Automatic polishing system and polishing method

Abstract

The present invention provides an automated polishing system and method that enables the automation of polishing operations. [Solution] The automatic polishing system and method include the steps of: acquiring an image of a panel with a camera; acquiring the location of a defective portion present in the panel based on the acquired image; bringing a polishing robot into surface contact with the panel at the location; and operating the polishing robot while maintaining surface contact between the polishing robot and the panel.

Description

Technical Field

[0001] The present invention relates to an automatic surface polishing system and method.

Background Art

[0002] In recent years, with the rapid spread of electric vehicles, solutions for weight reduction of vehicle bodies have been actively studied. For example, the application of aluminum panels to vehicle bodies is expanding. Due to the material characteristics of aluminum, a large number of defects such as chips, unevenness, scratches, and dents occur during the manufacturing process. Such defects can be visually confirmed after the painting process and may have an adverse effect on the aesthetics of the vehicle. For this reason, manual 100% inspection and 100% polishing are performed to remove the defects. However, such a polishing process manually performed by workers is a factor that causes excessive man-hours and increases the unit cost of parts. Also, aluminum dust is known as a harmful substance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] The present invention has been devised to solve the above-described problems, and provides an automatic polishing system and method that enable automation of the polishing work. An object of the present invention is to provide an automatic polishing system and method capable of reducing man-hours and costs. An object of the present invention is to provide an automatic polishing system and method capable of absorbing manufacturing tolerances and scatter that may occur during the manufacturing process. The objectives of the present invention are not limited to those stated above, and any other objectives not mentioned will be clearly understood by a person with ordinary skill in the art to which the present invention pertains (hereinafter referred to as "an ordinary person") from the following description. [Means for solving the problem]

[0005] The features of the present invention, which achieve the objectives of the present invention described above and perform the characteristic functions of the present invention described later, are as follows. According to one embodiment of the present invention, an automatic polishing method includes the steps of: acquiring an image of a panel with an imaging device; acquiring the location of a defective portion present in the panel with a computer based on the acquired image; directing a polishing robot including a polishing tool to the location; bringing the polishing tool into surface contact with the panel at the location; and operating the polishing robot while maintaining surface contact between the polishing tool and the panel.

[0006] According to one embodiment of the present invention, an automated polishing system includes an imaging device configured to acquire three-dimensional data of an object; an articulated robot including a polishing tool configured to polish the object; and a computer configured to determine the location of defective portions present on the object based on the three-dimensional data and to control the robot so that it polishes the defective portions at those locations. The computer is configured to control each axis of the robot so that polishing is performed while maintaining pressure that is evenly distributed to the defective portions by the polishing tool. [Effects of the Invention]

[0007] According to the present invention, an automated polishing system and method are provided that enable the automation of polishing work. According to the present invention, an automated polishing system and method are provided that can reduce man-hours and costs. According to the present invention, an automated polishing system and method are provided that can absorb manufacturing tolerances and scattering that may occur during the manufacturing process. The effects of the present invention are not limited to those described above, and other effects not mentioned can be clearly seen by those skilled in the art from the following description. [Brief explanation of the drawing]

[0008] [Figure 1] This is a diagram showing the configuration of an automatic polishing system according to an embodiment of the present invention. [Figure 2] This figure shows a deep learning model of an automated polishing system according to an embodiment of the present invention. [Figure 3] This figure shows a robot for an automated polishing system according to an embodiment of the present invention. [Figure 4] This figure shows a force torque sensor for an automatic polishing system according to an embodiment of the present invention. [Figure 5] This figure shows the polishing process of a polishing tool in an automated polishing system according to an embodiment of the present invention. [Figure 6] This is a flowchart of an automated polishing method according to an embodiment of the present invention. [Figure 7] This is a flowchart of the robot's operation in an automated polishing method according to an embodiment of the present invention. [Modes for carrying out the invention]

[0009] The specific structural or functional descriptions presented in the embodiments of the present invention are illustrative only for the purpose of illustrating embodiments based on the concept of the present invention, and embodiments based on the concept of the present invention can be implemented in various forms. Furthermore, it should be understood that the embodiments described herein should not be construed as being limited to those described herein, and that all modifications, equivalents, or substitutes included are part of the technical idea and scope of the present invention.

