Workpiece grabbing collision interference detection method, system and storage medium

Abstract

The present invention relates to the field of machine vision technology. Specifically, it relates to a method, system and storage medium for detecting collision interference of workpiece grasping. The method includes: obtaining a point cloud of a robot gripper at an initial position, and recording it as a first point cloud G of the robot gripper; obtaining a first point cloud S of a production scene through a 3D camera; obtaining a second point cloud of the production scene and a second point cloud of the robot gripper; obtaining a depth map of the point cloud of the production scene and a depth map of the point cloud of the robot gripper; obtaining an area where both the depth map Im_s and the depth map Im_g have projected depth values; obtaining the number of pixel points whose depth values ​​of the depth map Im_g are greater than the depth values ​​of the depth map Im_s; in response to the number of pixel points being greater than a preset collision threshold, determining that the robot gripper will collide with other objects when grasping a workpiece to be grasped. The method of the present invention can accurately determine whether the robot gripper will collide with an object in the production scene when it moves to the grasping point.

Description

Technical Field

[0001] The present invention generally relates to the field of machine vision technology, and more specifically, to a workpiece grasping collision interference detection method, system and storage medium. Background Art

[0002] With the reduction of labor costs and the development of robotics and computer vision technology, the proportion of robots used in the production process will become higher and higher. 3D vision-guided robot grasping is a key technology for realizing intelligent robot production. At present, due to the complexity of the production environment and the instability of existing 3D vision recognition technology, it is impossible to accurately predict whether the gripper will touch other workpieces during the grasping process. As a result, the gripper will touch other workpieces during the grasping process, resulting in collisions, unstable grasping, and even robot failure and shutdown. Summary of the invention

[0003] In order to solve one or more of the above technical problems, the present invention proposes to predict whether there will be collision interference between the gripper and the workpiece by predicting the point cloud of the robot gripper at the grasping point and combining it with the point cloud of the industrial production scene, so as to adjust the movement trajectory of the gripper to avoid collision between the gripper and other workpieces during the grasping process. To this end, the present invention provides solutions in the following aspects.

[0004] In a first aspect, the present invention provides a method for detecting collision interference in grasping a workpiece, comprising: obtaining a point cloud of a robot gripper at an initial position, and recording it as a first point cloud G of the robot gripper; obtaining a first point cloud S of a production scene through a 3D camera, and obtaining an internal parameter matrix of the 3D camera; the first point cloud of the production scene is composed of point cloud data of all workpieces near a grasping point; removing the point cloud of a workpiece to be grasped from the first point cloud of the production scene to obtain a second point cloud of the production scene ;According to the first point cloud G of the robot gripper, the point cloud of the robot gripper at the grasping point in the coordinate system of the 3D camera is obtained and recorded as the second point cloud of the robot gripper ; According to the internal parameter matrix, the second point cloud of the production scene Project it onto the plane image to obtain the production scene point cloud depth map Im_s; according to the internal parameter matrix, the second point cloud of the robot gripper Projecting onto a plane image, obtaining a robot gripper point cloud depth map Im_g; the direction of projection is from top to bottom; obtaining an area where both the production scene point cloud depth map Im_s and the robot gripper point cloud depth map Im_g have projected depth values; comparing the depth value of the production scene point cloud depth map Im_s with the depth value of the robot gripper point cloud depth map Im_g at each pixel point in the area, thereby obtaining the number of pixels where the depth value of the robot gripper point cloud depth map Im_g is greater than the depth value of the production scene point cloud depth map Im_s ; Response to the number of pixels If the collision value is greater than a preset collision threshold P, it is determined that the robot gripper will collide with other objects when grasping the workpiece to be grasped.

[0005] In one embodiment, the second point cloud of the production scene Projection to a plane image includes: traversing the point cloud For all the points in the image, the pixel position of each point on the projected image is calculated in combination with the camera intrinsic parameter matrix; the corresponding depth value is obtained according to the z-axis coordinate of each point.

[0006] In one embodiment, the robot gripper second point cloud Projection to a plane image includes: traversing the point cloud All points in the image are combined with the camera intrinsic parameter matrix to calculate the pixel position of each point on the projected image; based on the obtained pixel position, the point cloud with the largest Z-axis coordinate value at each pixel position is obtained The points in the graph are based on the point cloud with the largest Z-axis coordinate value at each pixel position. The Z-axis coordinate value of the point in is used to calculate the depth value of the pixel point at the corresponding pixel position.

