Robot teaching device and robot teaching method

Abstract

This robot teaching device (1) comprises: a three-dimensional data generation unit (15) that measures a workpiece (W) and a marker (M) disposed on the workpiece (W) and serving as a teaching point (P) on a movement path (PH) of a robot (4), and generates three-dimensional data; a marker recognition unit (13a) that recognizes the marker (M) from the three-dimensional data of the workpiece (W) and the marker (M) generated by the three-dimensional data generation unit (15); and a teaching point calculation unit (13b) that calculates the teaching point (P) of an operation of the robot (4) on the basis of the marker (M) recognized by the marker recognition unit (13a).

Description

Robot teaching device and robot teaching method

[0001] This disclosure relates to a robot teaching device and a robot teaching method.

[0002] Conventionally, a robot teaching device that generates a movement path of a robot is known.

[0003] Japanese Unexamined Patent Application Publication No. 2020-134221 discloses a scanning path generation device that generates a scanning path of an appearance inspection robot. In Japanese Unexamined Patent Application Publication No. 2020-134221, the appearance inspection robot includes a robot arm, a camera and a lighting source attached to the robot arm. In Japanese Unexamined Patent Application Publication No. 2020-134221, while moving the lighting source by the robot arm, the reflected light irradiated from the lighting source and reflected by the inspection object is continuously photographed by the camera. Based on the image photographed by the camera, the appearance of the inspection object is inspected. Also, in Japanese Unexamined Patent Application Publication No. 2020-134221, the scanning path generation device generates a scanning path of the lighting source moved by the robot arm. Specifically, the scanning path generation device generates a scanning path of the lighting source such that the reflected light from the inspection object passes through the imaging point based on the shape data of the inspection object, a virtual imaging point for imaging the inspection object by the imaging unit, and a virtual lighting point such that the imaging point is located on the optical path of the reflected light reflected by the inspection object. Here, the shape data of the inspection object is a model of the inspection object simulated on a computer.

[0004] Japanese Unexamined Patent Application Publication No. 2020-134221

[0005] In Japanese Unexamined Patent Application Publication No. 2020-134221, in order to generate a scanning path of the appearance inspection robot, it is necessary to obtain the shape data of the inspection object. However, there are cases where the shape data of the inspection object is not publicly available and difficult to obtain. Therefore, even when the shape data of the inspection object cannot be obtained, it is desired to calculate the teaching points of the movement path of the robot. Hereinafter, the shape data of the inspection object is referred to as the three-dimensional data of the workpiece.

[0006] This disclosure is made to solve the above-mentioned problems and aims to provide a robot teaching device and robot teaching method that can calculate teaching points for the robot's movement path even when 3D data of the workpiece is unavailable.

[0007] The robot teaching device according to the first aspect of this disclosure comprises a 3D data generation unit that measures a workpiece and a marker placed on the workpiece which serves as a teaching point for the movement path of a robot performing work on the workpiece, and generates 3D data; a marker recognition unit that recognizes a marker from the 3D data of the workpiece and the marker generated by the 3D data generation unit; and a teaching point calculation unit that calculates teaching points for the robot's movements based on the marker recognized by the marker recognition unit.

[0008] The robot teaching device according to the first aspect of this disclosure includes a 3D data generation unit that measures a workpiece and markers placed on the workpiece, which serve as teaching points for the robot's movement path when performing work on the workpiece, and generates 3D data. This ensures that even if 3D data of the workpiece is unavailable, the 3D data generation unit can generate 3D data. Then, from the 3D data of the workpiece and markers generated by the 3D data generation unit, a marker recognition unit recognizes the markers, and a teaching point calculation unit calculates teaching points for the robot's movement based on the markers recognized by the marker recognition unit. As a result, even if 3D data of the workpiece is unavailable, teaching points for the robot's movement path can be calculated.

[0009] The robot teaching method according to the second aspect of this disclosure comprises: measuring a workpiece and a marker placed on the workpiece that serves as a teaching point for the robot's movement path when performing work on the workpiece, and generating three-dimensional data; recognizing the marker from the generated three-dimensional data of the workpiece and the marker; and calculating teaching points for the robot's movements based on the recognized marker.

[0010] The robot teaching method according to the second aspect of this disclosure, as described above, comprises measuring a workpiece and markers placed on the workpiece that serve as teaching points for the robot's movement path when performing an operation on the workpiece, and generating three-dimensional data. This ensures that three-dimensional data can be generated even if three-dimensional data of the workpiece is unavailable. Then, the markers are recognized from the generated three-dimensional data of the workpiece and markers, and the teaching points for the robot's movement are calculated based on the recognized markers. As a result, a robot teaching method is provided that can calculate the teaching points for the robot's movement path even if three-dimensional data of the workpiece is unavailable.