[0010] On the other hand, while terms such as "first" and / or "second" in this invention can be used to describe various components, the components are not limited to those terms. For the purpose of distinguishing one component from another, for example, a first component may be called a second component, without departing from the scope of rights under the concept of this invention. Similarly, a second component may also be called a first component.

[0011] When one component is referred to as being "connected" to another, it should be understood that it is either directly connected to the other component, or may be connected, but there may be other components in between. On the other hand, when any component is referred to as being "directly connected" to or "in direct contact" with another component, it should be understood that there are no other components in between. Other expressions used to describe the relationship between components, such as "between" and "immediately between" or "adjacent to" and "directly adjacent to," should be interpreted similarly.

[0012] Throughout this specification, the same reference numerals represent the same component. On the other hand, the terms used herein are for the purpose of describing embodiments and do not limit the invention. In this specification, the singular form includes the plural form unless otherwise specified in the context. As used herein, “including” does not exclude the presence or addition of one or more other components, steps, actions and / or elements that are mentioned.

[0013] The present invention will be described in detail below with reference to the attached drawings. Manual polishing of materials such as aluminum, as described above, can cause an increase in man-hours and costs, and adverse effects on the health of workers. Therefore, the development of automatic polishing is an urgent situation, but it may be difficult to automate the polishing. For example, aluminum is difficult to automate the polishing due to the characteristics of the aluminum material. Specifically, in aluminum, due to the nature of the material, excessive dust is generated during the forming and manufacturing processes, and defects such as unevenness and indentations occur sporadically in non-uniform sites.

[0014] In order to automatically polish defects such as unevenness and indentations generated in the material at such non-fixed sites, a robot including a polishing tool must move to the position with the defect and polish while the polishing tool adapts to the curved surface in real time. Only by such a method can products without defects be obtained during mass production.

[0015] The method of teaching the operation of the robot can be automated. However, if the robot only moves along the pre-taught path and position, it is difficult to deal with product dispersion, robot dispersion, manufacturing process dispersion, etc. Therefore, it is difficult to ensure uniform polishing quality, and there is a high possibility of defects. In addition, calibration of the teaching work for all preset areas of the product is periodically required, resulting in extra man-hours.

[0016] The present invention provides an automatic polishing system and method that do not require teaching and enable adaptive control in which the polishing tool is in real-time surface contact with the curved surface of the polishing target.

[0017] As shown in FIG. 1, the polishing system 1 according to the present invention is configured to polish the surface of the object 2. The surface of the object 2 can include curved or bent portions. For example, the object 2 can be a panel of a vehicle body. As an example, the object 2 can be a portion including welded bending of a moving part of a vehicle body. As an example, the object 2 can be an aluminum panel. However, the object 2 polished by the polishing system 1 is not limited to this. The object 2 of the polishing system 1 can be applied not only to the vehicle body in the vehicle body process but also to objects in any field where polishing or sanding such as a painting process is required.

[0018] In particular, the polishing system 1 can polish the defective portion 4 formed on the surface of the object 2. The defective portion 4 can include irregularities, scratches, or dents formed on the surface of the object 2. Such defective portions 4 can be formed sporadically on the object 2. The polishing system 1 can adapt in real time to the surface during polishing even for the defective portion 4 in a portion with curvature or bending on the surface of the object 2 and polish the defective portion 4 without separate teaching.

[0019] In other words, the polishing system 1 is configured to determine the exact three-dimensional position of the defective portion 4 by image processing and analysis of the object 2 and perform polishing while adapting to the surface in real time at that position. For this purpose, the polishing system 1 includes an imaging device 20, a polishing device, and a computer 100. The computer 100 is configured to execute operations for the operation of the polishing system 1 and control the imaging device 20 and the polishing tool. In one embodiment, the computer 100 can execute an operation to obtain three-dimensional position information of the defective portion 4 based on the image acquired by the imaging device 20. Further, in one embodiment, the computer 100 can control the polishing tool to perform polishing on the acquired three-dimensional position information.

