is installed on the flange surface 21 of the robotic arm 2, the controller 20 controls the movement of the robotic arm 2 so that the tool working point 221 of the tool 22 enters the image sensing area established by the imaging device 3 33 in. When the tool working point 221 is located in the image sensing area 33, the controller 20 records the current posture of the robot arm 2 and records that the current position of the tool working point 221 corresponds to a specific coordinate on the coordinate system {I} of the imaging device. In this embodiment, the current posture may include, for example, the current displacement information (such as X-axis displacement, Y-axis displacement, and Z-axis displacement) of the flange surface 21 relative to the origin of the robot arm coordinate system (not shown in the figure). ) And rotation information (such as X-axis rotation, Y-axis rotation, and Z-axis rotation) are not limited here.

Then, the controller 20 can integrate the current posture and the specific coordinates into a pre-built conversion matrix to obtain the absolute position of the tool working point 221 on the robot arm coordinate system {B} through conversion. As mentioned above, the conversion matrix records the correspondence between the robot arm coordinate system {B} and the imaging device coordinate system {I}. In one embodiment, the transformation matrix is a homogeneous transformation matrix (Homogeneous transformation matrix), which is not limited herein.

In short, with the calibration system 1 of the present invention, the conversion matrix can be derived in advance after the robot arm coordinate system {B} and the imaging device coordinate system {I} are determined. In this way, when the tool 22 on the robotic arm 2 is to be calibrated, the tool can be obtained by the coordinates of the tool working point 221 in the imaging device coordinate system {I}, the posture of the robotic arm 2 and the conversion matrix The absolute position of the working point 221 on the robot arm coordinate system {B}. In this way, the operation accuracy of the robotic arm 2 can be effectively improved.

Please continue to refer to FIG. 2, which is a first specific embodiment of the flow chart of the calibration method of the present invention. First, the user determines whether the robot arm 2 / imaging device 3 in the calibration system 1 is reinstalled or replaced (step S10). If the calibration system 1 is installed for the first time, or when one of the robotic arm 2 and the imaging device 3 is replaced, the conversion between the robotic arm coordinate system {B} and the imaging device coordinate system {I} needs to be established first The relationship (step S12), that is, the conversion matrix is established.

Next, the user or the controller 20 of the calibration system 1 needs to determine whether the tool 22 set on the flange surface 21 of the robot arm 2 needs to be calibrated (step S14), that is, it is determined whether the tool 22 on the robot arm 2 is replaced (the same type Or different types), or whether the accuracy of the tool 22 is reduced due to long-term use.

If it is determined that the tool 22 needs to be calibrated, the calibration system 1 controls the robot arm 2 to move/rotate via the controller 20 to obtain a tool working point 221 on the tool 22 in the robot arm coordinate system through the above calibration procedure. } On the absolute position (step S16). After obtaining the absolute position of the tool working point 221 on the robot arm coordinate system {B}, the controller 20 can accurately grasp the position of the tool 22, thereby improving the accuracy of the robot arm 2.

Please refer to FIG. 3 and FIG. 4 at the same time. FIG. 3 is a second specific embodiment of the flowchart of the calibration method of the present invention, and FIG. 4 is the first specific embodiment of the calibration schematic diagram of the present invention. FIG. 3 discloses the calibration method of the robot tool of the present invention (hereinafter referred to as the calibration method in the specification), and the calibration method is applied to the calibration system 1 shown in FIG. 1.

As shown in FIG. 3, first, the user of the robot arm 2 manually or the controller 20 of the robot arm 2 determines whether the tool 22 on the robot arm 2 needs to be corrected (step S20). In one embodiment, the user/controller 20 is mainly when the tool 22 is replaced and the tool When the use time of 22 exceeds the first threshold, or the accuracy of the tool 22 is less than the second threshold, it is determined that the tool 22 needs to be corrected.

Then, when the tool 22 (a new tool or an old tool that needs to be calibrated) is set on the flange surface 21 of the robot arm 2, the controller 20 controls the robot arm 2 to move so that the tool working point 221 of the tool 22 enters In the image sensing area 33 created by the image device 3 (step S22). As mentioned above, the imaging device 3 mainly emits a light source through the light source device 31, and captures the light source emitted by the light source device 31 through the photosensitive device 32, whereby the light source creates an image sensing area within the moving range of the robotic arm 2 33.