[0011] According to this disclosure, as described above, teaching points for the robot's movement path can be calculated even when 3D data of the workpiece is unavailable.

[0012] This figure shows an appearance inspection system according to one embodiment. This figure shows a marker placed on a workpiece according to one embodiment. This figure shows teaching points calculated by the teaching point calculation unit of a robot teaching device according to one embodiment. This figure shows the illumination area of ​​the illumination light shining on the workpiece, simulated by the simulation unit of a robot teaching device according to one embodiment. This figure shows an image of the workpiece W captured by the imaging unit, simulated by the simulation unit of a robot teaching device according to one embodiment. This figure shows a band-shaped illumination area of ​​the workpiece with no gaps in the illumination light, simulated by the simulation unit of a robot teaching device according to one embodiment. This figure shows a band-shaped illumination area of ​​the workpiece with gaps in the illumination light, simulated by the simulation unit of a robot teaching device according to one embodiment. This figure shows a band-shaped illumination area displayed on the display unit of a robot teaching device according to one embodiment. This figure shows the illumination light shining on the actual workpiece as displayed on the display unit of a robot teaching device according to one embodiment. This is a flowchart of a robot teaching method according to one embodiment. This figure shows a marker placed on a workpiece according to a modified example.

[0013] The embodiments of this disclosure will be described below with reference to the drawings.

[0014] (Configuration of the Visual Inspection System) The configuration of the visual inspection system 100 will now be described. As shown in Figure 1, the visual inspection system 100 is a system for inspecting the appearance of a workpiece W to be inspected. For example, the workpiece W has a curved surface. The visual inspection system 100 includes a robot teaching device 1, a robot 4 which includes a real imaging unit 2 and a real illumination unit 3 that irradiates the workpiece W with illumination light. The robot 4 performs a visual inspection of the workpiece W by moving the real imaging unit 2 and the real illumination unit 3 relative to the workpiece W. The real imaging unit 2 and the real illumination unit 3 refer to the real imaging unit 2 and illumination unit 3 that actually exist, unlike the model M1 of the imaging unit 2 and the model M2 of the illumination unit 3 which will be described later. Note that the model M2 of the illumination unit 3 is an example of a virtual illumination unit.

[0015] The robot teaching device 1 is a device for offline teaching of the movement path PH shown in Figure 3 for a robot 4 to perform visual inspection of a workpiece W. The robot teaching device 1 performs offline teaching, teaching the operation of the robot 4 on a display screen through simulation without using an actual machine. The robot teaching device 1 is, for example, a personal computer. The robot teaching device 1 includes a display unit 11, an operation unit 12, a processing unit 13, a storage unit 14, and a 3D data generation unit 15. The display unit 11 includes a monitor such as an LCD monitor and displays a screen. The operation unit 12 includes input devices such as a mouse and a keyboard and accepts user input operations. The processing unit 13 includes a processor and performs various processes in the robot teaching device 1. The storage unit 14 includes non-volatile memory and stores the model M1 of the imaging unit 2, the model M2 of the illumination unit 3, the model M3 of the robot 4, and the model M4 of the workpiece W, which are used in the simulation.

[0016] The 3D data generation unit 15 is, for example, a 3D scanner. A 3D scanner generates 3D data by measuring countless points on the surface of an object. Examples of 3D scanners include optical scanners and laser scanners. An optical scanner measures the shape of an object and generates 3D data by irradiating the surface of an object with light and detecting the reflected or transmitted light. Similarly, a laser scanner measures the shape of an object and generates 3D data by irradiating the surface of an object with a laser and detecting the reflected or transmitted laser. The 3D data generation unit 15 may be a smartphone application or a handheld device held by the user. The 3D data generation unit 15 may also be attached to the robot 4.

[0017] The imaging unit 2, the illumination unit 3, and the robot 4 are a device that actually performs a visual inspection of a workpiece W based on the teaching result of the robot's movement path PH by the robot teaching device 1. The imaging unit 2 is a camera that images the workpiece W. The illumination unit 3 irradiates the workpiece W with illumination light. The robot 4 performs a visual inspection of the workpiece W by moving the imaging unit 2 and the illumination unit 3. The robot 4 is a vertical articulated robot and includes an arm 41. The arm 41 has multiple joints. The arm 41 also integrally holds the imaging unit 2 and the illumination unit 3 at its tip. The robot 4 moves the imaging unit 2 and the illumination unit 3 relative to the workpiece W by driving the multiple joints of the arm 41. The robot 4 also has a control unit 42 that controls the overall operation of the robot 4.