[0020] Computer 100 includes memory and a processor. The processor, which is hardware, can execute computer-readable code or a set of instructions stored in memory and process data. In non-limiting examples, a processor may include a central processing unit, a graphics processing unit, a multi-core processor, a multiprocessor, an ASIC (Application-Specific Integrated Circuit), or an FPGA (Field Programmable Gate Array).

[0021] Memory may store data, code, or a set of instructions that can be executed by the processor. Memory can be volatile or non-volatile. As an unspecified example, volatile memory may include DRAM (Dynamic Random Access Memory) or SRAM (Static Random Access Memory). As another unspecified example, non-volatile memory may include EEPROM (Electrically Erasable Programmable Read-Only Memory), flash memory, MRAM (Magnetic RAM), CD-ROM, or DVD-ROM.

[0022] The imaging device 20 is configured to acquire an image of object 2. In one embodiment, the imaging device 20 can acquire three-dimensional data of object 2 through a structured optical method. The imaging device 20 includes a camera and a projector. The camera is configured to acquire two-dimensional data or two-dimensional positional information of object 2 by capturing a light pattern projected onto object 2. For example, the camera may be a two-dimensional camera. The projector is configured to project a light pattern onto object 2. The imaging device 20 can construct three-dimensional data by combining depth information through the camera and projector. The computer 100 is configured to acquire positional coordinates in a three-dimensional coordinate system by matching the positional information of the two-dimensional data with the three-dimensional data through the constructed three-dimensional data. In one embodiment, the imaging device 20 may be a three-dimensional mechanical vision system.

[0023] Computer 100 can use the camera's intrinsic parameters, extrinsic parameters, and depth information to convert specific coordinates in 2D image data into 3D spatial coordinates.

[0024] The intrinsic parameters of a camera may include an intrinsic parameter matrix K based on the x-direction focal length fx, the y-direction focal length fy, the x-coordinate Cx of the principal point, and the y-coordinate Cy of the principal point. The intrinsic parameter matrix K can be determined by [Equation 1].

number

[0025] The camera's external parameters can include a rotation matrix R, a transformation vector T, and an external parameter matrix M1. The external parameter matrix M1 can be obtained by [Equation 2].

number

[0026] Computer 100 can perform calculations to convert two-dimensional position coordinates to three-dimensional spatial coordinates using internal and external parameters. Since this calculation can utilize known techniques for converting two-dimensional image coordinates to three-dimensional image coordinates, no further explanation is provided.

[0027] Referring to Figure 2, according to one embodiment of the present invention, the accuracy of identifying defective portions 4 in images acquired by the imaging device 20 can be further improved through deep learning training of the computer 100. For this purpose, the computer 100 can include a deep learning model 110. The deep learning model 110 is configured to learn an image of the object 2 containing the defective portion 4, the two-dimensional position of the defective portion 4 in the image of the object 2, and data of the defective portion 4 in the image of the object 2. Since the data of the defective portion 4 has a different range of values ​​than the data of the non-defective portion of the object 2, it can be determined that a portion with values ​​in that range contains a defective portion 4. When an image P1 of the object 2 is input to the deep learning model 110 based on the learned data, it can output the position coordinates of the defective portion 4 in the image of the object 2 and the data of the defective portion 4 in the image of the object 2 as output P2. Therefore, the computer 100 can identify the defective portion 4 with a high probability when determining the defective portion 4 in an image acquired from the imaging device 20. In other words, pixel values ​​on a 2D image can be calculated and matched with 3D data to obtain 3D data of a desired specific region, i.e., the defective part 4. The obtained 3D data can be converted into robot coordinates, as described later, and the robot 40 can move based on it. The present invention enables robot programming, which previously relied solely on teaching, to find the defective part 4 itself without teaching. In one example, the deep learning model 110 may be based on the YOLO (You Only Look Once) deep learning algorithm.