[0007] In one embodiment, removing the point cloud of the workpiece to be grasped from the first point cloud of the production scene includes: obtaining a 3D model M of the workpiece to be grasped; aligning the 3D model M with the first point cloud of the production scene, and obtaining a pose matrix of the workpiece to be grasped under the 3D camera. ; Obtain the matrix B composed of the vertex coordinates of the bounding box of the 3D model M at the initial position of the robot gripper, and combine it with the pose matrix A matrix consisting of the vertex coordinates of the bounding box at the grabbing point in the coordinate system of the 3D camera is obtained, and the point cloud inside the bounding box is deleted from the first point cloud of the production scene.

[0008] In one embodiment, the step of obtaining the matrix composed of the vertex coordinates of the bounding box at the grasping point in the coordinate system of the 3D camera comprises: adding the matrix B composed of the vertex coordinates of the bounding box at the initial position of the robot gripper to the pose matrix Multiply.

[0009] In one embodiment, obtaining the point cloud of the robot gripper at the grasping point in the 3D camera coordinate system includes: comparing the first point cloud of the robot gripper with the pose matrix Multiply.

[0010] In one embodiment, the method further comprises: responding to the number of the pixel points If the collision value is less than a preset collision threshold P, it is determined that the robot gripper will not collide with other objects when grasping the workpiece to be grasped.

[0011] In one embodiment, the method further includes: in response to determining that the robot gripper will collide with other objects when grasping the workpiece to be grasped, issuing an alarm.

[0012] In a second aspect, the present invention provides a workpiece grasping collision interference detection system, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the workpiece grasping collision interference detection method of the present invention is implemented.

[0013] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored therein, wherein the computer program is used to implement the workpiece grasping collision interference detection method of the present invention when executed by a processor.

[0014] The technical effect of the present invention is as follows: the workpiece grasping collision interference detection method of the present invention obtains the point cloud depth map corresponding to the production scene at the grasping point and the point cloud depth map corresponding to the robot gripper in the same coordinate system, and compares the two point cloud depth maps. In the area where both point cloud depth maps have projected depth values, the magnitude relationship of the depth values ​​of the points in the two point cloud depth maps is compared to determine whether the robot gripper is located under the object in the production scene when it moves to the grasping point, thereby accurately determining whether the robot gripper will collide with the object in the production scene when it moves to the grasping point. In order to avoid signal interference, by setting a collision threshold, only when the number of pixel points whose depth values ​​in the depth map Im_g are greater than the depth value of the depth map Im_s is greater than the threshold, is it determined that the robot gripper is located under the object in the production scene when it moves to the grasping point.

[0015] Furthermore, when it is determined that a collision will occur, an alarm is issued to remind on-site staff so that the staff can adjust the movement of the robot gripper or pause the robot in time, thereby avoiding collision between the gripper and other workpieces during the grasping process. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0017] Figure 1 is a flow chart schematically illustrating a method for detecting a workpiece grasping collision interference according to an embodiment of the present invention;

[0018] Figure 2 is a point cloud depth map schematically illustrating a production scene according to an embodiment of the present invention;

[0019] Figure 3 is a schematic diagram showing a point cloud depth map of a robot gripper according to an embodiment of the present invention;

[0020] Figure 4 The second point cloud of the production scene is schematically shown in an embodiment of the present invention. Flowchart of the method for projecting onto a planar image;

[0021] Figure 5 The second point cloud of the robot gripper is schematically shown in the embodiment of the present invention. Flowchart of the method for projecting onto a planar image;

[0022] Figure 6 is a flowchart schematically illustrating a method for removing a point cloud of a workpiece to be grasped from a first point cloud of a production scene according to an embodiment of the present invention;

[0023] Figure 7 Schematic diagram of a workpiece grasping collision interference detection system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0025] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0026] Workpiece grabbing collision interference detection method embodiment:

[0027] like Figure 1 As shown, the workpiece grasping collision interference detection method of the present invention includes:

[0028] S101. Obtain the first point cloud of the robot gripper and the first point cloud of the production scene, specifically: obtain the point cloud of the robot gripper at the initial position, and record it as the first point cloud G of the robot gripper; obtain the first point cloud S of the production scene through a 3D camera, and obtain the internal parameter matrix of the 3D camera; the first point cloud of the production scene is composed of point cloud data of all workpieces near the grasping point.

