Information processing system, information processing program, and information processing method

Abstract

Conventional techniques have the problem that trajectory information cannot be generated unless weld lines are defined as lines in three-dimensional model data. An information processing system according to the present invention is provided with at least one computing device configured to perform: master point cloud data generation processing (S3) for generating master point cloud data in which a surface shape of a work target object is defined by a point cloud from three-dimensional model data of the work target object; and first work information data generation processing (S5) for extracting a work target part of the work target object from the master point cloud data and generating work information data including at least a movement path of a work tool for performing work with respect to the extracted work target part.

Description

Information processing system, information processing program, and information processing method

[0001] The present invention relates to an information processing system, an information processing program, and an information processing method, and more particularly to an information processing system, an information processing program, and an information processing method for generating work information data used for controlling a robot.

[0002] In controlling a work robot that performs work such as welding, it is necessary to generate work information data that includes at least information on the trajectory of the arm movement. Patent Document 1 discloses a technology for generating the trajectory information required to generate work information data. Patent Document 1 also discloses a weld line selection method that quickly extracts weld lines based on three-dimensional CAD data.

[0003] JP 2010-184278 A

[0004] However, the technique described in Patent Document 1 has a problem in that it is not possible to generate a weld line (for example, trajectory information) unless the weld line is defined as a line in the three-dimensional model data.

[0005] One aspect of the present invention is an information processing system comprising at least one computing device configured to perform a master point cloud data generation process that generates master point cloud data that defines the surface shape of a work object using a point cloud from three-dimensional model data of the work object, and a first work information data generation process that extracts work target locations on the work object from the master point cloud data and generates work information data that includes at least the movement path of a work tool that will perform work on the extracted work target locations.

[0006] One aspect of the present invention is an information processing program that causes a computing device to execute a master point cloud data generation process that generates master point cloud data that defines the surface shape of a work object using a point cloud from three-dimensional model data of the work object, and a first work information data generation process that extracts a work object location on the work object from the master point cloud data and generates work information data that includes at least the movement path of a work tool that will perform work on the extracted work object location.

[0007] One aspect of the present invention is an information processing method for generating work information data including at least the movement path of a work tool from three-dimensional model data using a computer, the information processing method including: generating master point cloud data that defines the surface shape of a work object using a point cloud from the three-dimensional model data of the work object; extracting work target locations on the work object from the master point cloud data; and generating the work information data corresponding to the extracted work target locations.

[0008] One aspect of the present invention uses master point cloud data generated from three-dimensional model data to generate work information data that includes at least the movement path of a work tool that will perform work on a work target location.

[0009] According to one aspect of the present invention, it is possible to generate work information data that includes at least the movement path of a work tool that performs work on a location that is not defined as line data in three-dimensional model data.

[0010] 6 is a schematic diagram of a working robot system according to a first embodiment. FIG. 7 is a diagram illustrating an example of work using the working robot system according to the first embodiment. FIG. 8 is a flowchart illustrating the operation of an information processing system according to the first embodiment. FIG. 9 is a flowchart illustrating the flow of work information data generation processing in the information processing system according to the first embodiment. FIG. 10 is a diagram illustrating a specific example of the operations from step S10 to step S13 in the flowchart shown in FIG. 4. FIG. 11 is a flowchart illustrating the flow of tool path determination processing in the information processing system according to the first embodiment. FIG. 12 is a diagram illustrating a specific example of the operations in steps S20, S22, and S24 in the flowchart shown in FIG. 6. FIG. 13 is a flowchart illustrating the operation of an information processing system according to a second embodiment. FIG. 14 is a flowchart illustrating the operation of an information processing system according to a third embodiment. FIG. 15 is a flowchart illustrating the flow of work information data correction processing in the information system according to the third embodiment.

[0011] For clarity of explanation, the following description and drawings have been omitted and simplified as appropriate. Furthermore, each element shown in the drawings as a functional block performing various processes can be configured in hardware with a CPU (Central Processing Unit), memory, and other circuits, and in software with a program loaded into memory. Therefore, those skilled in the art will understand that these functional blocks can be realized in various forms using only hardware, only software, or a combination thereof, and are not limited to any of these. In addition, the same elements are designated by the same reference numerals in each drawing, and redundant explanations are omitted as necessary.