[0018] In the visual inspection of the workpiece W, the workpiece W is illuminated by the illumination unit 3, while the surface of the workpiece W is imaged by the imaging unit 2. Additionally, the imaging unit 2 and illumination unit 3 are moved relative to the workpiece W by the robot 4, while the surface of the workpiece W is imaged by the imaging unit 2. Since it is impossible to know where abnormalities such as foreign objects, scratches, and dents on the surface of the workpiece W are located, the workpiece W is generally imaged multiple times to ensure that its entire surface is covered. Based on the image results of the workpiece W taken by the imaging unit 2, it is then inspected to determine whether or not abnormalities such as foreign objects, scratches, and dents are present on the surface of the workpiece W.

[0019] (Offline Teaching) The processing unit 13 of the robot teaching device 1 performs offline teaching by simulation based on the model M1 of the imaging unit 2, the model M2 of the illumination unit 3, the model M3 of the robot 4, and the model M4 of the workpiece W stored in the memory unit 14. The processing unit 13 performs offline teaching by simulation, for example, by displaying the workpiece W on the display unit 11. The processing unit 13 includes a marker recognition unit 13a, a teaching point calculation unit 13b, a simulation unit 13c, a display processing unit 13d, and an output unit 13e. The marker recognition unit 13a, the teaching point calculation unit 13b, the simulation unit 13c, the display processing unit 13d, and the output unit 13e are, for example, functional blocks that function in software.

[0020] First, the 3D data generation unit 15 will be described. In this embodiment, the 3D data generation unit 15 measures the workpiece W and the marker M placed on the workpiece W, which serves as a teaching point P for the movement path PH of the robot 4 that performs work on the workpiece W, and generates 3D data. For example, as shown in Figure 2, a 3D object Ma, which serves as the marker M, is placed on the surface of the workpiece W in advance by the user. The 3D object Ma refers to an object that occupies a part of a 3D space enclosed by several planes and curved surfaces. The 3D object Ma has, for example, a cylindrical shape. However, the shape of the 3D object Ma is not limited to a cylindrical shape. Also, the 3D object Ma is attached to the workpiece W by, for example, adhesive tape. The 3D data generation unit 15 measures both the workpiece W and the marker M placed on the workpiece W and generates 3D data for both the workpiece W and the marker M. In this 3D data, the workpiece W and the marker M are not distinguished, and this 3D data is 3D data of an object in which the workpiece W and the marker M are integrated. The 3D data generated by the 3D data generation unit 15 is input to the sign recognition unit 13a of the processing unit 13.

[0021] In this embodiment, the sign recognition unit 13a recognizes the sign M from the three-dimensional data of the workpiece W and the sign M generated by the three-dimensional data generation unit 15. Specifically, the sign recognition unit 13a recognizes the sign M by segmenting the sign M from the three-dimensional data generated by the three-dimensional data generation unit 15. For example, the sign recognition unit 13a determines whether each point of the three-dimensional data belongs to the workpiece W or the sign M. For this determination, for example, image processing technology may be used. In the image processing technology, it is determined whether each point of the three-dimensional data belongs to the workpiece W or the sign M based on images of the workpiece W and the sign M that have been stored in advance. Alternatively, it may be determined whether each point of the three-dimensional data belongs to the workpiece W or the sign M using a pre-trained model in which the characteristics of the workpiece W and the sign M have been learned in advance. Furthermore, among the points of the three-dimensional data, the points that belong to the workpiece W are stored in the storage unit 14 as the model M4 of the workpiece W.

[0022] In this embodiment, the teaching point calculation unit 13b calculates teaching points P for the robot 4's movements based on the signs M recognized by the sign recognition unit 13a. For example, the teaching point calculation unit 13b calculates teaching points P for the robot 4's movements such that multiple signs M recognized by the sign recognition unit 13a become teaching points P for the robot 4's movement path PH. In Figure 3, for example, the robot 4's movement path PH is set along multiple signs M arranged in multiple rows along the horizontal direction of the paper. In Figure 3, multiple movement paths PH are generated along the horizontal direction of the paper to correspond to the positions of the multiple rows of signs M. The teaching point calculation unit 13b creates normal vectors V that are substantially perpendicular to the surface of the workpiece W at the locations through which the multiple movement paths PH pass. The tip of the normal vector V is the teaching point P. As a result, the teaching point calculation unit 13b obtains multiple teaching points P along the curved surface of the workpiece W. In Figure 3, the normal vector V is shown only for a portion of the movement path PH, but normal vector V is created for all of the multiple movement paths PH. The spacing of the normal vector V on the movement path PH is set so that the spacing of the teaching points P is a predetermined value. Alternatively, the spacing of the normal vector V may be set to a predetermined initial value. Furthermore, instead of using the tip of the normal vector V as the teaching point P as described above, the coordinates of the sign M recognized by the sign recognition unit 13a may be used as the teaching point P.