[0028] According to one embodiment of the present invention, at least a portion of the defective portion 4 before image acquisition by the imaging device 20 can undergo a pre-enhancement process. For example, at least a portion of the defective portion 4 of the object 2 can be visually displayed in advance by methods such as color processing. In another example, by artificially forming polishing marks on the visually visible defective portion 4 using an oil stone or sandpaper, the shading of the defective portion 4 with irregularities, dents, etc., can be enhanced, making it easier for the imaging device 20 to identify the defective portion 4. In the case of defective portions 4 in the object 2, if the degree of defect is weak, they may not be identified by the imaging device 20 or the computer 100, so the ability to identify the defective portion 4 can be further improved through the above process.

[0029] In one embodiment, a two-dimensional image acquired by a camera can be preprocessed by a computer 100. The computer 100 can preprocess the defective areas 4 in the two-dimensional image by Hue, Saturation, Value (HSV) color analysis. Preprocessing may include increasing the saturation in the HSV. By adjusting the saturation of the defective areas 4 in the preprocessing step, the detection rate of the defective areas 4 can be improved.

[0030] In one example, the imaging device 20 can be attached to an articulated robot 30. The computer 100 can control the movement of the articulated robot 30 and, through the control of the articulated robot 30, can control the imaging position of the imaging device 20 relative to the object 2.

[0031] The polishing system 1 includes a polishing device. In one embodiment, the polishing device may be a robot 40, which may be an articulated robot. In a non-limiting example, the polishing device may be a multi-axis controllable articulated collaborative robot. In a non-limiting example, the polishing device may be a 6-axis controllable articulated collaborative robot.

[0032] The robot 40 includes a polishing tool 50. The polishing tool 50 is configured to polish the surface of object 2 like sandpaper. The polishing tool 50 can be detachably attached to the working end of the robot 40. The polishing tool 50 can be rotated by a spindle attached to the robot 40. The polishing tool 50 can be rotated by the robot 40 while pressing it against the surface of object 2.

[0033] As shown in Figure 3, the robot 40 includes a force-torque sensor 60. In one example, the force-torque sensor 60 may be a multi-axis force-torque sensor. In a non-limiting example, the force-torque sensor 60 may be a 6-axis force-torque sensor. The force-torque sensor 60 senses the applied force, and the robot 40 can adjust the force applied to the surface of object 2 based on the sensed information.

[0034] Referring to Figure 4, according to an embodiment of the present invention, the force-torque sensor 60 may be a resistive sensor that converts deformation due to an external force into an electrical signal. In one embodiment, the force-torque sensor 60 includes a strain gauge 62. The force-torque sensor 60 can detect deformation through the strain gauge 62 as an electrical signal, i.e., a change in resistance, and provide the detected information to the robot 40.

[0035] According to an embodiment of the present invention, the force torque sensor 60 may include a plurality of strain gauges 62. The strain gauges 62 can be arranged on the force torque sensor 60 at predetermined intervals. As shown in the illustrated embodiment, the strain gauges 62 can be arranged at predetermined intervals along the circumferential direction of the force torque sensor 60. In one embodiment, the force torque sensor 60 includes four or more strain gauges 62, and each strain gauge 62 can be arranged at intervals of approximately 90° in the circumferential direction (in the figure, 62a, 62b, and 62c are shown, and 62d is omitted). However, the number of strain gauges 62 can be changed.

[0036] Specifically, the force torque sensor 60 can measure the pressure applied to the polishing tool 50 when it comes into contact with the surface of the target 2 as a change in resistance. According to the present invention, when polishing the surface of the target 2, uniform pressure control is possible by controlling each axis of the robot 40 so that the resistance values ​​measured by each strain gauge 62 become uniform. This enables automated polishing work by allowing the robot 40 to adapt to the surface of the target 2 in real time and perform polishing, regardless of the bending or curvature of the surface of the target 2.

[0037] In one embodiment, as shown in Figure 5, the movement of the robot 40 can be controlled so that the polishing tool 50 moves along a path P that surrounds the defective portion 4, with the defective portion 4 as the center. In one embodiment, the robot 40 can be controlled so that the polishing tool 50 moves along a rectangular path with the defective portion 4 as the center. For example, in the case of rough polishing, the polishing tool 50 can move along a path following a square with sides of 30 mm, and in the case of finishing, it can move along a path following a square with sides of 50 mm. Polishing uneven surfaces on a car body panel requires work over a wide area. That is, if polishing is concentrated on only one part, notches and marks may remain excessively. Therefore, the present invention can prevent the above problem by configuring the polishing tool 50 to move along a path surrounding the defective portion 4 during polishing.