As shown in FIG. 4, when the tool working point 221 is located in the image sensing area 33, the imaging device 3 can determine the current position of the tool working point 221 by using the truncated information captured by the photosensitive device 32 to obtain the tool working point 221 The current specific coordinates on the coordinate system of the imaging device {I}. In the embodiment of FIG. 4, the specific coordinates are represented by (X _t ^I , Y _t ^I , Z _t ^I ), where I represents the image device coordinate system {I}, and t represents the tool operating point 221. Moreover, the image device coordinate system {I} in the present invention is a coordinate system drawn by the robot arm coordinate system {B} as the reference coordinate system (detailed later). When the tool working point 221 is located in the image sensing area 33, the controller 20 of the robotic arm 2 can also know the specific coordinates of the tool working point 221.

It is worth mentioning that as the direction and angle of the robot arm 2 move, the robot arm 2 will also have different postures. In an embodiment, the posture may be, for example, displacement information and rotation information of the flange surface 21 relative to the origin of the robot arm coordinate system, which is not limited herein.

Go back to Figure 3. When the tool working point 221 is located in the image sensing area 33, the controller 20 records the current posture of the robot arm 2 (step S24), and at the same time records the specific coordinate (I) of the tool working point 221 currently on the image device coordinate system {I} Step S26). Then, the controller 20 further obtains the pre-established conversion matrix (step S28), and imports the current posture and specific coordinates into the conversion matrix to obtain the tool operating point 221 on the robot arm coordinate system {B} through conversion. Absolute position (step S30). In the embodiment, the absolute position of the tool working point 221 in the robot arm coordinate system {B} can be expressed by the following formula: P _t ^B =T _I ^B P _t ^I , where P _t ^B is the tool working point 221 in the robot arm A point on the coordinate system {B}, T _I ^B is a conversion matrix that records the correspondence between the robot arm coordinate system {B} and the imaging device coordinate system {I}, P _t ^I is the tool working point 221 in the imaging device coordinate Is a point in {I}.

It is worth mentioning that after obtaining the absolute position of the tool working point 221 in the robot arm coordinate system {B}, the controller 20 can further control the imaging device 3 to capture a two-dimensional image of the tool working point 221 (step S32 ). By capturing two-dimensional images, the image processing unit 30 can establish a vector relationship from the center point of the flange surface 21 to the tool working point 221, or directly create a complete three-dimensional image of the tool 22 (detailed later). In this way, the user of the robotic arm 2 can more clearly understand the state of the tool 22 currently set, and then decide how to set the tool working point 221 or how to operate the robotic arm 2.

Please continue to refer to FIG. 5, which is a first specific embodiment of the establishment flowchart of the present invention. As mentioned above, to accurately obtain the absolute position of the tool working point 221 on the robot arm coordinate system {B}, the imaging device coordinate system {I} and the conversion matrix must be established first. Therefore, the user of the robotic arm 2 first installs or replaces the robotic arm 2 and/or the imaging device 3 in the calibration system 1 (step S40), and only when the robotic arm 2 and/or the imaging device 3 is installed or replaced for the first time Each step shown in FIG. 5 needs to be performed to re-establish the image device coordinate system {I} and the conversion matrix. At this time, the tool 22 has not been set on the robot arm 2 yet.

After step S40, the controller 20 controls the movement of the robot arm 2 without the tool 22 on the robot arm 2, so that the flange surface 21 on the robot arm 2 enters the imaging device 3 In the standing image sensing area 33 (step S42). Next, the robot arm 2 moves based on the control of the controller 20 to make the flange surface 21 move in the image sensing area 33, and the controller 20 records the multiple pens of the robot arm 2 during the movement of the flange surface 21. Posture information (step S44).

In one embodiment, the posture information includes coordinate information on the robot arm coordinate system {B} when the flange surface 21 is located at a plurality of positions in the image sensing area 33. In another embodiment, the posture information further includes the X-axis rotation, Y-axis rotation, and Z-axis rotation of the robot arm 2 when the flange surface 21 is located at the multiple positions. However, the above are only the main embodiments of the present invention, but not limited thereto.

When the flange surface 21 moves in the image sensing area 33, the image device 3 can use the light source device 31 and the photosensitive device 32 to determine the position of the flange surface 21 in the image sensing area 33, and the controller 20 can directly Read the multi-pen coordinate information on the robot arm coordinate system {B} when the flange surface 21 moves, and thereby, the controller 20 can establish the imaging device coordinate system {I} for the imaging device 3 based on the multiple pen coordinate information (Step S46).

Moreover, in addition to the above-mentioned multiple coordinate information, the posture information recorded by the controller 20 also includes the amount of rotation of the robot arm 2 during the movement, so that the controller 20 can follow the known coordinate system of the robot arm { B}. The reconstructed image device coordinate system {I} and the recorded posture information are used to establish the conversion matrix (step S48). After step S48, the calibration system 1 already has the required parameters for the absolute position calibration of various tools 22 set on the robotic arm 2.

