Workpiece grabbing pose determination method and device, computer device, and storage medium

Abstract

This application discloses a workpiece grasping pose determination method, apparatus, computer device, and storage medium. The method includes: acquiring a scene point cloud corresponding to a target workpiece to be grasped in a workpiece placement area, and determining first candidate grasping pose information based on a preset point cloud template corresponding to the target workpiece and the scene point cloud; acquiring preset symmetry parameters corresponding to the target workpiece; transforming the first candidate grasping pose information according to the preset symmetry parameters to obtain second candidate grasping pose information; and determining the target grasping pose information from the first and second candidate grasping pose information based on a preset grasping pose determination strategy. The target grasping pose information is used to determine the target position and target posture when the robotic arm grasps the target workpiece. This application is beneficial for improving the workpiece grasping success rate of the robotic arm in complex grasping scenarios, thereby improving workpiece grasping efficiency.

Description

Technical Field

[0001] This application relates to the field of point cloud processing technology, and in particular to a method, apparatus, computer equipment and storage medium for determining the posture of a workpiece gripping. Background Technology

[0002] Automated workpiece gripping is an important technology in modern industrial production. It uses grippers in robotic arms to automatically hold, transport, and place workpieces, increasing the level of automation in the production process, greatly improving production efficiency, reducing labor costs, and enabling stable operation in harsh environments.

[0003] To guide a robotic arm in successfully grasping a workpiece, it's typically necessary to determine the arm's grasping pose information (including its position and orientation). Once the robotic arm reaches the corresponding grasping pose, it can then grasp the workpiece. Point cloud matching is a computer vision technology that uses point cloud data to identify and locate objects. In the field of automated workpiece grasping, point cloud matching technology helps robots identify the workpiece's position and orientation, thereby achieving precise grasping. This significantly improves grasping accuracy and brings many conveniences to industrial production.

[0004] To improve the gripping accuracy of robotic arms, existing methods for determining workpiece gripping poses typically rely on point cloud matching. For example, a point cloud template of the workpiece and an optimal gripping pose corresponding to that template are pre-established. The gripping pose in the actual scene is then determined based on the matching between the workpiece point cloud and the template. This method focuses on improving template matching accuracy but neglects the complexity of actual gripping scenarios. Even with high-precision matching between the point cloud template and the actual point cloud, and a high-precision gripping pose obtained from the optimal pose, this pose may still fail to successfully grip the workpiece in complex scenarios, resulting in low gripping efficiency. Therefore, a new method for determining workpiece gripping poses is urgently needed to improve the ability to handle complex gripping scenarios, increase the success rate of robotic arm gripping, and ultimately improve workpiece gripping efficiency. Summary of the Invention

[0005] This application provides a workpiece gripping pose determination method, apparatus, computer equipment, and storage medium, which helps to improve the success rate of robotic arms gripping workpieces in complex gripping scenarios, thereby improving the efficiency of automatic workpiece gripping.

[0006] In a first aspect, embodiments of this application provide a method for determining the gripping pose of a workpiece, including:

[0007] Obtain the scene point cloud corresponding to the target workpiece to be grabbed in the workpiece placement area;

[0008] The first candidate grasping pose information is determined based on the preset point cloud template corresponding to the target workpiece and the scene point cloud.

[0009] Obtain the preset symmetry parameters corresponding to the target workpiece;

[0010] The first candidate grasping pose information is transformed according to the preset symmetry parameters to obtain the second candidate grasping pose information;

[0011] Based on a preset grasping pose determination strategy, the target grasping pose information is determined from the first candidate grasping pose information and the second candidate grasping pose information; wherein, the target grasping pose information is used to determine the target position and target posture corresponding to the robotic arm grasping the target workpiece.

[0012] Secondly, embodiments of this application provide a workpiece gripping pose determination device, comprising:

[0013] The first acquisition unit is used to acquire the scene point cloud corresponding to the target workpiece to be grabbed in the workpiece placement area;

[0014] The first determining unit is used to determine the first candidate grasping pose information based on the preset point cloud template corresponding to the target workpiece and the scene point cloud.

[0015] The second acquisition unit is used to acquire the preset symmetry parameters corresponding to the target workpiece;

[0016] The pose transformation unit is used to transform the first candidate grasping pose information according to the preset symmetry parameters to obtain the second candidate grasping pose information.

[0017] The second determining unit is used to determine the target grasping pose information from the first candidate grasping pose information and the second candidate grasping pose information based on a preset grasping pose determination strategy; wherein, the target grasping pose information is used to determine the target position and target posture corresponding to the robotic arm grasping the target workpiece.

[0018] Thirdly, embodiments of this application also provide a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the workpiece gripping pose determination method of the first aspect described above.

[0019] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to perform the workpiece grasping pose determination method described in the first aspect.

[0020] This application provides a workpiece gripping pose determination method, apparatus, computer device, and storage medium. The method includes: acquiring a scene point cloud corresponding to a target workpiece to be gripped in a workpiece placement area, and determining a first candidate gripping pose information based on a preset point cloud template corresponding to the target workpiece and the scene point cloud; acquiring preset symmetry parameters corresponding to the target workpiece; transforming the first candidate gripping pose information according to the preset symmetry parameters to obtain a second candidate gripping pose information; and determining a target gripping pose information from the first candidate gripping pose information and the second candidate gripping pose information based on a preset gripping pose determination strategy; wherein the target gripping pose information is used to determine the target position and target posture corresponding to the robotic arm gripping the target workpiece. In this embodiment, after obtaining the grasping pose information (first candidate grasping pose information) of the target workpiece based on template matching, the preset symmetry parameters corresponding to the target workpiece are further obtained. Then, the grasping pose information can be transformed according to the preset symmetry parameters to obtain other grasping pose information. On this basis, the target grasping pose information that conforms to the complex grasping scenario can be determined from all the obtained grasping pose information based on the preset grasping pose determination strategy. This is beneficial to improving the robustness of the robotic arm grasping, improving the success rate of the robotic arm grasping workpieces in complex grasping scenarios, and thus improving the grasping efficiency of the workpiece. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of the first process of the workpiece gripping pose determination method provided in an embodiment of this application.