[0012] Furthermore, the above-described program includes a set of instructions (or software code) that, when loaded into a computer, causes the computer to perform one or more functions described in the embodiments. The program may be stored in a non-transitory computer-readable medium or a tangible storage medium. By way of example and not limitation, computer-readable media or tangible storage media include random-access memory (RAM), read-only memory (ROM), flash memory, solid-state drive (SSD) or other memory technologies, CD-ROM, digital versatile disc (DVD), Blu-ray (registered trademark) disc or other optical disk storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage device. The program may also be transmitted on a transitory computer-readable medium or communication medium. By way of example and not limitation, transitory computer-readable media or communication media include electrical, optical, acoustic, or other forms of propagated signals.

[0013] Embodiment 1 Hereinafter, an embodiment of the present invention will be described with reference to the drawings. Fig. 1 is a schematic diagram of a working robot system according to embodiment 1. As shown in Fig. 1, working robot system 1 according to embodiment 1 includes a computer 10 and a working robot 20. Furthermore, working robot 20 has a robot arm 21 mounted on a base, and a work tool 22 and a working sensor 23 attached to the tip of working robot 20.

[0014] Computer 10 receives three-dimensional model data generated elsewhere, or generates three-dimensional model data. Computer 10 also uses the three-dimensional model data to generate work information data used in controlling work robot 20. Here, the generation of work data involves multiple processes, but these processes may be performed using at least one computer 10; the number of computers 10 included in work robot system 1 is not limited to one. Computer 10 also transmits the generated work information data to work robot 20. In work robot system 1 according to the first embodiment, work robot 20 has a control unit (not shown) that controls robot arm 21, and the control unit controls robot arm 21 using the work information data received.

[0015] The work robot 20 is capable of performing at least one of welding, painting, coating, cutting, polishing, grinding, and turning. The robot arm 21 moves a work tool 22 attached to the tip of the robot arm 21 and a work sensor 23 along a trajectory defined by work information data.

[0016] The work tool 22 varies depending on the work being performed. For example, the work tool 22 may be a welding torch, nozzle, welding rod, etc. for welding work, or a painting tool such as a paint gun for painting work. Other possible work tools include coating tools (e.g., coating sprayers), cutting tools (e.g., drills, end mills, milling cutters), polishing tools (e.g., polishing brushes), cutting tools (grindstones, cup brushes, diamond bars), turning tools (e.g., turning bits), etc.

[0017] The working unit sensor 23 is a sensor that scans the shape of a work object. Possible sensors of this type include LiDAR (Light Detection and Ranging), line scanners, and three-dimensional scanners. In the working robot system 1 according to the first embodiment, the working unit sensor 23 is used to recognize the shape of the work object before and after the work is performed.

[0018] Figure 2 is a diagram illustrating an example of a task using working robot system 1 according to embodiment 1. In the example shown in Figure 2, non-weld members 31 are placed on base material 30, and the non-weld members 31 are welded along a welding target line WL set at the end thereof, thereby joining the non-weld members 31 to the base material 30. In this case, in working robot system 1 according to embodiment 1, working robot 20 controls robot arm 21 so that the tip of work tool 22 moves along welding target line WL at a predetermined speed and at a predetermined angle relative to base material 30.

[0019] In working robot system 1 according to the first embodiment, work information data must be created for each task to be performed on a work object before production activities begin. In working robot system 1 according to the first embodiment, work information data is created from three-dimensional model data in computer 10. By creating work information data from three-dimensional model data in this way, work robot 20 is used to create work information data without using an actual object such as a prototype.

[0020] Here, the three-dimensional model data includes three-dimensional computer-aided design (CAD) data and attribute data, which will be described later. CAD data may be data containing surface information indicating the shape of a work object. CAD data formats include STEP (Standard for the Exchange of Product model data), IGES (Intial Graphics Exchange Specification), STL (STereoLithography), etc. The attribute data also includes data such as part names, part numbers, materials, dimensions, work conditions, and contour shapes. As will be described later, the attribute data may also include work object lines. The attribute data may be data in a computer-readable data format included in the three-dimensional model data, or may be assembly and component data (e.g., assembly data) described on paper drawings. The work information data includes shape data of the contour lines of the work object, position data indicating the position of the contour lines on the work object, and data related to work information such as the dimensions of the work object, the work position, the work path, and the material and thickness of each part of the work object. For example, in the case of welding work, the attribute data may include at least one of data on welding conditions such as welding current, welding speed, and torch angle.

[0021] In working robot system 1 according to the first embodiment, a model-based work information data generation process is carried out on computer 10 to generate work information data from three-dimensional model data without using an actual object. This model-based work information data generation process is carried out by executing an information processing program stored in a storage device mounted on computer 10 using a calculation device such as a CPU.