It is worth mentioning that, after the imaging device coordinate system {I} and the conversion matrix are established, the controller 20 can further control the imaging device 3 to capture a two-dimensional image of the flange surface 21 (step S50). After the calibration system 1 obtains the two-dimensional image of the tool working point 221 in step S32 of FIG. 3, the image processing unit 30 can use the two-dimensional image of the flange surface 21 and the two-dimensional image of the tool working point 221 To establish a vector relationship from the center point of the flange surface 21 to the tool working point 221, or directly create a complete three-dimensional image of the tool 22 (detailed later).

Please refer to FIG. 6 and FIG. 7 at the same time. FIG. 6 is a first specific embodiment of a schematic diagram of establishing a coordinate system of an imaging device of the present invention, and FIG. 7 is a second specific embodiment of a flowchart of establishing a coordinate system of the present invention. FIG. 7 is used to describe in detail in the flowchart of FIG. 5 in conjunction with FIG. 6, how to create the image device coordinate system {I} and the conversion matrix according to various information.

First, when the controller 20 needs to establish the image device coordinate system {I} and/or the conversion matrix, it first controls the robot arm 2 to move so that the flange surface 21 enters the image sensing area 33 of the image device 3 (step S60).

It is worth mentioning that in step S60, the controller 20 mainly makes the Z-axis direction of the flange surface 21 perpendicular to the plane of the image sensing area 33, that is, makes the flange surface 21 parallel to the coordinate of the image device to be established. It is the coordinate plane of {I}. Thus, in the subsequent creation process of the imaging device coordinate system {I}, the controller 20 can set the Z-axis height of the robotic arm 2 when the imaging device 3 first detects the flange surface 21 as the imaging device coordinate system {I}The height of the upper Z axis is 0.

After the flange surface 21 enters the image sensing area 33 of the imaging device 3, the controller 20 then controls the flange surface 21 to move to the first position in the image sensing area 33, while recording the posture information of the robot arm 2 (step S62 ). The posture information includes at least the coordinate information of the flange surface 21 on the robot arm coordinate system {B}. Based on the posture information, the controller 20 defines the coordinate information on the robot arm coordinate system {B} when the flange surface 21 is at the first position as the basic positioning point of the imaging device coordinate system {I} (step S64).

In the embodiment of FIG. 6, the first position is represented by coordinates A (X ₁ ^B , Y ₁ ^B , Z ₁ ^B ), where X ₁ ^B represents when the flange surface 21 is located at the first position, The X axis coordinate on the robot arm coordinate system {B}, Y ₁ ^B represents the Y axis coordinate on the robot arm coordinate system {B} when the flange surface 21 is in the first position, and Z ₁ ^B represents the flange surface 21 When in the first position, the Z-axis coordinate on the robot arm coordinate system {B}. In step S64, the controller 20 mainly defines the above-mentioned coordinate information as the basic positioning point of the image device coordinate system, that is, the coordinate A (X ₁ ^B , Y ₁ ^B , Z ₁ ^B ) is used as the image device coordinate system { I}'s origin coordinates.

Then, the controller 20 further controls the flange surface 21 to move to the second position in the image sensing area 33 without changing the height of the Z axis, and simultaneously records the posture information of the robot arm 2 (step S66). The posture information includes at least the coordinate information of the flange surface 21 on the robot arm coordinate system {B}. Based on the posture information, the controller 20 can define the X-axis or Y-axis of the coordinate system {I} of the imaging device according to the relative relationship between the first position and the second position (step S68).

In the embodiment of FIG. 6, the second position is represented by coordinates B (X ₂ ^B , Y ₂ ^B , Z ₂ ^B ), where X ₂ ^B represents that when the flange surface 21 is at the second position, The X axis coordinate on the robot arm coordinate system {B}, Y ₂ ^B represents the Y axis coordinate on the robot arm coordinate system {B} when the flange surface 21 is in the second position, and Z ₂ ^B represents the flange surface 21 When in the second position, the Z-axis coordinate on the robot arm coordinate system {B}. In this embodiment, the controller 20 establishes a virtual straight line based on the first position and the second position, and defines the direction of the virtual straight line as the Y-axis direction Y ^{I of the} coordinate system {I} of the imaging device.

Then, the controller 20 further controls the flange surface 21 to move to the third position in the image sensing area 33 without changing the height of the Z axis, and simultaneously records the posture information of the robot arm 2 (step S70). The posture information includes at least the coordinate information of the flange surface 21 on the robot arm coordinate system {B}. Based on the posture information, the controller 20 can define the coordinate plane of the image device coordinate system {I} according to the relative relationship between the first position, the second position, and the third position (step S72). The coordinate plane refers to a plane formed by the X axis and the Y axis.