[0029] When the 3D camera obtains the first point cloud of the production scene, the location where the 3D camera is installed and the direction of shooting can be determined as needed. In this embodiment, the 3D camera is installed just above the workpiece to be grasped, and shoots from top to bottom.

[0030] S102, obtaining a second point cloud of the production scene and a second point cloud of the robot gripper, specifically: removing the point cloud of the workpiece to be grasped from the first point cloud of the production scene to obtain the second point cloud of the production scene ;According to the first point cloud G of the robot gripper, the point cloud of the robot gripper at the grasping point in the coordinate system of the 3D camera is obtained and recorded as the second point cloud of the robot gripper .

[0031] When grabbing a workpiece, the robot gripper moves along a preset trajectory and starts the grabbing action when it reaches the end point of the trajectory. The grabbing point refers to the end point of the robot gripper's moving trajectory.

[0032] Since the workpiece to be grasped must be in contact with the robot gripper before it can be grasped, when predicting whether the robot gripper will collide with an object other than the workpiece to be grasped, the point cloud of the workpiece to be grasped should be removed from the first point cloud of the production scene, and the remaining point cloud data of the first point cloud of the production scene and the point cloud data of the robot gripper are used to determine whether the robot gripper will collide with an object other than the workpiece to be grasped. Since the point cloud of the production scene is acquired by shooting with a 3D camera, the coordinate system corresponding to the point cloud of the production scene is the 3D camera coordinate system, and the coordinate system corresponding to the point cloud of the robot gripper is not the 3D camera coordinate system. The point clouds of the two are not in the same coordinate system. By converting the point cloud of the robot gripper to the 3D camera coordinate system, the point clouds of the two are made to be point clouds in the same coordinate system, so as to accurately judge the spatial position relationship between the robot gripper and the objects in the production scene.

[0033] S103, obtaining the production scene point cloud depth map and the robot gripper point cloud depth map, specifically: according to the internal reference matrix, the second point cloud of the production scene Project it onto the plane image to obtain the production scene point cloud depth map Im_s; according to the internal parameter matrix, the second point cloud of the robot gripper Project it onto the plane image to obtain the robot gripper point cloud depth map Im_g; the projection direction is from top to bottom.

[0034] The projection described in this embodiment refers to the projection from the horizontal plane where the 3D camera is located from top to bottom. The projection described in other embodiments may also refer to other forms of projection. And the second point cloud of the robot gripper Projection can be obtained as Figure 2 The production scene point cloud depth map Im_s shown in the figure and Figure 3 The point cloud depth map Im_g of the robot gripper is shown.

[0035] Second point cloud for production scene And the second point cloud of the robot gripper The projection method is to traverse the production scene point cloud separately And the second point cloud of the robot gripper The point cloud of the production scene is generated by calculating the coordinates of each point and the intrinsic parameter matrix of the 3D camera. And the second point cloud of the robot gripper Projected onto the plane in the Z-axis direction.

[0036] S104, obtaining areas where both the production scene point cloud depth map Im_s and the robot gripper point cloud depth map Im_g have projected depth values.

[0037] In this implementation, having a projected depth value means that the depth value is greater than zero.

[0038] If both the depth map Im_s and the depth map Im_g have projected depth values ​​in a certain area, it means that the robot gripper and the object in the production scene may have intersecting and overlapping parts in the spatial position area corresponding to the area. By further determining the Z-axis coordinates of the robot gripper and the object in the production scene in the area, it can be determined whether the two have intersecting and overlapping parts.

[0039] S105, compare the depth value of the production scene point cloud depth map Im_s with the depth value of the robot gripper point cloud depth map Im_g, specifically: compare the depth value of the production scene point cloud depth map Im_s with the depth value of the robot gripper point cloud depth map Im_g at each pixel point in the area, so as to obtain the number of pixel points whose depth value of the robot gripper point cloud depth map Im_g is greater than the depth value of the production scene point cloud depth map Im_s .