[0023] Furthermore, the teaching point calculation unit 13b also creates a connection path PHa in which the robot 4 moves, connecting multiple movement paths PH. In Figure 3, the connection path PHa is represented by a dotted line. Also, when the robot 4 is moving along the connection path PHa, no inspection is performed by the imaging unit 2 and the illumination unit 3.

[0024] In this embodiment, the simulation unit 13c performs a simulation in which illumination light is irradiated onto the workpiece W while moving a virtual illumination unit 3 along a plurality of teaching points P calculated by the teaching point calculation unit 13b. For example, as shown in Figure 4, the simulation unit 13c acquires the state of reflection of the illumination light on the workpiece W through simulation. The simulation unit 13c acquires the state of reflection of the illumination light on the workpiece W through simulation using a model M1 of the imaging unit 2, a model M2 of the illumination unit 3, a model M3 of the robot 4, and a model M4 of the workpiece W. Also, as shown in Figure 5, the simulation unit 13c acquires the irradiation area A of the virtual illumination unit 3 in the captured image IM of the workpiece W captured by the virtual imaging unit 2 as the state of reflection of the illumination light through simulation. In this embodiment, the irradiation area A corresponds to the inspection area of ​​the workpiece W.

[0025] Specifically, the simulation unit 13c acquires an image IM of the workpiece W, including the reflection of the illumination unit 3 when the robot 4 is in a predetermined posture. This image IM includes an illumination region A where the amount of reflected illumination light is large and the brightness is high. The illumination region A can also be described as the part in the image IM where the illumination unit 3 is reflected. As shown in Figure 6, the simulation unit 13c acquires the illumination region A for all of the multiple movement paths PH generated by the teaching point calculation unit 13b. The illumination region A has, for example, a band shape along the left-right direction of the paper. The simulation unit 13c also acquires multiple band-shaped illumination regions A corresponding to the multiple movement paths PH. In the example shown in Figure 6, a state is shown where there are no gaps C between the multiple band-shaped illumination regions A. If the user does not properly position the marker M, gaps C will occur between the band-shaped illumination regions A, as shown in Figure 7. In this case, inspection is not performed in the gaps C, so the movement path PH is inappropriate.

[0026] In this embodiment, as shown in Figure 8, the display processing unit 13d acquires an image of the irradiated area A when illumination light is shone on the workpiece W from the simulation results of the simulation unit 13c, and displays it on the display unit 11 by superimposing it on the image of the workpiece W. Specifically, the display processing unit 13d acquires multiple strip-shaped irradiated areas A when illumination light is shone on the workpiece W while moving the virtual illumination unit 3 along the calculated teaching point P, and displays the acquired images of the multiple strip-shaped irradiated areas A on the display unit 11 by superimposing them on the image of the workpiece W. In Figure 8, multiple strip-shaped irradiated areas A are shown in a state where there are no gaps C between the strip-shaped irradiated areas A.

[0027] The output unit 13e outputs the teaching point P calculated by the teaching point calculation unit 13b to the control unit 42 of the actual robot 4. As a result, the robot 4 operates based on the calculated teaching point P. At this time, illumination light is shone onto the workpiece W from the actual illumination unit 3, and the workpiece W is imaged by the imaging unit 2. As shown in Figure 9, the image of the captured workpiece W is displayed on the display unit 11. For example, the image of the illumination area A shown in Figure 9 changes in real time in conjunction with the operation of the robot 4. Alternatively, the illumination area A acquired by the simulation shown in Figure 8 and the illumination light shone from the actual illumination unit 3 shown in Figure 9 may be displayed adjacent to each other on the display unit 11 at the same time.

[0028] (Models of imaging unit, illumination unit, robot, and workpiece) The storage unit 14 stores models M1 of multiple imaging units 2. The simulation unit 13c acquires an image IM of the workpiece W captured by the imaging unit 2 according to the imaging unit 2 of the model selected from the multiple imaging unit 2 models M1 stored in the storage unit 14 based on user input using the operation unit 12. The model M1 of the imaging unit 2 includes information such as the dimensions of the imaging unit 2, the shape of the imaging unit 2, the field of view of the imaging unit 2, the working distance representing the distance from the lens of the imaging unit 2 to the focal position, and the type of image sensor of the imaging unit 2. Specifically, the models M1 of multiple imaging units 2 include models of line-type cameras and models of area-type cameras. The model of a line-type camera is a model of an imaging unit 2 that has a line-shaped image sensor and images the workpiece W in a line. In visual inspection of the workpiece W using a line-type camera, for example, scan imaging is performed in which the workpiece W is continuously imaged while the imaging unit 2 is moved relative to the workpiece W. Furthermore, the area-type camera model is a model of an imaging unit 2 that has a two-dimensional image sensor and images a workpiece W in a predetermined two-dimensional area. In the visual inspection of a workpiece W using an area-type camera, for example, intermittent imaging is performed in which the workpiece W is intermittently imaged while the imaging unit 2 is moved relative to the workpiece W.