In the embodiment of FIG. 6, the third position is represented by coordinates C (X ₃ ^B , Y ₃ ^B , Z ₃ ^B ), where X ₃ ^B represents that when the flange surface 21 is located at the third position, The X axis coordinate on the robot arm coordinate system {B}, Y ₃ ^B represents the Y axis coordinate on the robot arm coordinate system {B} when the flange surface 21 is in the third position, and Z ₃ ^B represents the flange surface 21 When in the third position, it is the Z axis coordinate on the robot arm coordinate system {B}. In this embodiment, the controller 20 establishes a virtual straight line based on the first position and the third position, and defines the direction of the virtual straight line as the X-axis direction X ^{I of the} coordinate system {I} of the imaging device. Thereby, the controller 20 can further form a coordinate plane according to the X-axis direction X ^I and the Y-axis direction Y ^I.

As mentioned above, the controller 20 can define the Z axis of the imaging device coordinate system {I} when the imaging device 3 detects the flange surface 21 for the first time. Therefore, after step S72, the controller 20 can The imaging device coordinate system {I} is established based on the coordinate plane and the Z-axis direction Z ^I perpendicular to the coordinate plane (step S74).

It is worth mentioning that after setting the basic positioning point, the controller 20 of the present invention may first define the X-axis direction X ^I , or first define the Y-axis direction Y ^I , which means that the Step S66 and step S70 have no execution sequence relationship, and should not be limited to those shown in FIG. 7.

；其中，T_I ^B為記錄了機器手臂座標系{B}與影像裝置座標系{I}的對應關係的轉換矩陣，轉換矩陣中的元素R_3*3、O_3*1、O_1*3及1_1*1分別指出機器手臂座標系{B}與影像裝置座標系{I}之間的旋轉、平移、透視和放大關係。在不存在透視和放大關係的情況下，控制器20只要計算出旋轉矩陣和平移矩陣，就可以成功建立所述轉換矩陣。 In one embodiment, the transformation matrix adopted by the controller 20 is a homogeneous transformation matrix (Homogeneous transformation matrix). The conversion matrix is for example as follows:

; Among them, T _I ^B is a conversion matrix that records the correspondence between the robot arm coordinate system {B} and the imaging device coordinate system {I}. The elements in the conversion matrix are R _3*3 , O _3*1 , O _1*3 And 1 _1*1 respectively indicate the rotation, translation, perspective and magnification relationship between the robot arm coordinate system {B} and the imaging device coordinate system {I}. In the absence of a perspective and magnification relationship, as long as the controller 20 calculates the rotation matrix and the translation matrix, the conversion matrix can be successfully established.

Specifically, in the posture information recorded by the controller 20 in the aforementioned steps S62, S66, and S70, in addition to recording the coordinate information of the flange surface 21 on the robot arm coordinate system {B} In addition, it also records various rotation information generated by the robot arm 2 when it is moving, such as X-axis rotation, Y-axis rotation, and Z rotation. Therefore, the controller 20 can also establish the conversion matrix according to the recorded posture information. As shown in FIG. 7, the controller 20 extracts the X-axis rotation, Y-axis rotation, and Z-axis rotation from the recorded posture information, and calculates the amount of rotation based on the X-axis rotation, Y-axis rotation, and Z-axis rotation. The rotation matrix is described (step S76).

In an embodiment, the rotation matrix (R _3*3 ) is a combination of three consecutive basic rotation matrices, and can be implemented by Euler angle, RPY angle (Roll Pitch Yaw) or standard rotation angle.

Specifically, the rotation mode described by the Euler angles is to first rotate around the Z axis, then rotate around the Y axis, and finally rotate around the Z axis again, by which the rotation matrix (R _3*3 )=R _zyz =R _{z,θ z} R _{y,θ y} R _{z,θ z} . The rotation mode described by the RPY angle is to first rotate around the Z axis, then rotate around the Y axis, and finally rotate around the X axis, by which the rotation matrix (R _3*3 )=R _zyx =R _{z,θ z} R _{y,θ y} R _{x,θ x} . The rotation mode described by the standard rotation angle is to first rotate around the X axis, then rotate around the Y axis, and finally rotate around the Z axis, whereby the rotation matrix (R _3*3 )=R _xyz = R _{x,θ x} R _{y,θ y} R _{z,θ z} .