[0023] Figure 2 A schematic diagram of the preset center axis of the target workpiece provided in the embodiments of this application;

[0024] Figure 3 A schematic diagram of the angular rotation transformation of the pose of the target workpiece provided in the embodiments of this application.

[0025] Figure 4A Another flowchart illustrating the workpiece gripping pose determination method provided in an embodiment of this application;

[0026] Figure 4B Another flowchart illustrating the workpiece gripping pose determination method provided in an embodiment of this application;

[0027] Figure 5A schematic diagram showing the dimensional data of the robotic arm and robotic arm gripper provided in the embodiments of this application;

[0028] Figure 6A Another flowchart illustrating the workpiece gripping pose determination method provided in an embodiment of this application;

[0029] Figure 6B A schematic diagram showing the robotic arm gripper reaching a candidate grasping pose according to an embodiment of this application;

[0030] Figure 6C This is a schematic diagram showing the range of the target grasping space and the target workpiece space provided in the embodiments of this application;

[0031] Figure 6D This is a schematic diagram of the range of the grasping interference space provided in an embodiment of this application;

[0032] Figure 7 Another flowchart illustrating the workpiece gripping pose determination method provided in an embodiment of this application;

[0033] Figure 8 A schematic block diagram of the workpiece gripping pose determination device provided in the embodiments of this application;

[0034] Figure 9 A schematic block diagram of a computer device provided in an embodiment of this application. Detailed Implementation

[0035] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0036] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0037] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of the application. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0038] It should also be further understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0039] This application provides a workpiece gripping pose determination method, apparatus, computer device, and storage medium, which helps improve the success rate of robotic arms gripping workpieces in complex gripping scenarios, thereby improving the efficiency of automatic workpiece gripping. The execution entity of this workpiece gripping pose determination method can be the workpiece gripping pose determination device provided in the embodiments of this application, or a computer device integrating the workpiece gripping pose determination device provided in the embodiments of this application. The workpiece gripping pose determination device can be implemented in hardware or software, and the computer device can be a terminal or a server. The terminal can be a smartphone, tablet computer, PDA, or laptop computer, etc.

[0040] Please see Figure 1 , Figure 1 This is a flowchart illustrating a workpiece gripping pose determination method according to an embodiment of this application. In a specific implementation, the execution subject of this method can be a computer device used to control a robotic arm, such as a host computer. The method specifically includes the following steps S101 to S105.

[0041] Step S101: Obtain the scene point cloud corresponding to the target workpiece to be grabbed in the workpiece placement area.

[0042] In this context, the target workpiece refers to the workpiece to be grasped; the scene point cloud refers to the point cloud data corresponding to the target workpiece, and the scene point cloud corresponding to the target workpiece is the point cloud data corresponding to the target workpiece in its current placement posture. In practice, a depth camera can be used to photograph the target workpiece in the workpiece placement area, thereby obtaining the scene point cloud corresponding to the target workpiece. In practice, this depth camera may or may not be mounted on the robotic arm. If mounted on the robotic arm, the depth camera can be fixed to the end effector of the robotic arm using an "eye-on-hand" method, moving with the end effector of the robotic arm. Based on this, the robotic arm can obtain the scene point cloud corresponding to the target workpiece by moving to the vicinity of the workpiece placement area, for example, above the workpiece placement area, and photographing it.

[0043] Step S102: Determine the first candidate grasping pose information based on the preset point cloud template corresponding to the target workpiece and the scene point cloud.

[0044] The aforementioned first candidate grasping pose information is candidate and is used to determine the position and orientation of the robotic arm when grasping the target workpiece. Specifically, the first candidate grasping pose information may include the first candidate position and first candidate orientation of the robotic arm when grasping the target workpiece. In this embodiment, the scene point cloud reflects the current orientation of the target workpiece to be grasped. To determine the grasping pose information of the target workpiece in this orientation, the grasping pose information (first candidate grasping pose information) is determined based on a preset point cloud template corresponding to the target workpiece and the scene point cloud.

[0045] Specifically, in some implementations, the point cloud template corresponding to the target workpiece may include the grasping pose corresponding to the point cloud template, that is, the template grasping pose. Then, by performing point cloud matching between the point cloud template and the scene point cloud, the transformation relationship between the point cloud template and the scene point cloud can be obtained. Then, the template grasping pose can be transformed accordingly based on the transformation relationship. The transformation result is further transformed into the robot arm coordinate system, that is, the first candidate grasping pose information can be obtained.

[0046] The robotic arm coordinate system refers to the coordinate system used to describe the position and movement of the robotic arm. Typically, a robotic arm has two important coordinate systems:

[0047] Base coordinate system: with the robot arm base as the origin, used to describe the overall position and movement of the robot arm.

[0048] End-effector coordinate system: With the end effector (usually the gripper or tool mounting position) of the robotic arm as the origin, it describes the position and motion of the end effector and helps control the robotic arm to perform precise operations. The relationship between the base coordinate system and the end-effector coordinate system can be calculated using forward and inverse kinematics.

[0049] When a gripping tool, such as a chuck, is mounted at the end effector of the robotic arm, the end effector coordinate system is also called the tool coordinate system. The origin and x, y, z axes of the tool coordinate system depend on the tool's mounting method and usage requirements. For example, in one specific embodiment, the origin of the tool coordinate system is located at the operating part of the tool, such as the tip of a drill bit or the center of a chuck; the z-axis is the direction the chuck points towards the workpiece when gripping the tool; and the x and y axes can be perpendicular to the z-axis and parallel to the robotic arm base plane.