[0029] Furthermore, the memory unit 14 stores models M2 of multiple lighting units 3. The processing unit 13 obtains the state of reflection of illumination light by simulation according to the lighting unit 3 of the model selected based on the user's input operation using the operation unit 12 from among the multiple lighting unit 3 models M2 stored in the memory unit 14. The lighting unit 3 model M2 includes information such as the dimensions of the lighting unit 3, the shape of the lighting unit 3, the type of light source of the lighting unit 3, the arrangement of the light sources of the lighting unit 3, and the color of the illumination light. For example, the multiple lighting unit 3 models M2 include a linear lighting model and a pattern lighting model. A linear lighting model is a model of a lighting unit 3 in which the light sources are arranged in a line and illuminate in a line. Line lighting is used, for example, in combination with a linear camera. A pattern lighting model is a model of a lighting unit 3 in which the light sources are arranged two-dimensionally and illuminate in a predetermined pattern such as a grid pattern. Pattern lighting is used, for example, in combination with an area camera.

[0030] Furthermore, the storage unit 14 stores the model M3 of the robot 4 and the model M4 of the workpiece W. The model M3 of the robot 4 includes information such as the dimensions and shape of the robot 4. The model M4 of the workpiece W is, as described above, the points belonging to the workpiece W from among the points of the 3D data measured by the 3D data generation unit 15. The model M4 of the workpiece W may also be formed by measuring only the workpiece W before the marker M is placed by the 3D data generation unit 15. The model M4 of the workpiece W includes information such as the dimensions and shape of the workpiece W. In addition, the model M4 of the workpiece W may also include information such as the surface color and reflectance of the workpiece W. Information such as the surface color and reflectance of the workpiece W may be input by the user.

[0031] (Robot Teaching Method) Referring to Figure 10, the flow of offline teaching using the robot teaching device 1 of this embodiment will be explained based on a flowchart. Note that offline teaching is just one example of a robot teaching method.

[0032] First, in step S1, the 3D data generation unit 15 measures the workpiece W and the marker M placed on the workpiece W, which will be a teaching point P for the movement path PH of the robot 4 that performs work on the workpiece W, and generates 3D data. For example, if the 3D data generation unit 15 is a smartphone application, the user holds the smartphone and takes images of the workpiece W and the marker M with the smartphone's camera. The 3D data generation unit 15, as a smartphone application, generates 3D data of the workpiece W and the marker M.

[0033] Next, in step S2, the sign recognition unit 13a recognizes the sign M from the generated 3D data of the workpiece W and the sign M. As described above, the sign recognition unit 13a recognizes the sign M using image processing technology and a learning model.

[0034] Next, in step S3, the teaching point calculation unit 13b calculates teaching points P for the robot 4's movements based on the recognized markers M. For example, if multiple markers M are arranged in multiple rows, the teaching point calculation unit 13b calculates teaching points P for multiple movement paths PH corresponding to the positions of the multiple rows of markers M.

[0035] Next, in step S4, the simulation unit 13c performs a simulation in which it irradiates the workpiece W with illumination light while moving the virtual illumination unit 3 along a plurality of teaching points P, based on the teaching points P calculated by the teaching point calculation unit 13b.

[0036] Next, in step S5, the display processing unit 13d acquires images of multiple strip-shaped irradiation areas A from the simulation results of the simulation unit 13c, superimposes them onto the image of the workpiece W, and displays them on the display unit 11.

[0037] Furthermore, proceeding from step S3 to step S6, the output unit 13e outputs the teaching point P calculated by the teaching point calculation unit 13b to the control unit 42 of the actual robot 4. As a result, the robot 4 operates based on the calculated teaching point P, illumination light is shone onto the workpiece W from the actual illumination unit 3, and the workpiece W is imaged by the actual imaging unit 2. Then, in step S7, the image of the workpiece W illuminated by the captured illumination light is displayed on the display unit 11.

[0038] The operations in steps S1 to S7 are repeated by the user moving the marker M or adding new marker M until the inspection range is appropriate.