In addition, the controller 20 further obtains the origin of the robot arm coordinate system {B} (for example (0,0,0)) and the basic positioning point of the imaging device coordinate system {I} (for example, the coordinate A(X ₁ ^B , Y ₁ ^B , Z ₁ ^B )) and calculate the translation matrix (O _3*1 ) based on the translation (step S78) (step S80). Wherein, the translation amount includes at least an X-axis translation amount, a Y-axis translation amount, and a Z-axis translation amount between the origin of the robot arm coordinate system {B} and the basic positioning point of the imaging device coordinate system {I}.

After step S80, the controller 20 can calculate the conversion matrix according to the calculated rotation matrix and translation matrix (step S82). After step S82, the controller 20 has completed to calibrate any tool 22 set on the flange surface 21 of the robot arm 2 (ie, find the tool working point 221 in the robot arm coordinate system {B} absolute position) preparation program. Accordingly, the controller 20 can execute the steps shown in FIG. 3 at any time to perform a calibration procedure on the tools set/replaced on the robot arm 2.

Please also refer to FIG. 8, which is a first specific embodiment of the schematic diagram of the conversion matrix of the present invention. As shown in Fig. 8, before the installation of the calibration system 1 is completed, the controller 20 already knows the robot arm coordinate system {B}. In the embodiment of FIG. 8, the robot arm coordinate system {B} is composed of the X axis direction X ^B , the Y axis direction Y ^B and the Z axis direction Z ^B.

Then, move the flange surface 21 of the robot arm 2 to the first position (with coordinates A (X ₁ ^B , Y ₁ ^B , Z ₁ ^B )) and the second position (with coordinates B) in the image sensing area 33 (X ₂ ^B , Y ₂ ^B , Z ₂ ^B )) and the third position (with coordinates C (X ₃ ^B , Y ₃ ^B , Z ₃ ^B )), the controller 20 can be based on the robot arm coordinate system {B } To create the image device coordinate system {I} used by the image device 3. In the embodiment of FIG. 8, the image device coordinate system {I} is composed of the X-axis direction X ^I , the Y-axis direction Y ^I, and the Z-axis direction Z ^I.

Finally, based on the known coordinate system of the robotic arm {B}, the established coordinate system of the imaging device {I}, and the posture information recorded during the movement of the robotic arm 2, the controller 20 can establish a The conversion matrix T _I ^B for converting between the arm coordinate system {B} and the imaging device coordinate system {I}.

When the translation amount of the robot arm 2 on the X axis is x, the conversion matrix is expressed as:

When the translation amount of robot arm 2 on the Y axis is y, the conversion matrix is expressed as:

When the translation amount of the robot arm 2 on the Z axis is z, the conversion matrix is expressed as:

When the rotation of the robot arm 2 on the X axis is θ _x , the conversion matrix is expressed as:

When the rotation of the robot arm 2 on the Y axis is θ _y , the conversion matrix is expressed as:

When the rotation of the robot arm 2 on the Z axis is θ _z , the conversion matrix is expressed as:

Through the conversion of the above conversion matrix, when the tool 22 has been calibrated and moved with the robot arm 2, the controller 20 can directly know the position information of the tool working point 221 on the tool 22 on the robot arm coordinate system {B}. In this way, the robot arm 2 (and the tool 22 thereon) can be controlled more accurately.

Please continue to refer to FIG. 9, which is a first specific embodiment of the two-dimensional image diagram of the present invention. In the aforementioned step S50 of FIG. 5, the imaging device 3 captures the two-dimensional image 5 of the flange surface of the flange surface 21 of the robotic arm 2 through the photosensitive device 32. In step S32 of FIG. 3, the imaging device 3 captures the two-dimensional image 4 of the tool working point 221 of the tool 22 through the photosensitive device 32. Through the two-dimensional images 4 and 5, the controller 20 can further establish the center point (not shown in the figure) of the flange surface 21 of the robot arm 2 to the work With 22 tool work points 221 vector relationship. In this way, it is more beneficial for the user of the robotic arm 2 to understand the type of the tool 22 currently used.

Specifically, in step S50 in FIG. 5, the robot arm 2 is controlled by the controller 20 to rotate, and the photosensitive device 32 captures a plurality of one-dimensional images corresponding to the flange surface 21 with different rotation angles. In an embodiment, the rotation angle may be, for example, a fixed angle (for example, 1 degree, 5 degrees, etc.), a half circle or a circle, etc., which is not limited herein. Then, the imaging device 3 receives the multiple one-dimensional images captured by the photosensitive device 32 through the image processing unit 30, and processes the multiple one-dimensional images through an algorithm to generate a two-dimensional image of the flange surface 21.

Similarly, in step S32 in FIG. 3, the robot arm 2 is controlled by the controller 20 to rotate, and the photosensitive device 32 captures a plurality of one-dimensional images corresponding to the tool working points 221 with different rotation angles. Then, the image processing unit 30 of the imaging device 3 processes the plural one-dimensional images through an algorithm, thereby generating a two-dimensional image of the tool working point 221.