[0050] Based on this, determining the template grasping pose involves determining the grasping point position and x, y, z axis directions within the point cloud template in the tool coordinate system. Further, according to the transformation relationship between the point cloud template and the scene point cloud, the grasping point position and x, y, z axis directions in the point cloud template are transformed accordingly to obtain the corresponding grasping point position and x, y, z axis directions in the scene point cloud. Then, based on the calibration relationship between the depth camera and the tool coordinate system, the corresponding grasping point position and x, y, z axis directions in the scene point cloud are transformed to the grasping point position and x, y, z axis directions in the tool coordinate system, thus obtaining the first candidate position and first candidate pose. After determining the grasping point position and x, y, z axis directions in the tool coordinate system, it is equivalent to determining the position that the center point of the robotic arm's gripper needs to reach as the grasping point position, and the x, y, z axes of the robotic arm in the tool coordinate system should reach this x, y, z axis direction, thereby performing workpiece grasping.

[0051] Step S103: Obtain the preset symmetry parameters corresponding to the target workpiece.

[0052] In this embodiment, after obtaining the first candidate grasping pose information based on template-based point cloud matching, in order to improve the workpiece grasping success rate of the robotic arm in complex grasping environments, the preset symmetry parameters corresponding to the target workpiece are further obtained, and the remaining candidate grasping poses (i.e., the second candidate grasping poses below) are determined based on the preset symmetry parameters.

[0053] The aforementioned preset symmetry parameters are parameters related to the symmetry properties of the target workpiece. For example, in one embodiment, the preset symmetry parameters corresponding to the target workpiece may include a preset central axis and a preset symmetry rotation angle corresponding to the preset central axis. The overlap between the target workpiece and its original position after rotating around the preset central axis by the preset symmetry rotation angle reaches a preset overlap threshold.

[0054] In specific implementation, the aforementioned preset central axis can be determined by calculating the point cloud template based on a central axis calculation strategy. For example, Principal Component Analysis (PCA) can be used to analyze the point cloud template to obtain the directions of each principal component, and then one of these principal component directions can be determined as the central axis. Optionally, in some embodiments, the aforementioned central axis can also be manually set in the point cloud template. For example, taking a wrench as the target workpiece, the preset central axis can be perpendicular to the wrench handle and pass through the center of the wrench point cloud template (i.e., the average value of all points in the point cloud template).

[0055] In practical implementation, when the target workpiece is symmetrical, the axis of symmetry of the target workpiece can be determined as the central axis. If the target workpiece has multiple axes of symmetry, the axis of symmetry that is most advantageous for the robotic arm to grasp the workpiece can be taken as the central axis. For example, the direction of the axis of symmetry that is most advantageous for the robotic arm to grasp the workpiece can be closest to the Z-axis direction in the template grasping pose corresponding to the point cloud template.

[0056] The aforementioned preset symmetrical rotation angle can include all feasible angles from which the target workpiece, after rotating around the preset central axis, can achieve a preset overlap threshold with the target workpiece. Taking a cube-shaped workpiece as an example, for instance... Figure 2 As shown, the preset central axis is the axis of symmetry of the cube workpiece, and the preset rotation angle can include 90 degrees, 180 degrees, and 270 degrees. It can be understood that after the cube workpiece is rotated 90 degrees, 180 degrees, and 270 degrees around the preset central axis, it will completely coincide with the cube workpiece in its original position.

[0057] Similarly, if the target workpiece is a hexagonal prism, and if the preset central axis is the axis of symmetry perpendicular to the hexagonal surface, then the preset symmetry rotation angle of the hexagonal prism workpiece can be 60 degrees, 120 degrees, 180 degrees, 240 degrees, and 300 degrees.

[0058] The aforementioned preset overlap threshold can be determined based on actual conditions, for example, 80% or 90%. It is understood that a higher preset overlap threshold indicates a higher degree of overlap between the target workpiece obtained after rotation and the target workpiece before rotation. When the preset overlap threshold is 100%, the target workpiece is a symmetrical workpiece, and the preset central axis of the target workpiece is the axis of symmetry of the symmetrical target workpiece.

[0059] Specifically, the aforementioned overlap degree can be calculated based on the degree of overlap between points in the point cloud template corresponding to the target workpiece before and after rotation. For example, in a specific implementation, the preset overlap degree threshold can be a preset overlap degree threshold corresponding to a preset overlap radius. Based on this, in the rotated point cloud template, it is determined whether there are any points in the point cloud template before rotation within the range formed by taking each point as the center and the preset overlap radius as the radius. Then, the proportion of the number of points in the point cloud template before rotation within the range to the total number of points in the point cloud template is calculated, thus obtaining the overlap degree.

[0060] Step S104: Transform the first candidate grasping pose information according to the preset symmetry parameters to obtain the second candidate grasping pose information.

[0061] In this embodiment, after obtaining the preset symmetry parameters corresponding to the target workpiece, the obtained first candidate grasping pose information can be transformed based on the preset symmetry parameters to obtain the second candidate grasping pose information.

[0062] In one embodiment, when the preset symmetry parameters include the preset central axis and the preset symmetry rotation angle corresponding to the preset central axis, the first candidate grasping pose information can be angularly rotated and transformed according to the preset central axis and the preset symmetry rotation angle to obtain the second candidate grasping pose information.

[0063] Specifically, the pose corresponding to the first candidate grasping pose information can be rotated around a preset central axis by a preset symmetrical rotation angle, and the resulting pose is the second candidate grasping pose information.

