Tire point cloud completion method, assembly method, control device and storage medium

By using motion parameters and initial reference point cloud when the robot grabs the tire, the tire point cloud is completed in real time, the problem of incomplete tire point cloud is solved, the accuracy and efficiency of vehicle tire assembly is improved, and the impact risk is reduced.

CN112700536BActive Publication Date: 2024-11-15GUANGDONG MIDEA WHITE HOME APPLIANCE TECH INNOVATION CENT CO LTD +1
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202011626444.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-12-31
Publication Date
2024-11-15
Estimated Expiration
2040-12-31

AI Technical Summary

Technical Problem

In the prior art, tire point clouds are incomplete due to tire movement and low black reflectivity, which affects the determination of collision risk between the vehicle and the tire.

Method used

During the process of the robot grabbing the tire, the real-time reference point cloud is determined using the robot's motion parameters and the initial reference point cloud, and converted it into a point cloud based on the depth camera coordinate system to complete the tire point cloud.

Benefits of technology

It improves the accuracy of the tire point cloud, enhances the efficiency and accuracy of vehicle tire assembly, and reduces the probability of collision between the vehicle and the tire.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN112700536B_ABST
    Figure CN112700536B_ABST
Patent Text Reader

Abstract

The present application discloses a tire point cloud completion method, an assembly method, a control device and a storage medium, wherein the tire point cloud completion method comprises: determining the tire point cloud of the target tire in real time during the robot grabbing the target tire; determining the real-time reference point cloud using the robot's motion parameters and the initial reference point cloud; wherein the initial reference point cloud is determined when the tire model is assembled to the robot and the robot moves to the initial position; and completing the tire point cloud using the real-time reference point cloud. By the above method, the accuracy of the tire point cloud acquired in real time can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of vehicle assembly, and in particular to a tire point cloud completion method, an assembly method, a control device and a storage medium. Background Art

[0002] The production line uses a robot to put the spare tire in the trunk of a dynamic vehicle to achieve active collision avoidance. The basic idea is to determine whether there is a collision risk by calculating the minimum distance between the two point clouds in real time based on the point cloud of the tire and the point cloud of the vehicle. An important prerequisite for this method is that the tire point cloud and the body point cloud must be complete.

[0003] Among them, for tires, since tires are moving, and are generally black with low reflectivity, and sometimes robots block the camera, the tire point cloud acquired in real time may be incomplete. Therefore, in order to calculate the minimum distance between the tire and the vehicle body, how to obtain a complete tire point cloud becomes a key issue. Summary of the invention

[0004] To solve the above problems, the present application provides a tire point cloud completion method, an assembly method, a control device and a storage medium, which can improve the accuracy of the tire point cloud acquired in real time.

[0005] A technical solution adopted in the present application is: to provide a tire point cloud completion method, the method comprising: determining the tire point cloud of the target tire in real time during the process of the robot grasping the target tire; determining the real-time reference point cloud using the robot's motion parameters and the initial reference point cloud; wherein the initial reference point cloud is determined when the tire model is assembled to the robot and the robot moves to the initial position; and completing the tire point cloud using the real-time reference point cloud.

[0006] The steps of obtaining the initial reference point cloud include: obtaining a first point cloud of the tire model and obtaining a second point cloud of the fixture; wherein the robot clamps the target tire using the fixture; combining the first point cloud and the second point cloud to obtain a third point cloud; after the fixture is fixedly connected to the robot, determining the position and posture of the third point cloud to obtain the initial reference point cloud.

[0007] Among them, the first point cloud and the second point cloud are point clouds based on the spatial coordinate system; the posture of the third point cloud is determined to obtain the initial reference point cloud, including: obtaining the motion parameters of the robot and the first conversion parameters of the robot end posture; the initial reference point cloud is determined by the third point cloud, the initial motion parameters of the robot, and the first conversion parameters.

[0008] The real-time reference point cloud is determined by using the motion parameters of the robot and the initial reference point cloud, including: determining the real-time reference point cloud by the third point cloud, the initial motion parameters of the robot, and the first conversion parameters.