[0038] The robot 40 includes a robot control device 70. The robot control device 70 is configured to receive sensing values ​​from the force-torque sensor 60. The robot control device 70 is also configured to control each axis of the robot 40 based on the sensing values ​​from the force-torque sensor 60.

[0039] The robot control device 70 is configured to communicate with the computer 100. The computer 100 is configured to transmit the three-dimensional position coordinates of the defective part 4 to the robot control device 70. The robot control unit 70 is then configured to have the robot 40 perform polishing work at the received three-dimensional position coordinates. In one embodiment, the Ethernet (registered trademark) TCP (Transmission Control Protocol) method can be used as the communication method between the computer 100 and the robot control device 70.

[0040] According to an embodiment of the present invention, the computer 100 is configured to perform calculations that relate the coordinate systems of the imaging device 20, the robot 40, and the polishing tool 50 to each other.

[0041] The computer 100 is configured to convert the coordinates of the defective portion 4 observed by the imaging device 20, i.e., the coordinate system of the imaging device 20 (xi, yi, zi), to the coordinate system of the robot 40 (xr, yr, zr). In one example, a transformation matrix R can be used to map the coordinate system of the imaging device 20 (xi, yi, zi) to the Cartesian coordinate system of the robot 40 (xr, yr, zr). In one example, the coordinate system of the robot 40 (xr, yr, zr) may be the Cartesian coordinate system of the robot 40. The Cartesian coordinate system shows the displacement of the x, y, and z axes relative to the origin of the robot 40. By calibrating the imaging device 20 and the robot 40, the distance between the three-dimensional position coordinates of the defective portion 4 measured by the imaging device 20 and the origin of the robot 40 can be calculated, thereby setting the distance the robot 40 can travel to the defective portion 4. In other words, the computer 100 can acquire the target coordinates (x,y,z) that the robot 40 must move to in the x, y, and z directions from the robot 40's origin to the defective part 4. The computer 100 can acquire the target coordinates (x,y,z) and transmit the acquired target coordinates (x,y,z) to the robot control device 70. As a result, the robot 40 can move to the target coordinates (x,y,z) representing the defective part 4 and perform polishing.

[0042] Furthermore, in one embodiment, the computer 100 is configured to transmit target coordinates (x,y,z) taking into account the rotation value of the polishing tool 50. The coordinate system (xt,yt,zt) of the polishing tool 50 is a coordinate system that moves relative to the endpoint of the polishing tool 50. Even if the movement of the robot 40 is determined through the robot 40's Cartesian coordinate system (xr,yr,zr), the rotation value of the robot 40 rotates with respect to each axis of the polishing tool 50's coordinate system (xt,yt,zt), so the amount of rotation must be calculated according to the curvature of object 2. For example, when determining the rotation direction of the coordinate system (xt,yt,zt) of the polishing tool 50, the axial direction can be aligned with the thumb of the right hand, and the direction in which the index finger and middle finger, which are extended perpendicular to the thumb, point can be set as the positive (+) direction, and the opposite direction as the negative (-) direction. Taking the above points into consideration, the computer 100 can transmit coordinates (x, y, z, Rx, Ry, Rz) to the robot control device 70 as the final target coordinates. x, y, and z can represent the linear movement amounts of the robot 40 along its three axes, and Rx, Ry, and Rz can represent the rotation amounts of the robot 40 along each axis.

[0043] Furthermore, the computer 100 is configured to calculate the curvature of the defective portion 4. The rotation value of the robot 40 must be determined according to the curvature of the object 2 so that the polishing tool 50 can make close contact with the defective portion 4 and adapt to the curvature. If polishing is performed on a part with a large curvature without reflecting the rotation correction value, surface contact adaptation control will not be performed smoothly. For this reason, the computer 100 can calculate the amount of curvature change of the object 2 by considering the normal vector in the 3D image. The normal vector can be determined when the computer 100 acquires 3D position information from the 2D image it has acquired.