[0064] In practice, the rotation method around the preset central axis can be determined based on whether the Z-axis in the first candidate gripping pose information of the target workpiece and the preset central axis are in the same or similar directions. When the preset central axis and the Z-axis in the first candidate gripping pose information are not in the same or similar directions, rotating around the preset central axis means rotating around the central axis. When the preset central axis and the Z-axis in the first candidate gripping pose information are in the same or similar directions, rotating around the preset central axis can mean keeping the Z-axis unchanged and rotating around the central axis.

[0065] When there are multiple preset symmetrical rotation angles, multiple corresponding second candidate grasping pose information can be obtained.

[0066] For example, when the target workpiece and the preset central axis are as follows Figure 2 As shown in the example, taking the preset central axis and the Z-axis in the first candidate grasping pose information as being in the same direction, the second candidate grasping pose information obtained by rotating around the central axis while keeping the Z-axis unchanged is as follows: Figure 3 As shown in the figure. If the gripping pose of the robotic arm gripper is as shown in the first candidate gripping pose information of the target workpiece, then the gripping pose of the robotic arm gripper is as shown in the figure. Figure 3 As shown in the upper half of the figure, the three second candidate gripping poses obtained by rotating the gripping pose corresponding to the first candidate gripping pose information by 90 degrees, 180 degrees, and 270 degrees around the preset central axis of the target workpiece are shown in the lower half of the figure. At this time, the Z-axis of the three second candidate gripping pose information is consistent with the Z-axis of the first candidate posture in the first candidate gripping pose information, and the gripping point position is consistent with the gripping point position in the first candidate gripping position information (i.e., the aforementioned first candidate position), except that the x and y axes in the posture are rotated by a preset symmetrical rotation angle.

[0067] In some implementations, there may be multiple preset center axes. In this case, for each target preset center axis, the target second candidate grasping pose information corresponding to the target preset center axis can be determined, so that the target second candidate grasping pose information corresponding to each preset center axis can be determined as the second candidate grasping pose information.

[0068] Step S105: Determine the target grasping pose information from the first candidate grasping pose information and the second candidate grasping pose information based on the preset grasping pose determination strategy.

[0069] The aforementioned preset grasping pose determination strategy is used to select the optimal grasping pose from multiple grasping poses; the aforementioned target grasping pose information is used to determine the target position and target posture corresponding to the robotic arm grasping the target workpiece. In this embodiment, after determining the first candidate grasping pose information and the second candidate grasping pose information, the target grasping pose information can be determined based on the preset grasping pose determination strategy.

[0070] For example, in some implementations, such as Figure 4A As shown, step S104 may specifically include steps S201 and S202.

[0071] Step S201: Based on the preset grasping interference judgment strategy, grasping interference judgment is performed on each candidate grasping pose information in the first candidate grasping pose information and the second candidate grasping pose information respectively to obtain the grasping interference judgment result corresponding to each candidate grasping pose information.

[0072] The aforementioned preset grasping interference judgment strategy is used to determine whether environmental interference will occur when the robotic arm grasps the target workpiece according to the corresponding candidate grasping pose information, thereby obtaining a grasping interference judgment result. In this embodiment, after obtaining the first candidate grasping pose information and the second candidate grasping pose information, grasping interference judgment is performed on each candidate grasping pose information to determine feasible grasping pose information.

[0073] In one embodiment, such as Figure 4B As shown, step S201 may specifically include steps S301 to S303.

[0074] Step S301: Obtain the preset gripping parameters corresponding to the target workpiece.

[0075] The preset gripping parameters are parameters related to the gripping process of the target workpiece. In specific implementation, these preset gripping parameters may include the size data of the target workpiece itself and the gripping parameters of the robotic arm gripper when gripping the target workpiece.

[0076] For example, in one embodiment, the preset gripping parameters include initial gripper stroke data, gripper size data, and workpiece size data of the target workpiece. The initial gripper stroke data is the initial distance between the two grippers after the robotic arm reaches the pose corresponding to the candidate gripping pose information and before it begins closing the grip; the gripper size data is the size data of the gripper itself, which may include, for example, length l, width w, and height h. Figure 5 As shown, the robotic arm can... Figure 5 As shown in the left half, the robotic arm gripper is mounted on the end of the robotic arm. The gripper's dimensions can be obtained as follows: Figure 5 As shown in the right half of the diagram, the z-axis direction in the tool coordinate system is perpendicular to the forward direction of the end effector. The workpiece size data refers to the dimensions of the target workpiece itself. This size data can be calculated and determined based on the point cloud template of the target workpiece. The target workpiece size data may include the workpiece size data near the gripping point position in the template gripping pose corresponding to the point cloud template. This workpiece size data near the gripping point position may include the distance data between the workpiece gripping point position and each surface of the target workpiece. In some embodiments, this distance data may be obtained through manual modification or setting.

[0077] Step S302: For each of the first candidate grasping pose information and the second candidate grasping pose information, determine the grasping interference space of the target workpiece corresponding to the candidate grasping pose information according to the preset grasping parameters and the candidate grasping pose information.

[0078] The aforementioned grasping interference space is the spatial range that would obstruct the gripper of the robotic arm when it grasps a workpiece. Under the grasping pose corresponding to the candidate grasping pose information, the gripper of the robotic arm grasps the target workpiece according to the target workpiece grasping parameters (initial gripper stroke data, gripper size data). Therefore, the grasping interference space corresponding to each candidate grasping pose information is determined according to the candidate grasping pose information and the preset grasping parameters.

[0079] In one embodiment, if the preset gripping parameters include the initial gripper stroke data, gripper size data, and the target workpiece size data, then as follows: Figure 6A As shown, step S302 may specifically include steps S401 to S403.

[0080] Step S401: Determine the target gripping space required for the gripper to close and grip the target workpiece based on the initial gripper stroke data, the gripper size data, the workpiece size data, and the candidate gripping pose information.