[0039] (Effects of this embodiment)

[0040] The robot teaching device 1 includes a 3D data generation unit 15 that measures a workpiece W and a marker M placed on the workpiece W, which will serve as a teaching point P for the movement path PH of the robot 4 performing work on the workpiece W, and generates 3D data. This ensures that even if 3D data of the workpiece W is unavailable, the 3D data generation unit 15 can generate the 3D data. Then, from the 3D data of the workpiece W and marker M generated by the 3D data generation unit 15, the marker recognition unit 13a recognizes the marker M, and the teaching point calculation unit 13b calculates the teaching points P for the robot 4's movements based on the marker M recognized by the marker recognition unit 13a. As a result, even if 3D data of the workpiece W is unavailable, the teaching points P for the robot 4's movement path PH can be calculated. Furthermore, since the teaching points P can be calculated simply by placing the marker M on the workpiece W without the user having to input the movement path PH into the robot teaching device 1, the user's work in obtaining the teaching points P is relatively easy.

[0041] The sign recognition unit 13a recognizes the sign M by segmenting the sign M from the 3D data generated by the 3D data generation unit 15. This allows for easy recognition of the sign M by segmenting it from the 3D data.

[0042] The robot teaching device 1 includes a simulation unit 13c that executes a simulation to irradiate the workpiece W with illumination light while moving a virtual illumination unit 3 along a plurality of teaching points P based on the teaching points P calculated by the teaching point calculation unit 13b, and a display processing unit 13d that acquires an image of the irradiation area A when the workpiece W is irradiated with illumination light from the simulation result of the simulation unit 13c, superimposes it on the image of the workpiece W, and displays it on the display unit 11. Thereby, the user can visually confirm whether the position of the label M previously arranged on the workpiece W, that is, the teaching point P, is appropriate by visually recognizing the image of the irradiation area A displayed on the display unit 11. When the teaching point P is inappropriate, the user can change the position of the label M arranged on the workpiece W so that an appropriate teaching point P is calculated. When the position of the label M is changed, the three-dimensional data generation unit 15 measures the workpiece W and the label M arranged on the workpiece W again.

[0043] The display processing unit 13d acquires a plurality of strip-shaped irradiation areas A when the workpiece W is irradiated with illumination light while moving the virtual illumination unit 3 along the calculated teaching point P, and superimposes the images of the acquired plurality of strip-shaped irradiation areas A on the image of the workpiece W and displays them on the display unit 11. Thereby, if there is a gap C between the irradiation areas A, it means that there is an area where the inspection is not performed. The user can easily confirm whether there is a leak in the inspection area by visually recognizing the image of the irradiation area A displayed on the display unit 11.

[0044] The robot 4 includes a real imaging unit 2 as a working unit that performs work on the workpiece W and a real illumination unit 3 that irradiates the workpiece W with illumination light. The real imaging unit 2 and the real illumination unit 3 are relatively moved with respect to the workpiece W to perform an appearance inspection of the workpiece W, and the irradiation area A corresponds to the inspection area of the workpiece W. Thereby, even when the three-dimensional data of the workpiece W cannot be obtained, the robot teaching device 1 can calculate the teaching point P of the movement path PH of the robot 4 for performing the appearance inspection of the workpiece W.

[0045] The robot teaching device 1 includes an output unit 13e that outputs the teaching point P calculated by the teaching point calculation unit 13b to the control unit 42 of the actual robot 4. As a result, the actual robot 4 operates based on the calculated teaching point P. The user can easily visually confirm whether the posture of the robot 4 is appropriate by visually recognizing the actually operating robot 4. Further, when the actual lighting unit 3 is attached to the actual robot 4, the user can easily visually confirm whether the irradiation light appropriately hits the workpiece W by visually recognizing the irradiation light irradiated from the actual lighting unit 3 to the workpiece W.

[0046] The marker M includes a three-dimensional object Ma that is previously arranged on the surface of the workpiece W by the user. As a result, since the three-dimensional object Ma is three-dimensional, it is possible to easily distinguish the workpiece W and the three-dimensional object Ma in the three-dimensional data as compared with a two-dimensional marker.

[0047] (Modification) It should be considered that the embodiments disclosed this time are illustrative in all respects and not restrictive. The scope of the present disclosure is shown by the claims rather than the description of the above embodiments, and further includes all changes (modifications) within the meaning and scope equivalent to the claims.

[0048] In the above embodiment, an example where the robot is a vertically articulated robot is shown, but the present disclosure is not limited to this. In the present disclosure, the robot may be a horizontally articulated robot or the like other than a vertically articulated robot.

[0049] Further, in the above embodiment, an example where the robot moves the imaging unit and the lighting unit is shown, but the present disclosure is not limited to this. In the present disclosure, the robot may move the workpiece.

[0050] Further, in the above embodiment, an example where the marker recognition unit, the teaching point calculation unit, the simulation unit, the display processing unit, and the output unit are included in one processing unit is shown, but the present disclosure is not limited to this. In the present disclosure, the marker recognition unit, the teaching point calculation unit, the simulation unit, the display processing unit, and the output unit may be distributed among a plurality of processing units.