In the embodiment of FIG. 9, the imaging device 3 only generates a two-dimensional image of the tool working point 221, so the calibration system 1 can only use the controller 20 to calculate the distance from the center point of the flange surface 21 to the tool working point 221 of the tool 22 Vector relationship. In other embodiments, the imaging device 3 can be controlled to generate multiple two-dimensional images of the tool 22, whereby the calibration system 1 can directly create a complete three-dimensional image of the tool 22 through the image processing unit 30 (as described below).

Please also refer to FIG. 10, which is a second specific embodiment of the two-dimensional image diagram of the present invention.

As shown in FIG. 10, in the image capturing process, the robot arm 2 can be controlled by the controller 20 to continuously rotate and change its Z-axis height. At the same time, the imaging device 3 is also controlled by the controller 20 to capture a two-dimensional image (for example, the flange surface two-dimensional image 5) when the robot arm 2 is at the first height Z1, and when the robot arm 2 is at the second height Z2 Capture a two-dimensional image (e.g. tool two-dimensional image Like 41), when the robot arm 2 is at the third height Z3, capture a two-dimensional image (for example, the tool two-dimensional image 42), ..., and when the robot arm 2 is at the nth height Zn, capture a two-dimensional image (for example, the tool Two-dimensional image 4n). Among them, the tool two-dimensional image 4n is the last two-dimensional image captured by the imaging device 3, and is a two-dimensional image used to describe the working point 221 of the tool.

In this embodiment, the imaging device 3 receives the flange surface two-dimensional image 5 and the plural tool two-dimensional images 41-4n through the image processing unit 30, and performs algorithms on these two-dimensional images 41-4n and 5 Processing to create a three-dimensional image of the tool 22 currently set on the robotic arm 2. In this way, the user can obtain information such as the specific appearance shape and size characteristics of the tool 22 through the three-dimensional image.

Please also refer to FIG. 11, which is a first specific embodiment of the 3D image creation flowchart of the present invention. In the embodiment of FIG. 11, the controller 20 controls the movement of the robotic arm 2 so that the tool 22 thereon enters the image sensing area 33. Next, the controller 20 controls the robot arm 2 to rotate, and controls the imaging device 3 to obtain multiple one-dimensional images corresponding to different rotation angles (step S90). Then, the image processing unit 30 processes the multiple one-dimensional images through an algorithm to create a two-dimensional image corresponding to a specific height (step S92).

Next, the controller 20 determines whether the image capturing process is completed (step S94), that is, whether the two-dimensional image of the flange surface 21 and the two-dimensional image of the tool working point 221 have been obtained. If the controller 20 determines that the image capturing process has not been completed, it controls the robot arm 2 to move to change the Z-axis height (step S96), and executes steps S90 and S92 again based on the changed Z-axis height to obtain the next two-dimensional image .

In one embodiment, the controller 20 can adjust the scanning height of the photosensitive device 32 so that the imaging device 3 first captures a two-dimensional image of the flange surface 21, and in the step S96, controls the robotic arm 2Raise to increase the height of the Z axis. In another embodiment, the controller 20 can adjust the scanning height of the photosensitive device 32 so that the imaging device 3 first captures the two-dimensional image of the tool working point 221, and in the step S96, the robot arm 2 is controlled to descend to reduce Z axis height.

If the controller 20 determines in step S94 that the image capturing process is completed, it means that the image processing unit 30 has obtained multiple two-dimensional images including the two-dimensional image 5 of the flange surface and the two-dimensional image 4 of the tool working point. Therefore, the image processing The unit 30 can process multiple two-dimensional images through an algorithm to create a three-dimensional image for describing the overall appearance and shape of the tool 22 (step S98). Through the three-dimensional images created by the image processing unit 30, the user of the robotic arm 2 can more clearly know the relevant information of the currently used tool 22, and then the robotic arm 2 can be more accurately set or controlled.

In contrast to the prior related art, the calibration system and calibration method of the present invention do not require manual teaching of the robotic arm, and use non-contact light sensing means, which can provide higher-precision calibration results without being affected by calibration procedures. This leads to wear of the tool itself.

Through the calibration system and calibration method of the present invention, the absolute position of one or more tool working points on the currently configured tool in the coordinate system of the robot arm can be effectively obtained, so that the robot arm can perform the tool more accurate Control to eliminate tolerances during tool manufacturing, deviations during assembly, or errors caused by tool wear. Compared with the non-contact correction method in the prior related art, there is no need to create a sample of the tool, and compared with the deviation value used in the prior related art, the correction accuracy can be further improved.