[0009] The first point cloud and the second point cloud are combined to obtain a third point cloud, including: combining the first point cloud and the second point cloud according to the clamping position of the clamp and the tire model to obtain the third point cloud.

[0010] Among them, the real-time reference point cloud is a point cloud based on the robot coordinate system, and the tire point cloud is a point cloud based on the depth camera coordinate system; before using the real-time reference point cloud to complete the tire point cloud, it also includes: converting the real-time reference point cloud into a point cloud based on the depth camera coordinate system.

[0011] Among them, converting the real-time reference point cloud into a point cloud based on a depth camera coordinate system includes: obtaining the position information of the robot's reference point in the depth camera coordinate system; determining a second conversion parameter of the robot coordinate system and the depth camera coordinate system according to the position information; and using the second conversion parameter to convert the real-time reference point cloud into a point cloud based on the depth camera coordinate system.

[0012] Another technical solution adopted in the present application is: providing a vehicle tire assembly method, the method comprising: obtaining a depth image of a scene in which a robot assembles a target tire onto a vehicle; using the depth image to construct a vehicle point cloud model of the vehicle and a tire point cloud model of the target tire; using the above method to complete the tire point cloud model; and performing corresponding processing according to the spatial position of the vehicle point cloud model and the completed tire point cloud model.

[0013] Among them, corresponding processing is performed according to the spatial positions of the vehicle point cloud model and the completed tire point cloud model, including: adjusting the running trajectory of the tire according to the spatial positions of the vehicle point cloud model and the completed tire point cloud model to assemble the tire to the vehicle.

[0014] Another technical solution adopted in the present application is: to provide a vehicle assembly control device, which includes: a depth camera for collecting depth images of a scene in which a robot assembles a target tire to a vehicle; a memory for storing program data; and a controller connected to the depth camera and the memory for executing the program data in the above-mentioned method.

[0015] Another technical solution adopted by the present application is to provide a computer-readable storage medium, wherein the computer-readable storage medium stores program data, and when the stored data is executed by the controller, it is used to implement the method as described above.

[0016] The tire point cloud completion method provided by the present application includes: determining the tire point cloud of the target tire in real time during the process of the robot grabbing the target tire; determining the real-time reference point cloud using the robot's motion parameters and the initial reference point cloud; wherein the initial reference point cloud is determined when the tire model is assembled to the robot and the robot moves to the initial position; and completing the tire point cloud using the real-time reference point cloud. Through the above method, when the tire point cloud is missing, it can be completed using a predetermined tire model, thereby improving the accuracy of the tire point cloud acquired in real time. Furthermore, due to the improved accuracy of the tire point cloud, the efficiency and accuracy of vehicle tire assembly are further improved, the failure rate is reduced, and the probability of collision between the vehicle and the tire is effectively reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0018] Figure 1 It is a flow chart of an embodiment of a tire point cloud completion method provided by the present application;

[0019] Figure 2 is a flow chart of an embodiment of a method for obtaining an initial point cloud;

[0020] Figure 3 It is a schematic diagram of the combination of the clamp and the tire;

[0021] Figure 4 yes Figure 1 Schematic diagram of the process of step 13;

[0022] Figure 5 It is a schematic flow chart of an embodiment of a vehicle tire assembly method provided by the present application;

[0023] Figure 6 It is a structural schematic diagram of an embodiment of a vehicle assembly control device provided by the present application;

[0024] Figure 7 It is a structural schematic diagram of an embodiment of a computer-readable storage medium provided by the present application. DETAILED DESCRIPTION

[0025] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It will be appreciated that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for ease of description, only some but not all structures related to the present application are shown in the drawings. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the art without making creative work are within the scope of protection of the present application.

[0026] The terms "first", "second", etc. in this application are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0027] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0028] See also Figure 1 , Figure 1 : is a flow chart of an embodiment of a tire point cloud completion method provided by the present application, the method comprising:

[0029] Step 11: While the robot is grasping the target tire, the tire point cloud of the target tire is determined in real time.