[0022] The model-based work information data generation process will be described in detail below. The model-based work information data generation process includes multiple processes, which may be executed by a single computer or multiple computers. Therefore, in the following description, the computer 10 will be referred to as an information processing system. For simplicity's sake, the model-based work information data generation process will be described below using an example in which work information data corresponding to welding work is generated. However, the model-based work information data generation process according to the first embodiment can be applied to various types of work other than welding work.

[0023] FIG. 3 is a flowchart illustrating the operation of the information processing system according to the first embodiment.

[0024] As shown in FIG. 3, the information processing system described below includes at least one arithmetic unit configured to perform a master point cloud data generation process (steps S1 to S3) that generates master point cloud data that defines the surface shape of a work object using a point cloud from three-dimensional model data of the work object, and a first work information data generation process (step S5) that extracts work target locations on the work object from the master point cloud data and generates work information data that includes at least the movement path of a work tool that will perform work on the extracted work target locations.

[0025] 3, in the model-based work information data generation process according to the first embodiment, first, a model data acquisition process is performed to acquire already-generated 3D model data of the work object (step S1). Next, in the model-based work information data generation process according to the first embodiment, a 3D mesh data generation process is performed to generate 3D mesh data of the work object from the 3D model data acquired in step S1 (step S2). These steps S1 and S2 may be collectively referred to as a data conversion process.

[0026] The three-dimensional mesh data generated in step S2 represents the overall shape of the work object using unit meshes defined at a predetermined size. The three-dimensional mesh data makes it possible to handle the shape of the work object as a set of surfaces.

[0027] Next, the model-based work information data generation process according to the first embodiment performs a master point cloud data generation process (step S3) to generate master point cloud data that defines the surface shape of the work object as a point cloud from the three-dimensional model data of the work object. In the example shown in FIG. 2 , the three-dimensional mesh data generated in step S2 is used as the three-dimensional model data in step S3. In the master point cloud data generation process according to the first embodiment, for example, a virtual cube that covers the entire three-dimensional mesh data is defined, and when a measurement line perpendicular to one face of the virtual cube is extended to the three-dimensional mesh data, the intersection of the measurement line and the three-dimensional mesh data is defined as one piece of master point cloud data. The master point cloud data includes intersections between numerous measurement lines and the three-dimensional mesh data and is a collection of countless point information. This point information may be, for example, coordinate information. Here, for example, if one side of the virtual cube is a side on which the work object is considered to be observable by the work unit sensor 23 when the work object is placed on the workbench, a correlation can be made with the actual point cloud data of the work object acquired by the work unit sensor 23 during actual work, thereby improving work efficiency.

[0028] Next, the model-based work information data generation process according to the first embodiment performs an attribute data extraction process to extract attribute data from the three-dimensional model data (step S4). The attribute data extracted in step S4 includes at least position information of the work target location defined by the dimensions of the work target, the work position, and the work path. The attribute data may also include position data indicating the position of the outline of the work target, and data related to work information such as the material and thickness of each part of the work target. Here, in the model-based work information data generation process according to the first embodiment, as an example, the position information of the work target location is defined by the attribute data included in the three-dimensional model data.

[0029] Next, the model-based work information data generation process according to the first embodiment performs a first work information data generation process (e.g., the work information data generation process of step S5) that extracts a work target location of the work object from the master point cloud data and generates work information data that includes at least an operation path (e.g., a tool path) of the work tool 22 that performs work on the extracted work target location. The work information data generation process of step S5 includes multiple processes, so the work information data generation process of step S5 will be described in detail with reference to FIGS. 4 and 5.

[0030] FIG. 4 is a flowchart illustrating the flow of work information data generation processing in the information processing system according to the first embodiment. FIG. 5 is a diagram illustrating a specific example of the operations from step S10 to step S13 of the flowchart shown in FIG. 4. Note that in FIG. 5, the surface of the work object on which the point cloud is set is indicated by hatching. Also in FIG. 5, a work object (e.g., a non-welded member 31) is placed on the surface of a base material 30, and a curved portion of the periphery of the non-welded member 31 is to be welded. That is, in the first embodiment, an example is described in which work information data corresponding to a curved work object line is generated.

[0031] As shown in Figure 4, the work information data generation process in step S5 includes a base material surface extraction process (steps S10, S11) that extracts the base material surface on which the work target line is set from the master point cloud data, a boundary line extraction process (step S12) that extracts a boundary line that defines the periphery of the base material surface on the master point cloud data, a continuous line data conversion process (step S13) that converts the line among the boundary lines that is designated as the work target line into continuous line data consisting of a series of approximation lines, and a tool path determination process (step S14) that determines a tool path that defines the movement trajectory of the work tool 22 attached to the robot arm 21 based on the continuous line data.