The above are only preferred specific examples of the present invention, and are not limited to the scope of the patent of the present invention. Therefore, all equivalent changes made by using the content of the present invention are included in the scope of the present invention in the same way. Bright.

S20~S32: Steps to obtain absolute position

Claims (18)

A method for calibrating a robot tool is applied to a calibration system having a robotic arm, an imaging device and a tool, wherein the imaging device creates an image sensing area within a movement range of the robotic arm, and the tool is set on the machine On a flange surface (Flange) of the arm and having a Tool Working Point (TWP), the correction method includes the following steps: a01) When the tool is not set on the robot arm, control the robot arm to move So that the flange surface enters the image sensing area, where the robot arm uses a robot arm coordinate system; a02) controls the flange surface to move in the image sensing area, and records the posture information of the robot arm at the same time, Wherein the posture information includes at least coordinate information on the robot arm coordinate system when the flange surface is located at a plurality of positions in the image sensing area, and a coordinate information of the robot arm when the flange surface is located at each position X-axis rotation, Y-axis rotation, and Z-axis rotation; a03) based on the coordinate information to establish an imaging device coordinate system used by the imaging device; a04) based on the robot arm coordinate system and the imaging device coordinate system And the posture information to establish the conversion matrix; a) when the tool is set on the robotic arm, control the movement of the robotic arm so that the tool working point enters the image sensing area; b) record a current of the robotic arm Posture, and a specific coordinate of the tool operating point on the imaging device coordinate system; c) obtaining the conversion matrix, wherein the conversion matrix records the correspondence between the robotic arm coordinate system and the imaging device coordinate system; and d ) Import the current posture and the specific coordinates into the conversion matrix to obtain an absolute position of the tool operating point on the robot arm coordinate system through conversion. The method for calibrating a robot tool according to claim 1, wherein the imaging device includes a light source device and a photosensitive device arranged in parallel to each other, the light source device emits a light source toward the photosensitive device, and the photosensitive device captures the light emitted by the light source device A light source, the imaging device uses the light source to establish the image sensing area. The method for calibrating a robot tool according to claim 2, wherein the step a02 includes the following steps: a021) controlling the flange surface to move to a first position in the image sensing area, and placing the flange surface at the The coordinate information on the coordinate system of the robot arm at the first position is defined as a basic positioning point of the coordinate system of the imaging device; a022) Control the flange surface to move to the image sensing area without changing the height of the Z axis And define an X-axis direction or a Y-axis direction of the coordinate system of the imaging device according to the relative relationship between the first position and the second position; a023) Control the flange surface without changing the Z-axis In the case of height, move to a third position in the image sensing area, and define a plane of the image device coordinate system according to the relative relationship between the first position, the second position, and the third position; wherein, The step a03 is to establish the coordinate system of the imaging device according to the coordinate plane and a Z-axis direction perpendicular to the coordinate plane. The method for calibrating a robot tool according to claim 3, wherein in the step a01, the flange surface is parallel to the coordinate plane of the coordinate system of the imaging device, and the robot arm detects the imaging device for the first time The height of the Z-axis when reaching the flange surface is set to the height where the Z-axis of the coordinate system of the imaging device is 0. The method for calibrating a robot tool according to claim 3, wherein the step a04 includes the following steps: a041) calculating a rotation matrix based on the X-axis rotation, the Y-axis rotation and the Z-axis rotation in the attitude information A042) Obtain an x-axis translation amount, a y-axis translation amount and a z-axis translation amount between an origin of the robot arm coordinate system and the basic positioning point of the imaging device coordinate system; a043) according to the x-axis The translation amount, the y-axis translation amount, and the z-axis translation amount calculate a translation matrix; and a044) calculate the conversion matrix according to the rotation matrix and the translation matrix. The method for correcting a robot tool according to claim 2, wherein the transformation matrix is a homogeneous transformation matrix (Homogeneous transformation matrix). The method for calibrating a robot tool according to claim 2, wherein after step a04, it further includes a step a05): obtaining a two-dimensional image of the flange surface from the imaging device. The method for calibrating a robot tool according to claim 7, which further includes the following steps: e) obtaining a two-dimensional image of the working point of the tool through the imaging device; and f) according to the two-dimensional image of the flange surface and The two-dimensional image of the working point of the tool establishes a vector relationship from a center point of the flange surface to the working point of the tool. The method for calibrating a robot tool according to claim 7, which further includes the following steps: g) obtaining a two-dimensional image of the tool through the imaging device; h) determining whether an image capturing process is completed; i) Change the Z-axis height of the robotic arm before the image capture process is completed, and perform step g again; and j) After the image capture process is completed, based on the two-dimensional image of the flange