[0030] It can be understood that this embodiment is applied to assembling tires on dynamic vehicles on a production line. The production line mainly uses a conveyor mechanism (large plate) to carry the vehicle, and the movement of the conveyor mechanism drives the movement of the vehicle. Then the tire is grabbed by a robot and assembled into the trunk of the vehicle.

[0031] In this embodiment, a depth camera is disposed above the conveying mechanism to acquire a depth image.

[0032] Specifically, the point data set of the product appearance surface obtained by measuring instruments in reverse engineering is also called a point cloud. In the embodiment of the present application, step 11 can be specifically as follows:

[0033] (1) Obtain the contour information of the depth image.

[0034] Among them, edge feature points of the depth image may be extracted first, and then affine transformation may be performed on the extracted edge feature points to obtain contour information of the depth image.

[0035] When performing contour extraction, depth difference detection or gradient difference detection can be used:

[0036] Depth difference detection mainly obtains the difference in depth values ​​between adjacent pixels. If the difference is large, it is considered that the two pixels are discontinuous in three-dimensional space.

[0037] Gradient difference detection mainly obtains the difference in depth gradient values ​​between adjacent pixels. If the difference is large, it is considered that the two points may be located on the contour line of the depth image (ie, not in the same plane).

[0038] After contour extraction, the edge feature points need to be affine transformed. Affine transformation refers to a linear transformation of a vector space followed by a translation in geometry to transform it into another vector space.

[0039] (2) The camera coordinate system is converted into the space coordinate system.

[0040] First, the parameters of the depth camera and the position information of the depth camera based on the spatial coordinate system are obtained; and the coordinate conversion parameters are determined according to the parameters and the position information.

[0041] Among them, the parameters of the depth camera are mainly the camera intrinsic parameters. A camera intrinsic parameter matrix is ​​given below:

[0042]

[0043] Among them, fx and fy are focal lengths, which are generally equal to each other, x0 and y0 are the coordinates of the principal point (relative to the imaging plane), and s is the coordinate axis tilt parameter, which is 0 in an ideal case.

[0044] Among them, the position of the depth camera in the spatial coordinate system, that is, the external parameters of the camera, that is, the position of the camera in space and the rotation parameters relative to each axis.

[0045] Then, using the above external parameters and internal parameters, we can get the coordinate transformation parameters, that is, the transformation matrix from the camera coordinate system to the space coordinate system. Then, we use the coordinate transformation parameters to transform the point cloud in step (1) to get the point cloud in the space coordinate system.

[0046] For example:

[0047]

[0048] Where [Xc, Yc, Zc] TRepresents the camera coordinates, [Xw, Yw, Zw] T It represents the world coordinates of the object, R is the rotation matrix (3 degrees of freedom), T is the translation matrix, and the two form a 3×4 matrix, which is the external parameter matrix of the camera.

[0049] Step 12: Determine the real-time reference point cloud using the robot's motion parameters and the initial reference point cloud.

[0050] The initial reference point cloud is determined when the tire model is assembled to the robot and the robot moves to the initial position. Figure 2 As shown, Figure 2 1 is a flow chart of an embodiment of a method for obtaining an initial point cloud. The specific process of obtaining an initial reference point cloud is as follows:

[0051] Step 21: Acquire a first point cloud of the tire model and acquire a second point cloud of the fixture; wherein the robot clamps the target tire using the fixture.

[0052] The robot end is provided with a first flange, one end of the clamp is provided with a second flange, the first flange and the second flange are fixed, and the other end of the clamp is used to clamp the tire. It can be understood that since the tire has a hollow shaft, the clamp can be an expansion type, that is, the clamp extends into the tire axis and then expands outward to clamp the tire.

[0053] The first point cloud of the tire model and the second point cloud of the fixture can be acquired by using a high-precision 3D scanner, which has higher accuracy.

[0054] Step 22: Combine the first point cloud and the second point cloud to obtain a third point cloud.

[0055] like Figure 3 As shown, Figure 3 It is a schematic diagram of the combination of the clamp and the tire.