[0032] 4, in the work information data generation process of step S5, first, the master point cloud data generated in step S3 is acquired (step S10). The master point cloud data acquired in step S10 is, for example, the data shown in the diagram corresponding to step S10 in FIG. 5, in which non-welded members 31 are placed on the surfaces of base material 30. In the first embodiment, the curved portions of the outer periphery of non-welded members 31 are to be welded.

[0033] Next, in the work information data generation process of step S5, a base material surface extraction process is performed to extract the base material surface on which the work target line is set from the master point cloud data (step S11). As shown in the diagram corresponding to step S11 in Figure 5, in step S11, an area excluding the point cloud data of the portion corresponding to the non-welded member 31 is extracted from the master point cloud data as the base material surface.

[0034] Next, in the work information data generation process of step S5, a boundary line extraction process is performed to extract a boundary line that defines the periphery of the base material surface on the master point cloud data (step S12). As shown in the diagram corresponding to step S12 in Figure 5, in step S12, a boundary line 32 that defines the periphery of the base material surface is extracted from the master point cloud data.

[0035] Next, in the work information data generation process of step S5, a continuous line data conversion process is performed to convert the line designated as the work target line 33 from the boundary line 32 into continuous line data consisting of a series of approximation lines (step S13). In the diagram corresponding to step S13 in FIG. 5, only the work target line 33 extracted in step S13 is shown. This work target line 33 is defined as continuous line data by combining multiple approximation lines, each of which is a straight line. The approximation lines may also include curved lines. This work target line 33 may be designated by the attribute data extracted in step S4, or may be designated by the user while referring to the screen.

[0036] Next, in the work information data generation process of step S5, a tool path determination process is performed to determine a tool path that defines the movement trajectory of a tool (e.g., work tool 22) attached to the robot arm based on the continuous line data (step S14). Here, the tool path determination process of step S14 is performed by a plurality of processes, so the tool path determination process will be described in detail with reference to Figures 6 and 7.

[0037] Fig. 6 is a flowchart illustrating the flow of a tool path determination process in the information processing system according to the first embodiment. Fig. 7 is a diagram illustrating a specific example of the operations of steps S20, S22, and S24 of the flowchart shown in Fig. 6. Note that in Fig. 7 as well, the surfaces of the workpiece on which the point cloud is set are indicated by hatching. Fig. 7 shows the same example as Fig. 5. Note that, for the sake of explanation, the continuous line data shown in Fig. 7 is a schematic data having fewer approximation lines than the number of approximation lines that make up the actual continuous line data.

[0038] As shown in FIG. 6 , the tool path determination process in step S14 includes an endpoint extraction process (step S20) that extracts endpoints 34 of each approximation line included in the continuous line data; a normal calculation process (steps S22 and S23) that calculates a normal 36 for each endpoint 34; a Cartesian coordinate system calculation process (step S24) that normalizes the Z, X, and Y axes to calculate Cartesian coordinate system data for each endpoint 34, with the normal 36 defined as the Z axis, the direction toward the next endpoint 34 defined as the X axis, and the cross product of the Z and X axes defined as the Y axis; and a data generation process (step S27) that generates work information data by connecting the Cartesian coordinate system data calculated for each endpoint 34 in order from the start point of the continuous line data toward the endpoint.

[0039] As shown in Fig. 6, the tool path determination process in step S14 first performs an endpoint extraction process (step S20) to extract endpoints of each approximation line included in the continuous line data obtained in step S13. As shown in Fig. 7, which corresponds to step S20, the endpoint extraction process sets a large number of endpoints 34 on the continuous line data. Note that of the multiple endpoints 34, the starting point of the continuous line data is numbered 0, and the ending point is numbered Ne. Also, in Fig. 6, N is used as a variable indicating the endpoint number.

[0040] Next, in the tool path determination process of step S14, N is initialized to 0 (step S21), and N is incremented by one (step S26), and the processes of steps S22 to S24 are repeated until the endpoint number N becomes the endpoint number Ne of the end point (step S25).