surface and the The plural of the two-dimensional images of the tool create a three-dimensional image of the tool. A calibration system for a robot tool includes: a robot arm having a flange surface (Flange) and using a robot arm coordinate system; a tool arranged on the flange surface and having a tool working point (Tool Working Point) Point, TWP); an imaging device that creates an image sensing area within a moving range of the robotic arm, and uses an imaging device coordinate system; and a controller, which is electrically connected to the robotic arm; wherein, in the machine When the tool is not set on the arm, the controller controls the movement of the robotic arm so that the flange surface moves in the image sensing area while recording the posture information of the robotic arm, wherein the posture information includes at least the flange surface is located at The coordinate information on the coordinate system of the robot arm at multiple positions in the image sensing area, and an X-axis rotation amount, a Y-axis rotation amount of the robot arm when the flange surface is located at each of the positions, and A Z-axis rotation amount, and the controller establishes the imaging device coordinate system based on the coordinate information, and establishes a conversion matrix according to the robot arm coordinate system, the imaging device coordinate system, and the posture information; wherein, in the robot arm When the tool is installed, the controller controls the movement of the robotic arm so that the tool operating point enters the image sensing area, and records a current posture of the robotic arm and the tool operating point on the image device coordinate system At a specific coordinate; Wherein, the controller imports the current posture and the specific coordinates into the conversion matrix to obtain an absolute position of the tool operating point on the robot arm coordinate system through conversion, wherein the conversion matrix records the robot arm coordinate system and Correspondence between the coordinate systems of the image device. The calibration system for a robot tool according to claim 10, wherein the imaging device includes: a light source device that emits a light source outward; and a photosensitive device that is arranged in parallel with the light source device to capture the light source emitted by the light source device to The image sensing area is established by the light source. The calibration system of the robot tool according to claim 11, wherein when the coordinate system of the imaging device is established, the controller first controls the flange surface to move to a first position in the image sensing area, and the method When the blue surface is at the first position, the coordinate information on the coordinate system of the robot arm is defined as a basic positioning point of the coordinate system of the imaging device; then, the controller controls the flange surface without changing the height of the Z axis Move down to a second position in the image sensing area, and define an X-axis direction or a Y-axis direction of the image device coordinate system according to the relative relationship between the first position and the second position; then, the control The device controls the flange surface to move to a third position in the image sensing area without changing the height of the Z axis, and defines the flange surface according to the relative relationship between the first position, the second position and the third position A coordinate plane of the image device coordinate system; then, the controller establishes the image device coordinate system based on the coordinate plane and a Z-axis direction perpendicular to the coordinate plane. The calibration system of the robot tool according to claim 12, wherein when the coordinate system of the imaging device is established, the controller makes the flange surface parallel to the coordinate level of the coordinate system of the imaging device When the imaging device detects the flange surface for the first time, the Z-axis height of the robotic arm coordinate system is set to the height of the Z-axis of the imaging device coordinate system as 0. The calibration system of the robot tool according to claim 12, wherein when the conversion matrix is established, the controller calculates a rotation amount of the X axis, the rotation amount of the Y axis, and the amount of the Z axis in the attitude information Rotation matrix, and then obtain an x-axis translation amount, a y-axis translation amount and a z-axis translation amount between an origin of the coordinate system of the robot arm and the basic positioning point of the coordinate system of the imaging device, according to the x-axis translation Calculate a translation matrix based on the amount, the y-axis translation amount, and the z-axis translation amount, and then calculate the conversion matrix according to the rotation matrix and the translation matrix. The correction system of the robot tool according to claim 11, wherein the transformation matrix is a homogeneous transformation matrix (Homogeneous transformation matrix). The calibration system of the robot tool according to claim 11, wherein the imaging device obtains a two-dimensional image of the flange surface when the imaging device coordinate system is established. The calibration system of the robot tool according to claim 16, wherein the imaging device obtains a two-dimensional image of the working point of the tool, and the imaging device further includes an image processing unit electrically connected to the photosensitive device, the image processing unit According to the two-dimensional image of the flange surface and the two-dimensional image of the tool operating point, a vector relationship from a center point of the flange surface to the tool operating point is established. The calibration system for a robot tool according to claim 16, wherein the imaging device obtains a plurality of two-dimensional images corresponding to different Z-axis heights of the working point of the tool, and the imaging device further includes an image electrically connected to the photosensitive device A processing unit, the image processing unit creates a three-dimensional image of the tool according to the two-dimensional image of the flange surface and the plurality of two-dimensional images of the tool.

Images