[0044] The computer 100 can obtain the amount of rotation of the normal vector at the defective part 4 using Rodrigues' rotation formula. Rodrigues' rotation formula can determine the amount of rotation on each axis between two vectors. This allows the computer to calculate the curvature of the measurement point relative to an arbitrarily set reference point. However, since this amount of rotation is the amount of rotation of the imaging device 20, the computer 100 can convert it into the amount of rotation of the robot 40 by multiplying the amount of rotation of the normal vector obtained in the 3D image by a rotation matrix. Furthermore, since the amount of rotation of the robot 40 follows the amount of rotation of the coordinate system of the polishing tool 50, an additional process can be performed to convert the amount of rotation of the Cartesian coordinate system of the robot 40 to the coordinate system value of the polishing tool 50. Through such calculations, the amount of rotation of the robot 40 can be determined and the position coordinates (x,y,z,Rx,Ry,Rz) of the final target point can be transmitted to the robot control device 70.

[0045] As described above, the position of the defective part 4 obtained by image analysis is position information generated based on the coordinate system (xi, yi, zi) of the imaging device 20. When this position information of the defective part 4 is input to the coordinate system (xr, yr, zr) of the robot 40, the robot 40 can move along the robot coordinate system. By mapping the coordinate system (xi, yi, zi) of the imaging device 20 to the coordinate system (xr, yr, zr) of the robot 40, the robot 40 can be moved to the defective part 4. A transformation matrix R can be used as a method to unify two different coordinate systems in this way. The transformation matrix R is a 4x4 matrix, where the leading 3x3 matrix part (rotation transformation part) represents two different coordinate systems, and the remaining 4x1 part (translation transformation part) represents the amount of origin shift between the two coordinate systems.

number

[0046] The difference between the rotation amounts of two different coordinate systems and the displacement of the origin can be calculated using Euler's formula [Equation 4] to obtain the rotational transformation portion in the transformation matrix R.

number

[0047] Here, Ax, Ay, and Az represent the x-axis rotation value, y-axis rotation value, and z-axis rotation value, respectively. Once the amount of rotation is known, the rotation conversion unit can be calculated using [Equation 4].

[0048] However, even if the imaging device 20 and the robot 40 are set up as predetermined, the exact amount of rotation cannot be determined due to factors such as the slope of the ground at the installation site and errors in the installation position. Therefore, the present invention is configured to actually calculate the amount of rotation in order to reduce this error. That is, by grasping the movement of each axis of the robot 40 within the imaging area of ​​the imaging device 20, the displacement of the robot 40 coordinate system (xr,yr,zr) can be obtained with respect to the coordinate system (xi,yi,zi) of the imaging device 20. By inputting the error value grasped based on this into [Equation 4], the rotational transformation portion can be calculated.

[0049] The movement transformation portion can be obtained by setting three more equations. These three equations can be obtained through the coordinate system (xr,yr,zr) of the imaging device 20 and the robot 40. The coordinates of the robot 40 can be obtained through the product of the transformation matrix R and the coordinates of the imaging device 20. Coordinate information of the defective part 4 (in the coordinate system of the imaging device 20) can be obtained through the imaging device 20, and coordinate information of the robot 40 at the defective part 4 can be obtained through teaching the robot 40. Since the 3D coordinates of the robot 40 and the 3D coordinates of the imaging device 20 are known, the three unknown movement transformation portions can be calculated using the three equations, and the robot 40 can move to the defective part 4 based on the mapped coordinate system.

[0050] As shown in Figure 6, in step S600, the polishing system 1 is configured to acquire a three-dimensional image of the surface of the object 2. As described above, the imaging device 20 can acquire a two-dimensional image of the object 2.

[0051] In step S610, the position coordinates of the defective part 4 can be obtained. The computer 100 performs calculations that associate the coordinate systems of the robot 40, the imaging device 20, and the polishing tool 50, thereby determining the position coordinates to which the robot 40 can move up to the defective part 4 in the robot 40's coordinate system.