[0041] In the tool path determination process of step S14, normal calculation processes for calculating normals for each endpoint are performed in steps S22 and S23. In step S22, point cloud data within a range of radius R centered on the endpoint 34 of the target object is extracted. As shown in FIG. 7 corresponding to step S22, in step S22, only point cloud data within a range of radius R centered on the endpoint 34 of the target object is extracted. Then, in step S23, a normal 36 is calculated at the endpoint 34 of the target object using the point cloud data extracted in step S22. As shown in FIG. 7 corresponding to step S23, in step S23, a normal 36 is calculated starting from the endpoint 34 of the target object. This normal is calculated by analyzing the point cloud data within the range of radius R centered on the endpoint 34 using a method such as principal component analysis (PCA).

[0042] Next, in the tool path determination process of step S14, a Cartesian coordinate system calculation process is performed for each endpoint, where the normal is defined as the Z axis, the direction toward the next endpoint is defined as the X axis, and the cross product of the Z axis and the X axis is defined as the Y axis, and the Z, X, and Y axes are normalized to calculate Cartesian coordinate system data (step S24). In the diagram corresponding to step S24 shown in Figure 7, the Z, X, and Y axes are defined for three endpoints 34, but in step S24, Cartesian coordinate system data is calculated by normalizing the Z, X, and Y axes for all endpoints.

[0043] Next, in the tool path determination process of step S14, a data generation process is performed in which the Cartesian coordinate system data calculated for each endpoint calculated in step S24 of the repeated processes of steps S22 to S26 are connected in order from the start point of the continuous line data toward the endpoint to generate work information data (tool path) (step S27). Also, in step S27, the generated work data is stored in a work information data storage unit, which is one of the storage units on the computer 10.

[0044] As explained above, the information processing system according to the first embodiment generates work information data using master point cloud data generated from three-dimensional model data. By converting the three-dimensional model data into master point cloud data in this way, it becomes possible to define a work target line 33 in the master point cloud data, even if the work target area is not defined as line data in the three-dimensional model data. Then, by using this work target line 33, the information processing system according to the first embodiment can generate work information data applicable to the work robot 20.

[0045] Furthermore, the information processing system according to the first embodiment generates work information data without using the actual product, eliminating the need to generate work information data using the work robot 20 and a prototype, and therefore, for example, work information data can be generated regardless of the installation location of the work robot 20. In other words, with the work robot system 1 according to the first embodiment, work information data can be created without operating the work robot 20 in advance, which can improve business efficiency.

[0046] Furthermore, when work information data is created using an actual work object and work robot 20, variations in the placement of the work object and other factors can adversely affect the accuracy of the work information data. However, with work robot system 1 according to the first embodiment, work information data is generated from three-dimensional model data that is free of variations, making it possible to generate highly accurate work information data that is unaffected by variations in the placement of the work object and other factors.

[0047] In the second embodiment, another embodiment of the model-based work information data generation process according to the first embodiment will be described. In the description of the second embodiment, the same reference numerals as those in the first embodiment will be used to denote the same items as those in the first embodiment, and the description thereof will be omitted.

[0048] Fig. 8 is a flowchart illustrating the operation of the information processing system according to the second embodiment. As shown in Fig. 8, the model-based work information data generation process according to the second embodiment is obtained by adding steps S30 to S33 to the model-based work information data generation process according to the first embodiment.

[0049] The information processing system of the second embodiment includes a data conversion process (steps S1, S2) that acquires three-dimensional model data of a work object and generates three-dimensional mesh data of the work object; a shape determination process (step S30) that determines whether the work object location on the three-dimensional mesh data is defined as line data on the three-dimensional model data; a primitive shape data extraction process (step S31) that extracts a primitive shape from the three-dimensional mesh data and generates primitive shape data if the work object location is defined as line data on the three-dimensional model data in the shape determination process (step S30); and a second work information data generation process (steps S32, S33) that generates work information data in accordance with the line data on the primitive shape data.If the work object location is defined as line data on the three-dimensional model data in the shape determination process (step S30), the information processing system performs a master point cloud data generation process (step S3) and a first work information data generation process (step S5).

[0050] In the model-based work information data generation process according to the second embodiment, after the three-dimensional mesh data generation process (step S2), a shape determination process is carried out to determine whether the work target area is defined as line data on the three-dimensional model data (step S30).

[0051] In the three-dimensional model data, lines defining the shape of components are defined as line data. Therefore, when joining end faces, line data defining the periphery of the component end faces is defined in the three-dimensional model data, and work information data can be generated using this line data as a work target line. On the other hand, for example, when performing processing such as overlaying one component on the surface of another component, no line data is defined on the surface of the one component, and it is necessary to perform the work information data generation process (step S5) using the master point cloud data (step S3) described in the first embodiment.