[0051] Furthermore, while the above embodiment shows an example in which the robot teaching device includes a simulation unit that performs a simulation of irradiating a workpiece with illumination light while moving a virtual illumination unit along a plurality of teaching points based on teaching points calculated by the teaching point calculation unit, the disclosure is not limited to this. In this disclosure, the robot teaching device may not include a simulation unit and may only calculate teaching points for the robot's movement based on signs recognized by the sign recognition unit.

[0052] Furthermore, while the above embodiment shows an example in which the display processing unit superimposes the image of the irradiation area obtained from the simulation onto the image of the workpiece and displays it on the display unit, the present disclosure is not limited to this. In the present disclosure, the display processing unit may display the image of the irradiation area obtained from the simulation and the image of the workpiece side by side on the display unit.

[0053] Furthermore, while the above embodiment shows an example in which images of multiple strip-shaped illumination regions are superimposed on the workpiece image and displayed on the display unit, the present disclosure is not limited thereto. In the present disclosure, when there is only one movement path, an image of one strip-shaped illumination region is superimposed on the workpiece image and displayed on the display unit.

[0054] Furthermore, while the above embodiment shows an example of a robot performing a visual inspection of a workpiece, including a real imaging unit as a work unit that performs work on the workpiece and a real illumination unit that irradiates the workpiece with illumination light, this disclosure is not limited to this. In this disclosure, the robot may perform tasks other than visual inspection. In this case, the work unit may be at least one of the following, other than the imaging unit and illumination unit: a three-dimensional shape generation unit, a distance measuring sensor, a coating unit, a pasting unit, a spraying unit, a welding unit, a sewing unit, an ultrasonic flaw detection unit, an eddy current flaw detection unit, a tapping inspection unit, a peeling laser irradiation unit, and a hardening UV irradiation unit. This makes it possible to calculate teaching points for the robot's movement path for various work units as described above, even when three-dimensional data of the workpiece is not available.

[0055] Furthermore, the above embodiment shows an example in which both the image of the irradiation area obtained from the simulation is superimposed on the image of the workpiece and displayed on the display unit, and the teaching points calculated by the teaching point calculation unit are output to the control unit of the actual robot, but the disclosure is not limited to this. In the disclosure, only one of the above may be performed.

[0056] Furthermore, while the above embodiment shows an example where the marker is a three-dimensional object pre-placed on the surface of the workpiece by the user, the disclosure is not limited to this. In this disclosure, as shown in Figure 11, the marker M may be a tape Mb attached to the workpiece W. In Figure 11, as an example, a plurality of circular tapes Mb are attached to the workpiece W. This simplifies the placement of the marker M, as the user only needs to attach the tape Mb to the workpiece W.

[0057] Furthermore, although the above embodiments illustrate an example of applying the present disclosure to a workpiece having a curved surface, the present disclosure is not limited thereto. The present disclosure may also be applied to workpieces that do not have a curved surface, such as flat plates.

[0058] The functions of the elements disclosed herein can be performed using circuits or processing circuits, including general-purpose processors, dedicated processors, integrated circuits, ASICs (Application Specific Integrated Circuits), conventional circuits, and / or combinations thereof, configured or programmed to perform the disclosed functions. A processor is considered a processing circuit or circuit because it includes transistors and other circuits. In this disclosure, a circuit, unit, or means is hardware that performs the enumerated functions, or hardware programmed to perform the enumerated functions. The hardware may be hardware disclosed herein, or other known hardware that is programmed or configured to perform the enumerated functions. If the hardware is a processor, which is considered a type of circuit, then the circuit, means, or unit is a combination of hardware and software, and the software is used to configure the hardware and / or the processor.

[0059] [Embodiments] The exemplary embodiments described above will be understood by those skilled in the art to be specific examples of the following embodiments.

[0060] (Aspect 1) A robot teaching device comprising: a 3D data generation unit that measures a workpiece and a marker placed on the workpiece which serves as a teaching point for the movement path of a robot that performs work on the workpiece and generates 3D data; a marker recognition unit that recognizes the marker from the 3D data of the workpiece and the marker generated by the 3D data generation unit; and a teaching point calculation unit that calculates teaching points for the robot's movements based on the marker recognized by the marker recognition unit.

[0061] (Aspect 2) The robot teaching device according to aspect 1, wherein the sign recognition unit recognizes the sign by segmenting the sign from the three-dimensional data generated by the three-dimensional data generation unit.

[0062] (Aspect 3) The robot teaching device according to aspect 2, comprising: a simulation unit that performs a simulation of irradiating the workpiece with illumination light while moving a virtual illumination unit along a plurality of teaching points based on the teaching points calculated by the teaching point calculation unit; and a display processing unit that acquires an image of the illumination area when illumination light is irradiated onto the workpiece from the simulation results of the simulation unit and displays it on a display unit by superimposing it on the image of the workpiece.