[0081] In this embodiment, the candidate gripping pose information is used to guide the robotic arm gripper to reach the desired pose. In specific implementation, the target gripping space includes the space enclosed by the gripper and the gripping point of the target workpiece before the robotic arm gripper begins to close the gripper to grip the target workpiece after reaching the gripping pose corresponding to the candidate gripping position information.

[0082] Optionally, in some implementations, to ensure that the gripper can successfully reach the candidate grasping pose information, the target grasping space may further include part or all of the space occupied by the robotic arm gripper itself when reaching the pose.

[0083] For example, exemplarily, such as Figure 6B As shown in the figure, when the target workpiece is a wrench, if the x, y, z axis directions and the position of the gripping point in a candidate gripping pose of the target workpiece are as follows: Figure 6B As shown in the figure (the figure shows the point cloud data of the plane where the gripping point of the wrench is located), the pose of the gripper when it reaches the candidate gripping pose is as follows: Figure 6B As shown. At this time, the target grab space is Figure 6C In the upper middle part of the diagram, the space enclosed by the dashed line includes the space occupied by the gripper itself, as well as the space enclosed by the gripper and the target workpiece near the gripping point. This target gripping space can be calculated based on the gripper's length l, width w, height h, and initial gripper stroke s.

[0084] In one specific implementation, the coordinates of each edge point in the gripper can be determined based on the candidate gripping pose information (in the tool coordinate system), the calibration relationship between the gripper position and the origin of the tool coordinate system, the initial gripper stroke data, and the gripper size data. Then, based on the coordinates of each edge point in the gripper, the gripping point position of the target workpiece, and the distance data between the target workpiece and its surfaces, the target gripping space can be determined. For example, the target gripping space can be defined as a rectangular area with a gripper width of l, a length of 2×w+s, and a height of m (the distance m from the gripping point position to the workpiece surface in the opposite direction along the Z-axis).

[0085] Step S402: Determine the target workpiece space corresponding to the gripper gripping the target workpiece based on the workpiece size data, the gripper size data, and the candidate gripping pose information.

[0086] The aforementioned target workpiece space should be understood as the space occupied by the target workpiece itself; specifically, the target workpiece space can be the space occupied by the grasped portion of the target workpiece when the robotic arm gripper reaches the target workpiece in the pose corresponding to the candidate grasping pose information. For example... Figure 6CThe area shown by the dashed line in the lower part of the figure is the target workpiece space. The target workpiece space can be determined based on the gripping point position in the candidate gripping pose information, the gripper size data, and the target workpiece size data.

[0087] Step S403: Determine the grasping interference space based on the target grasping space and the target workpiece space.

[0088] In some implementations, such as Figure 6D The grasping interference space shown in the figure is achieved by... Figure 6C The target gripping space shown is obtained by subtracting the target workpiece space occupied by the workpiece itself. This gripping interference space indicates a spatial range in which interference may occur.

[0089] In some implementations, the space obtained by subtracting the target workpiece space occupied by the workpiece itself from the target grasping space can be supplemented or removed according to the actual situation, thereby obtaining the grasping interference space.

[0090] Step S303: Determine the grasping interference judgment result corresponding to the candidate grasping pose information based on the number of interference points included in the grasping interference space, and obtain the grasping interference judgment result corresponding to each candidate grasping pose information respectively.

[0091] In this embodiment, the interference points are points in the grasping environment that do not belong to the target workpiece itself. The number of interference points in the grasping interference space reflects the obstruction present in that space; therefore, the grasping interference judgment result corresponding to the candidate grasping pose information is determined based on the number of interference points in the grasping interference space.

[0092] In the specific implementation process, after determining the grasping interference space, the total number of points belonging to the grasping environment located in the grasping interference space can be determined, that is, the interference point data. Then, it can be determined whether the interference point data is greater than the preset interference number threshold. If so, the grasping interference judgment result corresponding to the candidate grasping pose information is that there is interference; if not, the grasping interference judgment result is that there is no interference.

[0093] The points in the aforementioned capture environment can be obtained based on the scene point cloud corresponding to the target workpiece obtained in step S101.

[0094] Step S202: Determine the target grasping pose information from the first candidate grasping pose information and the second candidate grasping pose information based on the grasping interference judgment results.

[0095] In this embodiment, after obtaining the grasping interference judgment results corresponding to each candidate grasping pose information, the target grasping pose information can be determined based on the grasping interference judgment results.

[0096] Specifically, candidate grasping pose information whose grasping interference judgment result is that there is no interference can be determined as target grasping pose information.

[0097] In practical implementation, when multiple candidate grasping pose information results in no interference, the candidate grasping pose information can be further selected. For example, in one embodiment, such as Figure 7 As shown, step S202 may further include steps S501 to S503.

[0098] Step S501: Determine the current candidate grasping pose information from the first candidate grasping pose information and the second candidate grasping pose information based on the grasping interference judgment result.

[0099] In the specific implementation process, among all candidate grasping pose information composed of the first candidate grasping pose information and the second candidate grasping pose information, all candidate grasping pose information whose grasping interference judgment result is that there is no interference can be determined as the current candidate grasping pose information.

[0100] Step S502: When there is only one candidate grasping pose information, the current candidate grasping pose information is determined as the target grasping pose information.

[0101] In this embodiment, when there is only one candidate grasping pose information, the current candidate grasping pose information is determined as the target grasping pose information.

[0102] Step S503: When there are multiple current candidate grasping pose information, determine the pose transformation amount corresponding to each current candidate grasping pose information based on the initial pose corresponding to the robotic arm, and determine the target grasping pose information from each current candidate grasping pose information based on the pose transformation amount corresponding to each current candidate grasping pose information.

[0103] The initial pose of the robotic arm is the pose of the robotic arm before grasping. In some embodiments, since the robotic arm restores to a default pose (e.g., restores to the home point) after each grasping, the initial pose can be the default pose of the robotic arm.