[0052] In the model-based work information data generation process according to the second embodiment, if the work target area is not defined as line data on the three-dimensional model data in the shape determination process of step S30, the processes of steps S3 to S5 described in the first embodiment are carried out.

[0053] On the other hand, in the model-based work information data generation process according to the second embodiment, if the work target area is defined as line data on the three-dimensional model data in the shape determination process of step S30, a primitive shape data extraction process is performed in which a primitive shape is extracted from the three-dimensional mesh data and primitive shape data is generated (step S31).

[0054] Here, a primitive shape is a relatively simple, predetermined shape such as a cube, a rectangular parallelepiped, a sphere, a cylinder, or a cone. Data containing information about the lines and faces of such shapes is called primitive shape data. Then, in the primitive shape data extraction process in step S31, the work object is replaced with model data that combines primitive shapes.

[0055] Next, in the model-based work information data generation process according to the second embodiment, an attribute data extraction process is performed to extract attribute data from the three-dimensional model data in the same manner as in step S4 (step S32). After that, in the model-based work information data generation process according to the second embodiment, a second work information data generation process (e.g., work information data generation process (step S33)) is performed to generate work information data along the line data on the primitive shape data.

[0056] Replacing model data using primitive shape data can reduce the computational load compared to creating master point cloud data. Therefore, by using the model-based work information data generation process according to the second embodiment, the amount of calculation required by the information processing system can be appropriately reduced depending on the complexity of the work object and the work content. In other words, by using the model-based work information data generation process according to the second embodiment, it is possible to reduce the time required to generate work information data.

[0057] In the third embodiment, another embodiment of the model-based work information data generation process according to the first embodiment will be described. In the description of the third embodiment, the same reference numerals as those in the first embodiment will be used to denote the items described in the first embodiment, and the description thereof will be omitted.

[0058] Fig. 9 is a flowchart illustrating the operation of the information processing system according to the third embodiment. Fig. 10 is a flowchart illustrating the flow of work information data correction processing in the information system according to the third embodiment.

[0059] As shown in Figures 9 and 10, the model-based work information data generation process according to the third embodiment includes an actual point cloud data acquisition process (step S40) that acquires actual point cloud data from a work object using the working unit sensor 23, and a work information data correction process (step S41) that generates corrected work information data by correcting the work information data based on the actual point cloud data. The work information data correction process (step S41) includes a transformation matrix calculation process (steps S50, S51) that calculates a transformation matrix for transforming the master point cloud data so that the positions of pre-specified characteristic locations between the master point cloud data and the actual point cloud data coincide with those between the master point cloud data and the actual point cloud data, and a data correction process (step S52) that applies the inverse matrix of the transformation matrix to the work information data to generate corrected work information data.

[0060] 9, the model-based work information data generation process according to the third embodiment is obtained by adding steps S40 and S41 to the model-based work information data generation process according to the first embodiment. Specifically, the model-based work information data generation process according to the third embodiment corrects the work information data generated in step S5 in accordance with the status of the work target (actual object) placed on the workbench in steps S40 and S41.

[0061] Therefore, in the model-based work information data generation process according to the third embodiment, after the work information data generation process of step S5, an actual point cloud data acquisition process is performed (step S40) to acquire actual point cloud data from the work object using a sensor (e.g., work unit sensor 23).

[0062] Next, in the model-based work information data generation process according to the third embodiment, a work information data correction process is performed to generate corrected work information data by correcting the work information data based on the actual point cloud data (step S41). Here, the work information data correction process in step S41 involves multiple processes, which will be described with reference to FIG. 10.

[0063] 10, in the work information data correction process of step S41, a feature extraction process is performed (step S50) to extract feature parts that are geometric characteristics from both the master point cloud data created in step S3 and the actual point cloud data acquired in step S40. These feature parts can be, for example, notches, holes, or normals to specific locations on the work object.

[0064] Next, in the work information data correction process of step S41, a transformation matrix calculation process is performed to calculate a transformation matrix that converts the master point cloud data so that the positions of pre-specified characteristic points (e.g., characteristic parts extracted in step S50) between the actual point cloud data and the master point cloud data match (step S51).

[0065] Next, in the work information data correction process of step S41, the inverse matrix of the transformation matrix is ​​applied to the work information data generated in step S5 to generate corrected work information data (step S52). By this data correction process of step S52, corrected work information data is generated by offsetting the work information data generated in step S5 by an offset amount corresponding to the amount of deviation of the actual point cloud data from the master point cloud data.