[0040] Since the 3D camera shoots from top to bottom, the farther away from the 3D camera, that is, the lower the height, the greater the depth value. In the area, the depth value of the depth map Im_g is greater than the depth value of the depth map Im_s, indicating that when the robot gripper moves to the grasping point, a part of the robot gripper is directly below the object in the production scene. The robot gripper moves from top to bottom when grasping the workpiece. If the robot gripper moves directly below the object in the production scene, it will inevitably collide with the object in the production scene.

[0041] S106, determining whether the robot gripper will collide with other objects when grasping the workpiece to be grasped, specifically: responding to the number of pixel points If the collision value is greater than a preset collision threshold P, it is determined that the robot gripper will collide with other objects when grasping the workpiece to be grasped.

[0042] By setting the collision threshold P, noise interference can be prevented and more accurate detection can be achieved.

[0043] The workpiece grasping collision interference detection method of the present invention obtains the point cloud depth map corresponding to the production scene at the grasping point and the point cloud depth map corresponding to the robot gripper in the same coordinate system, and compares the two point cloud depth maps. In the area where both point cloud depth maps have projected depth values, the size relationship of the depth values ​​of the points in the two point cloud depth maps is compared to determine whether the robot gripper is located under the object in the production scene when it moves to the grasping point, thereby accurately determining whether the robot gripper will collide with the object in the production scene when it moves to the grasping point. In order to avoid signal interference, by setting a collision threshold, only when the number of pixel points whose depth value of the depth map Im_g is greater than the depth value of the depth map Im_s is greater than the threshold, it is determined that the robot gripper is located under the object in the production scene when it moves to the grasping point.

[0044] In the above embodiment, it is mentioned that the point cloud of the production scene is calculated based on the coordinates of each point and the internal parameter matrix of the 3D camera. Projected onto the plane in the Z-axis direction, in one embodiment, the following can be used: Figure 4 Specifically, as shown in Figure 4 As shown, the second point cloud of the production scene Projecting onto a flat image includes:

[0045] S401, obtaining the pixel position of each point in the production scene point cloud on the projected image, specifically: traversing the point cloud All points in the image are combined with the camera intrinsic parameter matrix to calculate the pixel position of each point on the projected image;

[0046] Assuming point cloud The point cloud coordinates of a point in is ( , , ), then the calculation expressions for the number of rows m and columns n corresponding to the point on the projected image are as follows:

[0047] (1)

[0048] (2)

[0049] In the formula, , , and These are the intrinsic parameters of the 3D camera.

[0050] S402 : Obtain the corresponding depth value of each point according to its z-axis coordinate.

[0051] Point Cloud Different points in the image have different z-axis coordinates, so their depth values ​​are different.

[0052] In the above embodiment, it is mentioned that the second point cloud of the robot gripper is calculated based on the coordinates of each point and the internal parameter matrix of the 3D camera. Projected onto the plane in the Z-axis direction, in one embodiment, the following can be used: Figure 5 Specifically, as shown in Figure 5 As shown, the second point cloud of the robot gripper Projecting onto a flat image includes:

[0053] S501, obtaining the pixel position of each point in the second point cloud of the robot gripper on the projection image, specifically: traversing the point cloud All points in the image are combined with the camera intrinsic parameter matrix to calculate the pixel position of each point on the projected image;

[0054] Assuming point cloud The point cloud coordinates of a point in is ( , , ), then the calculation expressions for the number of rows m and columns n corresponding to the point on the projected image are as follows:

[0055] (3)

[0056] (4)

[0057] In the formula, , , and These are the intrinsic parameters of the 3D camera.

[0058] S502, obtaining the depth value of the pixel point at each pixel position on the projection image, specifically: obtaining the point cloud with the largest Z-axis coordinate value at each pixel position according to the obtained pixel position The points in the graph are based on the point cloud with the largest Z-axis coordinate value at each pixel position. The Z-axis coordinate value of the point in is used to calculate the depth value of the pixel point at the corresponding pixel position.

[0059] Since there are multiple points in the gripper point cloud projected to the same image pixel position, it is necessary to determine the point with the largest z value and project it onto the image to obtain the projected image Im_g.