[0104] The aforementioned pose transformation amounts reflect the pose transformation amounts of the robotic arm from the initial pose to the current candidate grasping pose; essentially, they are the pose transformation amounts between two poses. In practical implementation, the pose transformation amounts between the initial pose and the current candidate grasping pose can be determined according to a preset pose transformation amount measurement strategy, thereby identifying the current candidate grasping pose information with the smallest pose transformation amount as the target grasping pose information.

[0105] The aforementioned pose transformation measurement strategy is used to measure the calculation method of the pose transformation between two poses. For example, in one specific embodiment, it may specifically include position transformation and attitude transformation. The position transformation can be determined by the transformation between two positions, while the attitude transformation can be determined based on the rotation transformation between two attitudes. For example, the sum of the rotation angles along the x, y, and z axes from attitude a to attitude b can be determined as the rotation transformation. In the specific comparison of pose transformations, the position transformation and attitude transformation can be compared separately.

[0106] In the specific implementation process, if the second candidate grasping pose information is obtained by angular rotation transformation of the first candidate grasping pose information, the position transformation amount between the obtained second candidate grasping pose information and the first candidate grasping position information is small. That is, the position transformation amounts corresponding to the first candidate grasping pose information and the second candidate grasping pose information are similar or equal. Therefore, the above pose transformation amount can only be considered.

[0107] In summary, this application provides a workpiece gripping pose determination method, apparatus, computer device, and storage medium. The method includes: acquiring a scene point cloud corresponding to a target workpiece to be gripped in a workpiece placement area, and determining a first candidate gripping pose information based on a preset point cloud template corresponding to the target workpiece and the scene point cloud; acquiring preset symmetry parameters corresponding to the target workpiece; transforming the first candidate gripping pose information according to the preset symmetry parameters to obtain a second candidate gripping pose information; and determining a target gripping pose information from the first candidate gripping pose information and the second candidate gripping pose information based on a preset gripping pose determination strategy; wherein the target gripping pose information is used to determine the target position and target posture corresponding to the robotic arm gripping the target workpiece. In this embodiment, after obtaining the grasping pose information (first candidate grasping pose information) of the target workpiece based on template matching, the preset symmetry parameters corresponding to the target workpiece are further obtained. Then, the grasping pose information can be transformed according to the preset symmetry parameters to obtain other grasping pose information. On this basis, the target grasping pose information that conforms to the complex grasping scenario can be determined from all the obtained grasping pose information based on the preset grasping pose determination strategy. This is beneficial to improving the robustness of the robotic arm grasping, improving the success rate of the robotic arm grasping workpieces in complex grasping scenarios, and thus improving the grasping efficiency of the workpiece.

[0108] This application also provides a workpiece gripping pose determination device, which is used to perform the steps in any of the aforementioned embodiments of the workpiece gripping pose determination method. Specifically, please refer to... Figure 8 , Figure 8 This illustration shows a structural schematic diagram of a workpiece gripping pose determination device 800 according to an embodiment of this application. The workpiece gripping pose determination device 800 specifically includes a first acquisition unit 801, a first determination unit 802, a second acquisition unit 803, a pose transformation unit 804, and a second determination unit 805.

[0109] The first acquisition unit 801 is used to acquire the scene point cloud corresponding to the target workpiece to be grabbed in the workpiece placement area.

[0110] The first determining unit 802 is used to determine the first candidate grasping pose information based on the preset point cloud template corresponding to the target workpiece and the scene point cloud.

[0111] The second acquisition unit 803 is used to acquire the preset symmetry parameters corresponding to the target workpiece;

[0112] The pose transformation unit 804 is used to transform the first candidate grasping pose information according to the preset symmetry parameters to obtain the second candidate grasping pose information.

[0113] The second determining unit 805 is used to determine the target grasping pose information from the first candidate grasping pose information and the second candidate grasping pose information based on a preset grasping pose determination strategy; wherein, the target grasping pose information is used to determine the target position and target posture corresponding to the robotic arm grasping the target workpiece.

[0114] In some embodiments of this application, the second determining unit 805 may be specifically used to perform grasping interference judgment on each candidate grasping pose information in the first candidate grasping pose information and the second candidate grasping pose information based on a preset grasping interference judgment strategy, and obtain grasping interference judgment results corresponding to each candidate grasping pose information; wherein, the preset grasping interference judgment strategy is used to determine whether environmental interference will occur when the robotic arm grasps the target workpiece according to the corresponding candidate grasping pose information; and to determine the target grasping pose information from the first candidate grasping pose information and the second candidate grasping pose information according to each grasping interference judgment result.

[0115] In some embodiments of this application, the second determining unit 805 may be specifically used for,

[0116] Obtain preset grasping parameters corresponding to the target workpiece; for each candidate grasping pose information in the first candidate grasping pose information and the second candidate grasping pose information, determine the grasping interference space of the target workpiece corresponding to the candidate grasping pose information according to the preset grasping parameters and the candidate grasping pose information; determine the grasping interference judgment result corresponding to the candidate grasping pose information based on the number of interference points included in the grasping interference space, and obtain the grasping interference judgment result corresponding to each candidate grasping pose information respectively.

[0117] In some embodiments of this application, the preset gripping parameters include initial gripper stroke data, gripper size data, and workpiece size data of the target workpiece; the gripping interference space of the target workpiece corresponding to the candidate gripping pose information is determined based on the preset gripping parameters and the candidate gripping pose information; the second determining unit 805 may be specifically used to: determine the target gripping space required for the gripper to close and grip the target workpiece based on the initial gripper stroke data, the gripper size data, the workpiece size data, and the candidate gripping pose information; determine the target workpiece space corresponding to the gripper gripping the target workpiece based on the workpiece size data, the gripper size data, and the candidate gripping pose information; and determine the gripping interference space based on the target gripping space and the target workpiece space.