[0066] As explained above, with the model-based work information data generation process according to the third embodiment, even if an actual object is placed on the work table at a position or angle that deviates from the shape defined by the master point cloud data, it is possible to operate the work robot 20 based on work information data that has been corrected to account for the deviation. By generating such work information data, it is possible to improve the accuracy of work performed by the work robot 20.

[0067] The actual point cloud data acquisition process (step S40) and the work information data correction process (step S41) of the model-based work information data generation process according to the third embodiment may be performed by the computer 10, or may be performed by a computer placed near the work robot 20 or a processor in a control unit built into the work robot 20.

[0068] The present invention is not limited to the above-described embodiments, and can be modified as appropriate without departing from the spirit of the present invention. For example, a person skilled in the art can easily conceive of implementing the present invention by appropriately combining the features described in each embodiment.

[0069] Finally, the embodiments of the present invention will be summarized with reference to the drawings etc. As shown in Figures 1 to 10, the embodiments of the present invention are as follows.

[0070] (Supplementary Note 1) An information processing system comprising at least one arithmetic device configured to perform: a master point cloud data generation process (S1 to S3) for generating master point cloud data that defines the surface shape of a work object using a point cloud from three-dimensional model data of the work object; and a first work information data generation process (S5) for extracting work target locations on the work object from the master point cloud data, and generating work information data that includes at least the movement path of a work tool that will perform work on the extracted work target locations.

[0071] (Supplementary Note 2) The information processing system described in Supplementary Note 1, wherein the first work information data generation process (S5) includes: a base material surface extraction process (S10, S11) that extracts a base material surface on which a work target line is set from the master point cloud data; a boundary line extraction process (S12) that extracts a boundary line that defines the periphery of the base material surface on the master point cloud data; a continuous line data conversion process (S13) that converts a line of the boundary line that is designated as the work target line into continuous line data consisting of a series of approximation lines; and a tool path determination process (S14) that determines a tool path that defines a movement trajectory of a tool attached to a robot arm based on the continuous line data.

[0072] (Supplementary Note 3) The information processing system according to Supplementary Note 2, comprising: an endpoint extraction process (S20) that extracts endpoints of each approximation line included in the continuous line data; a normal calculation process (S22, S23) that calculates a normal for each of the endpoints; a Cartesian coordinate system calculation process (S24) that calculates Cartesian coordinate system data for each of the endpoints by normalizing the Z axis, the X axis, and the Y axis, with the normal as the Z axis, the direction toward the next endpoint as the X axis, and the cross product of the Z axis and the X axis as the Y axis; and a data generation process (S27) that generates the work information data by connecting the Cartesian coordinate system data calculated for each of the endpoints in order from the start point of the continuous line data toward the endpoints.

[0073] (Supplementary Note 4) The information processing system according to Supplementary Note 3, wherein the position information of the work target location is defined by attribute data included in the three-dimensional model data.

[0074] (Supplementary Note 5) An information processing system according to any one of Supplementary Notes 1 to 4, comprising: a data conversion process (S1, S2) of acquiring three-dimensional model data of the work object and generating three-dimensional mesh data of the work object; a shape determination process (S30) of determining whether the work object location is defined as line data on the three-dimensional mesh data; a primitive shape data extraction process (S31) of extracting a primitive shape from the three-dimensional mesh data and generating primitive shape data if the work object location is defined as line data on the three-dimensional model data in the shape determination process; and a second work information data generation process (S32, S33) of generating the work information data along the line data on the primitive shape data, wherein the information processing system performs the master point cloud data generation process (S3) and the first work information data generation process (S5) if the work object location is defined as line data on the three-dimensional model data in the shape determination process (S30).

[0075] (Supplementary Note 6) An information processing system according to any one of Supplementary Notes 1 to 5, comprising: an actual point cloud data acquisition process (S40) for acquiring actual point cloud data from the work object using a sensor; and a work information data correction process (S41) for generating corrected work information data by correcting the work information data based on the actual point cloud data, wherein the work information data correction process (S41) comprises: a transformation matrix calculation process (S50, S51) for calculating a transformation matrix for transforming the master point cloud data so that positions of pre-specified characteristic points between the master point cloud data and the actual point cloud data coincide with those of the master point cloud data; and a data correction process (S52) for applying an inverse matrix of the transformation matrix to the work information data to generate the corrected work information data.