[0060] In the above embodiment, it is mentioned that the point cloud of the workpiece to be grasped is removed from the first point cloud of the production scene to obtain the second point cloud of the production scene. In one embodiment, the following method can be used: Figure 6 The method shown is implemented, specifically, as Figure 6 As shown, removing the point cloud of the workpiece to be grasped from the first point cloud of the production scene includes:

[0061] S601, obtaining a 3D model M of a workpiece to be grasped;

[0062] Point cloud data of the workpiece to be grasped can be accurately obtained based on the 3D model M of the workpiece to be grasped.

[0063] S602, obtaining the pose matrix of the workpiece to be grasped under the 3D camera, specifically: aligning the 3D model M with the first point cloud of the production scene, and obtaining the pose matrix of the workpiece to be grasped under the 3D camera ;

[0064] S603, obtaining the bounding box of the 3D model M and deleting the point cloud inside the bounding box from the first point cloud of the production scene, specifically: obtaining the matrix B composed of the vertex coordinates of the bounding box of the 3D model M at the initial position of the robot gripper, and combining it with the pose matrix A matrix consisting of the vertex coordinates of the bounding box at the grabbing point in the coordinate system of the 3D camera is obtained, and the point cloud inside the bounding box is deleted from the first point cloud of the production scene.

[0065] The bounding box algorithm is an algorithm for finding the optimal bounding space of a discrete point set. The basic idea is to use a geometric body (called a bounding box) with a slightly larger volume and simpler characteristics to approximately replace a complex geometric object.

[0066] The step of obtaining a matrix composed of vertex coordinates of the bounding box at the grasping point in the coordinate system of the 3D camera includes: combining the matrix B composed of the vertex coordinates of the bounding box at the initial position of the robot gripper with the posture matrix Multiply, the expression is as follows:

[0067] = (5)

[0068] In the formula, Represents the matrix composed of the vertex coordinates of the bounding box at the grab point in the coordinate system of the 3D camera.

[0069] The above embodiments mention obtaining the point cloud of the robot gripper at the grasping point in the 3D camera coordinate system. In one embodiment, obtaining the point cloud of the robot gripper at the grasping point in the 3D camera coordinate system includes: comparing the first point cloud of the robot gripper with the pose matrix Multiply, the expression is as follows:

[0070] = (6)

[0071] In the formula, represents the point cloud of the robot gripper at the grasping point in the 3D camera coordinate system, Represents the first point cloud of the robot gripper.

[0072] The first point cloud of the robot gripper and the pose matrix After multiplication, the point cloud data of the robot gripper in the 3D camera coordinate system can be obtained.

[0073] The above embodiment is only used to judge whether a robot gripper will collide when grasping a workpiece under certain working conditions. In another embodiment, it also includes: If the collision value is less than a preset collision threshold P, it is determined that the robot gripper will not collide with other objects when grasping the workpiece to be grasped.

[0074] If an alarm is issued when it is detected that the robot gripper will collide with the workpiece to be grasped, the on-site staff can be reminded to adjust the movement trajectory of the robot gripper. Therefore, in another embodiment, it also includes: in response to determining that the robot gripper will collide with other objects when grasping the workpiece to be grasped, an alarm is issued.

[0075] When it is determined that a collision will occur, an alarm is issued to alert on-site staff so that they can adjust the movement of the robot gripper or pause the robot in time to avoid collision between the gripper and other workpieces during the grasping process.

[0076] Workpiece grabbing collision interference detection system embodiment:

[0077] The present invention also provides a workpiece grasping collision interference detection system. Figure 7As shown, the workpiece grasping collision interference detection system includes a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the workpiece grasping collision interference detection method according to the first aspect of the present invention is implemented.

[0078] The workpiece grasping collision interference detection system also includes other components familiar to those skilled in the art, such as a communication bus and a communication interface, whose configuration and functions are known in the art and will not be described in detail here.

[0079] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, apparatus or device. For example, a computer-readable storage medium may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random-Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high-bandwidth memory HBM (High-Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device. Any application or module described in the present invention may be implemented using computer-readable / executable instructions that may be stored or otherwise maintained by such a computer-readable medium.

[0080] Computer readable storage medium embodiment:

[0081] The present invention further provides a readable storage medium, wherein the storage medium stores a computer program, and the computer program is used to implement the workpiece grasping collision interference detection method described in the first aspect of the present invention when executed by a processor.

[0082] Computer-readable storage media may be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or components, or any combination thereof. More specific examples (non-exhaustive examples) of computer-readable storage media may include, for example, an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, device, or device.