[0118] In some embodiments of this application, the second determining unit 805 may be specifically used to: determine the current candidate grasping pose information from the first candidate grasping pose information and the second candidate grasping pose information based on the grasping interference judgment result; when there is only one current candidate grasping pose information, determine the current candidate grasping pose information as the target grasping pose information; when there are multiple current candidate grasping pose information, determine the pose transformation amount corresponding to each current candidate grasping pose information based on the initial pose corresponding to each current candidate grasping pose information and the robotic arm, and determine the target grasping pose information from each current candidate grasping pose information based on the pose transformation amount corresponding to each current candidate grasping pose information.

[0119] In some embodiments of this application, the preset symmetry parameters corresponding to the target workpiece include a preset central axis and a preset symmetry rotation angle corresponding to the preset central axis. After the target workpiece rotates around the preset central axis by the preset symmetry rotation angle, the overlap between the target workpiece and the target workpiece in its original position reaches a preset overlap threshold. The pose transformation unit 804 can be specifically used to perform angular rotation transformation on the first candidate grasping pose information according to the preset central axis and the preset symmetry rotation angle to obtain the second candidate grasping pose information.

[0120] It should be noted that those skilled in the art can clearly understand that the specific implementation process of the above-mentioned workpiece gripping pose determination device and each unit can be referred to the corresponding description in the foregoing method embodiments. For the sake of convenience and brevity, it will not be repeated here.

[0121] The aforementioned workpiece gripping pose determination device can be implemented as a computer program, which can, for example, Figure 9 It runs on the computer equipment shown.

[0122] Please see Figure 9 , Figure 9 This is a schematic block diagram of a computer device provided in an embodiment of this application. The computer device 900 can be a smartphone, tablet, personal computer, server, or other device used to control a robotic arm. See also... Figure 9 The computer device 900 includes a processor 902, a memory, and a network interface 905 connected via a device bus 901, wherein the memory may include a storage medium 903 and internal memory 904.

[0123] The storage medium 903 may store an operating system 9031 and a computer program 9032. When the computer program 9032 is executed, it enables the processor 902 to execute a workpiece gripping pose determination method.

[0124] The processor 902 provides computing and control capabilities to support the operation of the entire computer device 900.

[0125] The internal memory 904 provides an environment for the operation of the computer program 9032 in the storage medium 903. When the computer program 9032 is executed by the processor 902, the processor 902 can execute the workpiece gripping pose determination method.

[0126] The network interface 905 is used for network communication, such as providing data transmission. Those skilled in the art will understand that... Figure 9 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device 900 to which the present application is applied. The specific computer device 900 may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0127] The processor 902 is used to run the computer program 9032 stored in the memory to implement the workpiece gripping pose determination method disclosed in the embodiments of this application.

[0128] Those skilled in the art will understand that Figure 9 The embodiments of the computer device shown do not constitute a limitation on the specific configuration of the computer device. In other embodiments, the computer device may include more or fewer components than illustrated, or combine certain components, or have different component arrangements. For example, in some embodiments, the computer device may include only memory and a processor. In such embodiments, the structure and function of the memory and processor are different from those shown. Figure 9 The embodiments shown are consistent and will not be repeated here.

[0129] It should be understood that in the embodiments of this application, the processor 902 may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0130] In another embodiment of this application, a computer-readable storage medium is provided. This computer-readable storage medium may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. The computer-readable storage medium stores a computer program, wherein when executed by a processor, the computer program implements the workpiece gripping pose determination method disclosed in the embodiments of this application.

[0131] Those skilled in the art will readily understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0132] In the embodiments provided in this application, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Units with the same function may be grouped into one unit. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, or it may be an electrical, mechanical, or other form of connection.

[0133] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of this application, depending on actual needs.

[0134] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0135] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a backend server, or a network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), magnetic disks, or optical disks.

[0136] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for determining the gripping posture of a workpiece, characterized in that, include: Obtain the scene point cloud corresponding to the target workpiece to be grabbed in the workpiece placement area; The first candidate grasping pose information is determined based on the preset point cloud template corresponding to the target workpiece and the scene point cloud; Obtain the preset symmetry parameters corresponding to the target workpiece; The first candidate grasping pose information is transformed according to the preset symmetry parameters to obtain the second candidate grasping pose information; Based on a preset grasping pose determination strategy, the target grasping pose information is determined from the first candidate grasping pose information and the second candidate grasping pose information; wherein, the target grasping pose information is used to determine the target position and target posture corresponding to the robotic arm grasping the target workpiece; The preset symmetry parameters corresponding to the target workpiece include a preset central axis and a preset symmetry rotation angle corresponding to the preset central axis. After the target workpiece rotates around the preset central axis by the preset symmetry rotation angle, the degree of overlap between the target workpiece and the target workpiece in its original position reaches a preset overlap threshold. The preset symmetry rotation angle includes all feasible angles from which the target workpiece can achieve the preset overlap threshold after rotating around the preset central axis. The preset overlap threshold is the preset overlap threshold corresponding to the preset overlap radius. In the rotated point cloud template, it is determined whether there are any points in the point cloud template before rotation within the range formed by taking each point as the center of the sphere and the preset overlap radius as the radius. The proportion of the number of points in the point cloud template before rotation within the range to the total number of points in the point cloud template is calculated to obtain the preset overlap threshold. The step of transforming the first candidate grasping pose information according to the preset symmetry parameters to obtain the second candidate grasping pose information includes: The first candidate grasping pose information is subjected to angular rotation transformation based on the preset central axis and the preset symmetrical rotation angle to obtain the second candidate grasping pose information; The pose corresponding to the first candidate grasping pose information is rotated around the preset central axis by the preset symmetrical rotation angle to obtain the pose as the second candidate grasping pose information. The rotation method around the preset central axis is determined based on whether the Z-axis in the first candidate grasping pose information of the target workpiece and the preset central axis are in the same or similar directions. If the preset central axis and the Z-axis in the first candidate grasping pose information are not in the same or similar directions, then rotating around the preset central axis refers to rotating around the central axis. If the preset central axis and the Z-axis in the first candidate grasping pose information are in the same or similar directions, then rotating around the preset central axis means keeping the Z-axis unchanged while rotating around the central axis.