[0052] Based on the position coordinates acquired in step S620, the robot 40 can perform automatic polishing on the surface of object 2. The polishing tool 50 can adapt to the surface of object 2 via the force torque sensor 60, regardless of whether the surface of object 2 has curvature, and perform the polishing work.

[0053] Referring to Figure 7, in the automated polishing process, in step S700, the robot 40 moves to the defective portion 4 at the acquired position coordinates. If multiple defective parts 4 exist, the computer 100 can generate an optimal movement path for the robot 40. In one example, the optimal movement path can mean the shortest distance to move through all defective parts 4 when there are multiple defective parts 4 present on the object 2.

[0054] In step S710, the robot 40 checks whether the polishing tool 50 is in surface contact with the surface of the target 2. Here, surface contact means that the polishing tool 50 is in even contact with the surface of the target 2. The computer 100 can determine whether or not surface contact is occurring based on the directional resistance values ​​of the strain gauge 62 of the force torque sensor 60. If the directional resistance values ​​are uniform, the computer 100 can determine that the polishing tool 50 and the surface of the target 2 are in surface contact.

[0055] When it is determined that surface contact has been made, the computer 100 starts polishing with the robot 40 (S720). The robot control device 70 rotates the spindle to rotate the polishing tool 50 and can apply a preset pressure to the surface of the object 2 through axis control of the robot 40.

[0056] After polishing begins, the robot control device 70 is configured to collect the values ​​sensed by the force torque sensor 60 in real time (S730). The robot control device 70 can receive the resistance values ​​of each strain gauge 62 and determine whether these values ​​are uniform or not.

[0057] The robot control device 70 is configured to control each axis of the robot 40 so that the pressure from the force torque sensor 60 is uniform in each direction (S740). Uniform pressure in each direction means that the polishing tool 50 is polishing in close contact with the surface of the object 2. Therefore, according to the present invention, polishing work can be performed automatically without teaching the robot 40.

[0058] In conventional techniques, teaching was performed for each sample. With such conventional techniques, it is difficult to respond flexibly when scattering occurs on a curved surface, resulting in polishing defects. However, according to the present invention, an automated polishing system is provided that can change the robot's position along the target curved surface without teaching it.

[0059] Conventionally, after an operator identified a defective area in a polishing target divided into multiple regions, the polishing robot was configured to perform polishing work by moving along a predetermined pattern in the region where the defective area was located, according to the teaching content set by the operator. In other words, pre-teaching of the robot's path was required. Furthermore, the polishing robot was configured to move only in a predetermined pattern, without taking into account the curvature of the polishing target. On the other hand, according to the present invention, the coordinates of the defective area (i.e., the precise position) are obtained based on the acquired image of the target, and the polishing robot moves to the acquired position and can automatically polish along the curvature of the target at that position without teaching.

[0060] Due to the characteristics of press forming, aluminum has curvatured particles. The present invention provides an automated polishing system that can flexibly adhere to the surface and perform polishing regardless of the presence or absence of these particles. The polishing system and method according to the present invention can be applied to all processes that require surface polishing.

[0061] The present invention is not limited by the embodiments described above and the accompanying drawings, and it will be obvious to those with ordinary skill in the art to which the present invention pertains that various substitutions, modifications, and changes are possible without departing from the technical spirit of the present invention. [Explanation of symbols]

[0062] 1. Polishing System 2. Target 4. Defective parts 20 Imaging device 30 Articulated Robots 40 Robots 50 polishing tools 60 Force Torque Sensor 62 Strain Gauges 70 Robot control devices 100 Computers 110 Deep Learning Models P path

Claims

1. The steps include acquiring an image of the panel using an imaging device, The steps include: obtaining the location of the defective portion present in the panel by a computer based on the acquired image; The steps include directing a polishing robot, including a polishing tool, to the aforementioned position, The steps include bringing the polishing tool into surface contact with the panel at the aforementioned position, An automated polishing method characterized by comprising the step of operating the polishing robot while maintaining surface contact between the polishing tool and the panel.