[0083] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0084] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A method for detecting collision interference of workpiece grasping, characterized in that: include: Obtain the point cloud of the robot gripper at the initial position and record it as the first point cloud G of the robot gripper; obtain the first point cloud S of the production scene through the 3D camera, and obtain the intrinsic parameter matrix of the 3D camera; The first point cloud of the production scene consists of point cloud data of all workpieces near the gripping point; The second point cloud of the production scene is obtained by removing the point cloud of the workpiece to be grasped from the first point cloud of the production scene , including: obtaining the 3D model M of the workpiece to be grasped; aligning the 3D model M with the first point cloud of the production scene, and obtaining the pose matrix of the workpiece to be grasped under the 3D camera ; Get the matrix B composed of the vertex coordinates of the bounding box of the 3D model M at the initial position of the robot gripper, and combine it with the pose matrix Get the matrix composed of the vertex coordinates of the bounding box at the grasping point in the coordinate system of the 3D camera, including: the matrix B composed of the vertex coordinates of the bounding box at the initial position of the robot gripper and the pose matrix Multiply; and delete the point cloud inside the bounding box from the first point cloud of the production scene; According to the first point cloud G of the robot gripper, the point cloud of the robot gripper at the grasping point in the coordinate system of the 3D camera is obtained and recorded as the second point cloud of the robot gripper ; According to the internal parameter matrix, the second point cloud of the production scene Project it onto the plane image to obtain the production scene point cloud depth map Im_s; according to the internal parameter matrix, the second point cloud of the robot gripper Project it onto the plane image to obtain the robot gripper point cloud depth map Im_g; the projection direction is from top to bottom; Get the area where both the production scene point cloud depth map Im_s and the robot gripper point cloud depth map Im_g have projected depth values; The depth value of the production scene point cloud depth map Im_s is compared with the depth value of the robot gripper point cloud depth map Im_g at each pixel point in the area, so as to obtain the number of pixels where the depth value of the robot gripper point cloud depth map Im_g is greater than the depth value of the production scene point cloud depth map Im_s. ; Responding to the number of pixels Greater than the preset collision threshold , determine whether the robot gripper will collide with other objects when grasping the workpiece to be grasped.

2. The workpiece grasping collision interference detection method according to claim 1, characterized in that: The second point cloud of the scene will be produced Projecting onto a flat image includes: Traversing the point cloud All points in the image are combined with the camera intrinsic parameter matrix to calculate the pixel position of each point on the projected image; Get the corresponding depth value of each point according to its z-axis coordinate.

3. The workpiece grasping collision interference detection method according to claim 1, characterized in that: The second point cloud of the robot gripper Projecting onto a flat image includes: Traversing the point cloud All points in the image are combined with the camera intrinsic parameter matrix to calculate the pixel position of each point on the projected image; Get the point cloud with the largest Z-axis coordinate value at each pixel position based on the obtained pixel position The points in the graph are based on the point cloud with the largest Z-axis coordinate value at each pixel position. The Z-axis coordinate value of the point in is used to calculate the depth value of the pixel point at the corresponding pixel position.

4. The workpiece grabbing collision interference detection method according to claim 1, characterized in that: The step of obtaining the point cloud of the robot gripper at the grasping point in the 3D camera coordinate system comprises: comparing the first point cloud of the robot gripper with the position matrix Multiply.

5. The workpiece grabbing collision interference detection method according to claim 1, characterized in that: Also includes: In response to the number of pixels If the collision value is less than a preset collision threshold P, it is determined that the robot gripper will not collide with other objects when grasping the workpiece to be grasped.

6. The workpiece grasping collision interference detection method according to any one of claims 1 to 5, characterized in that: Also includes: In response to determining that the robot gripper will collide with other objects when grasping the workpiece to be grasped, an alarm is issued.

7. A workpiece grasping collision interference detection system, comprising a memory and a processor, wherein a computer program is stored in the memory, characterized in that: When the computer program is executed by the processor, the workpiece grasping collision interference detection method according to any one of claims 1 to 6 is implemented.

8. A computer-readable storage medium, wherein a computer program is stored in the storage medium, characterized in that: The computer program is used to implement the workpiece grasping collision interference detection method described in any one of claims 1 to 6 when executed by a processor.

Images