[0063] (Aspect 4) The robot teaching device according to aspect 3, wherein the display processing unit acquires a plurality of strip-shaped irradiation areas when the workpiece is irradiated with illumination light while moving the virtual illumination unit along the calculated teaching points, and displays the acquired images of the plurality of strip-shaped irradiation areas superimposed on the image of the workpiece on the display unit.

[0064] (Aspect 5) The robot teaching device according to aspect 4, wherein the robot includes a real imaging unit as a work unit that performs work on the workpiece and a real illumination unit that irradiates illumination light onto the workpiece, and the real imaging unit and the real illumination unit are moved relative to the workpiece to perform an external inspection of the workpiece, and the illumination area corresponds to the inspection area of ​​the workpiece.

[0065] (Aspect 6) The robot teaching device according to any one of aspects 1 to 5, wherein the robot includes at least one of the following as a work unit for performing work on the workpiece: a three-dimensional shape measuring unit, a distance measuring sensor, a coating unit, a pasting unit, a spraying unit, a welding unit, a sewing unit, an ultrasonic flaw detection unit, an eddy current flaw detection unit, a tapping inspection unit, a peeling laser irradiation unit, and a hardening UV irradiation unit.

[0066] (Aspect 7) A robot teaching device according to any one of aspects 1 to 6, further comprising an output unit that outputs the teaching points calculated by the teaching point calculation unit to the control unit of an actual robot.

[0067] (Aspect 8) The robot teaching device according to any one of aspects 1 to 7, wherein the sign includes a three-dimensional object or seal that has been previously placed on the surface of the workpiece by the user.

[0068] (Aspect 9) A robot teaching method comprising: measuring a workpiece and a marker placed on the workpiece which serves as a teaching point for the movement path of a robot that performs work on the workpiece, and generating three-dimensional data; recognizing the marker from the generated three-dimensional data of the workpiece and the marker; and calculating teaching points for the robot's movements based on the recognized marker.

Claims

1. A robot teaching device comprising: a 3D data generation unit that measures a workpiece and a marker placed on the workpiece which serves as a teaching point for the movement path of a robot performing work on the workpiece and generates 3D data; a marker recognition unit that recognizes the marker from the 3D data of the workpiece and the marker generated by the 3D data generation unit; and a teaching point calculation unit that calculates teaching points for the robot's movements based on the marker recognized by the marker recognition unit.

2. The robot teaching device according to claim 1, wherein the sign recognition unit recognizes the sign by segmenting the sign from the three-dimensional data generated by the three-dimensional data generation unit.

3. The robot teaching device according to claim 2, comprising: a simulation unit that performs a simulation of irradiating the workpiece with illumination light while moving a virtual illumination unit along a plurality of teaching points based on the teaching points calculated by the teaching point calculation unit; and a display processing unit that acquires an image of the illumination area when illumination light is irradiated onto the workpiece from the simulation results of the simulation unit and displays it on a display unit superimposed on the image of the workpiece.

4. The robot teaching device according to claim 3, wherein the display processing unit acquires a plurality of strip-shaped irradiation areas when the workpiece is irradiated with illumination light while moving the virtual illumination unit along the calculated teaching points, and displays the acquired images of the plurality of strip-shaped irradiation areas superimposed on the image of the workpiece on the display unit.

5. The robot teaching device according to claim 4, wherein the robot includes a real imaging unit as a work unit for performing work on the workpiece and a real illumination unit for irradiating the workpiece with illumination light, and the real imaging unit and the real illumination unit are moved relative to the workpiece to perform an external inspection of the workpiece, and the illumination area corresponds to the inspection area of ​​the workpiece.

6. The robot teaching device according to claim 1, wherein the robot includes at least one of the following as a work unit for performing work on the workpiece: a three-dimensional shape measuring unit, a distance measuring sensor, a coating unit, a pasting unit, a spraying unit, a welding unit, a sewing unit, an ultrasonic flaw detection unit, an eddy current flaw detection unit, a tapping inspection unit, a peeling laser irradiation unit, and a hardening UV irradiation unit.

7. The robot teaching device according to claim 1, further comprising an output unit that outputs the teaching points calculated by the teaching point calculation unit to the control unit of an actual robot.

8. The robot teaching device according to claim 1, wherein the marker includes a three-dimensional object or seal that has been previously placed on the surface of the workpiece by the user.

9. A robot teaching method comprising: measuring a workpiece and a marker placed on the workpiece which serves as a teaching point for the movement path of a robot performing work on the workpiece, and generating three-dimensional data; recognizing the marker from the generated three-dimensional data of the workpiece and the marker; and calculating teaching points for the robot's movements based on the recognized marker.

Images