2. The method according to claim 1, characterized in that, The target grasping pose information is determined from the first candidate grasping pose information and the second candidate grasping pose information based on the preset grasping pose determination strategy, including: Based on a preset grasping interference judgment strategy, grasping interference judgment is performed on each candidate grasping pose information in the first candidate grasping pose information and the second candidate grasping pose information respectively to obtain the grasping interference judgment result corresponding to each candidate grasping pose information; wherein, the preset grasping interference judgment strategy is used to determine whether environmental interference will occur when the robotic arm grasps the target workpiece according to the corresponding candidate grasping pose information. The target grasping pose information is determined from the first candidate grasping pose information and the second candidate grasping pose information based on the grasping interference judgment results.

3. The method according to claim 2, characterized in that, The preset grasping interference judgment strategy performs grasping interference judgment on each candidate grasping pose information in the first candidate grasping pose information and the second candidate grasping pose information respectively, to obtain the grasping interference judgment result corresponding to each candidate grasping pose information, including: Obtain the preset gripping parameters corresponding to the target workpiece; For each of the first candidate grasping pose information and the second candidate grasping pose information, the grasping interference space of the target workpiece corresponding to the candidate grasping pose information is determined according to the preset grasping parameters and the candidate grasping pose information. Based on the number of interference points included in the grasping interference space, the grasping interference judgment result corresponding to the candidate grasping pose information is determined, and the grasping interference judgment result corresponding to each candidate grasping pose information is obtained.

4. The method according to claim 3, characterized in that, The preset gripping parameters include initial gripper stroke data, gripper size data, and workpiece size data of the target workpiece; determining the gripping interference space of the target workpiece corresponding to the candidate gripping pose information based on the preset gripping parameters and the candidate gripping pose information includes: The target gripping space required for the gripper to close and grip the target workpiece is determined based on the initial gripper stroke data, the gripper size data, the workpiece size data, and the candidate gripping pose information. The target workpiece space corresponding to the gripper gripping the target workpiece is determined based on the workpiece size data, the gripper size data, and the candidate gripping pose information. The grasping interference space is determined based on the target grasping space and the target workpiece space.

5. The method according to claim 2, characterized in that, The step of determining the target grasping pose information from the first candidate grasping pose information and the second candidate grasping pose information based on the grasping interference judgment results includes: Based on the grasping interference judgment result, the current candidate grasping pose information is determined from the first candidate grasping pose information and the second candidate grasping pose information; When there is only one candidate grasping pose information, the current candidate grasping pose information is determined as the target grasping pose information; When there are multiple current candidate grasping pose information, the pose transformation amount corresponding to each current candidate grasping pose information is determined based on the initial pose corresponding to each current candidate grasping pose information and the robot arm, and the target grasping pose information is determined from each current candidate grasping pose information based on the pose transformation amount corresponding to each current candidate grasping pose information.

6. A workpiece gripping posture determination device, characterized in that, include: The first acquisition unit is used to acquire the scene point cloud corresponding to the target workpiece to be grabbed in the workpiece placement area; The first determining unit is used to determine the first candidate grasping pose information based on the preset point cloud template corresponding to the target workpiece and the scene point cloud. The second acquisition unit is used to acquire the preset symmetry parameters corresponding to the target workpiece; The pose transformation unit is used to transform the first candidate grasping pose information according to the preset symmetry parameters to obtain the second candidate grasping pose information. The second determining unit is used to determine the target grasping pose information from the first candidate grasping pose information and the second candidate grasping pose information based on a preset grasping pose determination strategy; wherein, the target grasping pose information is used to determine the target position and target posture corresponding to the robotic arm grasping the target workpiece; The preset symmetry parameters corresponding to the target workpiece include a preset central axis and a preset symmetry rotation angle corresponding to the preset central axis. After the target workpiece rotates around the preset central axis by the preset symmetry rotation angle, the degree of overlap between the target workpiece and the target workpiece in its original position reaches a preset overlap threshold. The preset symmetry rotation angle includes all feasible angles from which the target workpiece can achieve the preset overlap threshold after rotating around the preset central axis. The preset overlap threshold is the preset overlap threshold corresponding to the preset overlap radius. In the rotated point cloud template, it is determined whether there are any points in the point cloud template before rotation within the range formed by taking each point as the center of the sphere and the preset overlap radius as the radius. The proportion of the number of points in the point cloud template before rotation within the range to the total number of points in the point cloud template is calculated to obtain the preset overlap threshold. The step of transforming the first candidate grasping pose information according to the preset symmetry parameters to obtain the second candidate grasping pose information includes: The first candidate grasping pose information is subjected to angular rotation transformation based on the preset central axis and the preset symmetrical rotation angle to obtain the second candidate grasping pose information; The pose corresponding to the first candidate grasping pose information is rotated around the preset central axis by the preset symmetrical rotation angle to obtain the pose as the second candidate grasping pose information. The rotation method around the preset central axis is determined based on whether the Z-axis in the first candidate grasping pose information of the target workpiece and the preset central axis are in the same or similar directions. If the preset central axis and the Z-axis in the first candidate grasping pose information are not in the same or similar directions, then rotating around the preset central axis refers to rotating around the central axis. If the preset central axis and the Z-axis in the first candidate grasping pose information are in the same or similar directions, then rotating around the preset central axis means keeping the Z-axis unchanged while rotating around the central axis.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 5.

Images