2. The step of obtaining the location of the defective portion is: The steps include: identifying the defective portion in the image using the computer; The automatic polishing method according to claim 1, characterized by comprising the step of calculating the position coordinates of the defective portion identified by the computer.

3. The step of identifying the defective portion is performed by a pre-trained deep learning model of the computer. The automatic polishing method according to claim 2, characterized in that the deep learning model is configured to provide positional data of the defective portion when an image of the panel is input.

4. The step of obtaining the location of the defective portion is: The steps include acquiring two-dimensional data of an object by capturing a light pattern projected onto the panel by the imaging device, The steps include projecting the light pattern onto the target using the imaging device, The steps include: extracting depth information by the imaging device based on the deformation of the captured light pattern; The automatic polishing method according to claim 1, characterized by comprising the step of acquiring three-dimensional data from the two-dimensional data based on the internal and external parameters of the imaging device and the depth information.

5. The step of obtaining the location of the defective portion is: The automatic polishing method according to claim 1, characterized by including the step of mapping two-dimensional data of the defective portion acquired by the imaging device to three-dimensional data.

6. The steps include obtaining the normal vector of the defective portion from the three-dimensional data, A step of calculating the amount of rotation of the normal vector with respect to the defective portion, The steps include determining the amount of change in the curvature of the defective portion based on the amount of rotation of the normal vector, The automatic polishing method according to claim 5, further comprising the step of activating the polishing tool based on the determined amount of change in curvature.

7. The step of obtaining the location of the defective portion is: The automatic polishing method according to claim 1, characterized by including the step of converting the position of the defective portion acquired by the imaging device into a position relative to the polishing robot.

8. The step of operating the polishing robot is: The automatic polishing method according to claim 1, characterized by including the step of controlling each axis of the polishing robot based on the value sensed by the force torque sensor of the polishing robot.

9. The automatic polishing method according to claim 8, further comprising the step of controlling each axis of the polishing robot such that the deformations of a plurality of strain gauges of the force torque sensor are substantially equal to each other.

10. The step of operating the polishing robot is: The automatic polishing method according to claim 1, characterized in that it includes the step of rotating the polishing tool at a predetermined speed while applying pressure to the panel through the control of the polishing robot.

11. The step of operating the polishing robot is: The automatic polishing method according to claim 1, characterized by including the step of rotating the polishing tool while moving along a path surrounding the defective portion, with the defective portion as the center.

12. The automatic polishing method according to claim 1, characterized in that the panel is made of aluminum and the panel includes curvature.

13. The automatic polishing method according to claim 1, characterized in that the defective portion includes dents, scratches, or irregularities formed on the surface of the panel.

14. The automatic polishing method according to claim 1, characterized in that the imaging device is a three-dimensional machine vision camera.

15. An imaging device configured to acquire three-dimensional data of a target, A multi-joint robot including a polishing tool configured to polish the aforementioned object, Includes a computer configured to determine the location of a defective portion present in the object based on the three-dimensional data, and to control the robot so that the robot polishes the defective portion at that location, The automated polishing system is characterized in that the computer is configured to control each axis of the robot so that polishing is performed while maintaining a uniform pressure distribution of the polishing tool to the defective portion.

16. The automated polishing system according to claim 15, characterized in that the robot includes a force torque sensor, and the computer is configured to determine in real time whether the pressure is uniform based on the value sensed by the force torque sensor.

17. The robot includes a force torque sensor, The force torque sensor includes at least four strain gauges arranged at predetermined intervals. The automated polishing system according to claim 15, characterized in that the computer is configured to determine whether the pressure is uniform based on whether the deformations of each strain gauge are substantially equal to each other.

18. The automated polishing system according to claim 15, wherein the computer includes a pre-trained deep learning model, and the deep learning model is configured to identify the defective portion and provide the location of the defective portion when an image of the target is input.

19. The automatic polishing system according to claim 15, characterized in that the target is a car body panel made of aluminum.

20. The automatic polishing system according to claim 15, characterized in that the defective portion includes dents, scratches, or irregularities formed on the surface of the object, and the object includes curvature.

Images