[0076] (Supplementary Note 7) An information processing program that causes a computing device to execute the following steps: a master point cloud data generation process (S1 to S3) that generates master point cloud data that defines the surface shape of a work object using a point cloud from three-dimensional model data of the work object; and a first work information data generation process (S5) that extracts work target locations on the work object from the master point cloud data and generates work information data that includes at least the movement path of a work tool that will perform work on the extracted work target locations.

[0077] (Supplementary Note 8) An information processing method for generating work information data including at least a movement path of a work tool from three-dimensional model data using a computer, the information processing method comprising the steps of: generating master point cloud data that defines the surface shape of a work object using a point cloud from the three-dimensional model data of the work object (S1 to S3); extracting work target locations on the work object from the master point cloud data; and generating the work information data corresponding to the extracted work target locations (S5).

[0078] This application claims priority based on Japanese Patent Application No. 2023-124823, filed on July 31, 2023, the disclosure of which is incorporated herein by reference in its entirety.

[0079] REFERENCE SIGNS LIST 1 Working robot system 10 Computer 20 Working robot 21 Robot arm 22 Working tool 23 Working sensor 30 Base material 31 Non-weld member WL Welding target line

Claims

1. An information processing system comprising at least one computing device configured to perform: a master point cloud data generation process for generating master point cloud data that defines the surface shape of a work object as a point cloud from three-dimensional model data of the work object; and a first work information data generation process for extracting work target locations of the work object from the master point cloud data, and generating work information data including at least the movement path of a work tool that will perform work on the extracted work target locations.

2. The information processing system of claim 1, wherein the first work information data generation process includes: a base material surface extraction process for extracting a base material surface on which a work target line is set from the master point cloud data; a boundary line extraction process for extracting a boundary line defining the periphery of the base material surface on the master point cloud data; a continuous line data conversion process for converting a line among the boundary lines designated as the work target line into continuous line data consisting of a series of approximation lines; and a tool path determination process for determining a tool path defining the movement trajectory of a tool attached to a robot arm based on the continuous line data.

3. An information processing system as described in claim 2, comprising: an endpoint extraction process that extracts endpoints of each approximation line included in the continuous line data; a normal calculation process that calculates a normal for each of the endpoints; a Cartesian coordinate system calculation process that calculates Cartesian coordinate system data for each of the endpoints by normalizing the Z axis, the X axis, and the Y axis, with the normal as the Z axis, the direction toward the next endpoint as the X axis, and the cross product of the Z axis and the X axis as the Y axis; and a data generation process that generates the work information data by connecting the Cartesian coordinate system data calculated for each of the endpoints in order from a starting point of the continuous line data toward the endpoint.

4. An information processing system according to claim 1, wherein the position information of the work target area is defined by attribute data included in the three-dimensional model data.

5. An information processing system as described in claim 1, comprising: a data conversion process for acquiring three-dimensional model data of the work object and generating three-dimensional mesh data of the work object; a shape determination process for determining whether the work object is defined as line data on the three-dimensional mesh data; a primitive shape data extraction process for extracting a primitive shape from the three-dimensional mesh data and generating primitive shape data when the work object is defined as line data on the three-dimensional model data in the shape determination process; and a second work information data generation process for generating the work information data along the line data on the primitive shape data, wherein the information processing system as described in claim 1 performs the master point cloud data generation process and the first work information data generation process when the work object is defined as line data on the three-dimensional model data in the shape determination process.

6. An information processing system as described in claim 1, comprising: an actual point cloud data acquisition process for acquiring actual point cloud data from the work object using a sensor; and a work information data correction process for generating corrected work information data by correcting the work information data based on the actual point cloud data, wherein the work information data correction process includes a transformation matrix calculation process for calculating a transformation matrix for transforming the master point cloud data so that positions of pre-specified characteristic points between the master point cloud data and the actual point cloud data coincide with those of the master point cloud data; and a data correction process for applying an inverse matrix of the transformation matrix to the work information data to generate the corrected work information data.

7. An information processing program that causes a computing device to execute the following: a master point cloud data generation process that generates master point cloud data that defines the surface shape of a work object as a point cloud from three-dimensional model data of the work object; and a first work information data generation process that extracts work target locations of the work object from the master point cloud data, and generates work information data that includes at least the movement path of a work tool that will perform work on the extracted work target locations.

8. An information processing method for generating work information data including at least a movement path of a work tool from three-dimensional model data using a computer, the information processing method comprising: generating master point cloud data that defines a surface shape of a work object as a point cloud from the three-dimensional model data of the work object; extracting work target locations of the work object from the master point cloud data; and generating the work information data corresponding to the extracted work target locations.

Images