[0056] Specifically, step 22 may be: combining the first point cloud and the second point cloud according to the clamping position of the fixture and the tire model to obtain a third point cloud. Since the point cloud of the fixture and the point cloud of the tire model are known, the relative positional relationship between the two is determined according to the clamping position of the fixture and the tire model, that is, the third point cloud formed by combining the two point clouds can be determined.

[0057] Step 23: After the fixture is fixedly connected to the robot, determine the pose of the third point cloud to obtain the initial reference point cloud.

[0058] It can be understood that each robot has a conversion relationship between the world coordinates of the robot end and the robot coordinates, and in this embodiment, the conversion relationship is defined as the first conversion parameter. Therefore, by obtaining the third point cloud and the first conversion parameter, the third point cloud and the first conversion parameter are calculated to obtain the point cloud of the third point cloud relative to the robot coordinate system, that is, the initial reference point cloud.

[0059] Among them, the motion parameters of the robot and the first conversion parameters of the robot end posture are obtained; and the initial reference point cloud is determined by the third point cloud, the initial motion parameters of the robot, and the first conversion parameters.

[0060] Among them, the robot's motion parameters include the joint angle of each joint.

[0061] In one embodiment, the corresponding first conversion parameter can be obtained according to the specific model of the robot. For example, when using a KUKA robot, it can be extracted from the official simulation software KUKA SIM PRO. The current joint angle obtained in real time from the working robot is substituted into the simulation conversion matrix, which is the current simulation conversion matrix of the robot (i.e., the first conversion parameter).

[0062] In another embodiment, the calibration can also be completed by the robot coordinates and the world coordinates of the robot end. Specifically, a calibration block is provided, and the world coordinates of multiple calibration points on the calibration block are calibrated by three-dimensional scanning, shooting depth images, etc. Then the robot end is controlled to move to these calibration points, and the first conversion parameters of the world coordinate system and the robot coordinate system are established through the corresponding relationship.

[0063] Specifically, the first conversion parameter may be embodied in the form of a conversion matrix, and the point cloud in the robot coordinate system may be obtained by multiplying the point cloud in the world coordinate system by the conversion matrix.

[0064] Furthermore, the real-time point cloud is obtained in the following manner: a real-time reference point cloud is determined by the third point cloud, the real-time motion parameters of the robot, and the first conversion parameters.

[0065] It can be understood that the initial reference point cloud can be determined by the third point cloud and the first conversion parameters. Because the clamping state of the robot on the tire through the clamp is unchanged, the relative position between the initial reference point cloud and the robot is also unchanged, so the transformation of the tire point cloud is completely caused by the movement of the robot. Therefore, it is only necessary to combine the real-time motion parameters of the robot to determine the real-time reference point cloud by the third point cloud, the real-time motion parameters of the robot, and the first conversion parameters.

[0066] Step 13: Use the real-time reference point cloud to complete the tire point cloud.

[0067] Understandably, to complete the point cloud, we first need to make sure that the two point clouds are in the same coordinate system.

[0068] Since the tire point cloud obtained in step 11 is obtained by taking a photo with a depth camera, the default is the camera coordinate system. Therefore, it is necessary to convert the real-time reference point cloud into a point cloud based on the depth camera coordinate system.

[0069] like Figure 4 As shown, Figure 4 yes Figure 1 The flowchart of step 13 in the embodiment of the present invention may include:

[0070] Step 131: Obtain the position information of the robot's reference point in the depth camera coordinate system.

[0071] It can be understood that in this embodiment, the number of cameras is one, and the position and angle of the camera are fixed. Therefore, the position of the robot's reference point (robot coordinate origin) in the camera coordinate system can be determined.

[0072] Step 132: Determine a second conversion parameter of the robot coordinate system and the depth camera coordinate system according to the position information.

[0073] First, define: OXY is the robot coordinate system, O'X'Y' is the camera coordinate system. theta is the angle between the two coordinate systems. Assume that any point P has a coordinate in the robot coordinate system and also has a coordinate in the camera coordinate system.

[0074] OXY is the robot coordinate system, O'X'Y' is the camera coordinate system. The coordinate transformation relationship can be seen from the above figure:

[0075] x=x'*r*cos(theta)-y'*r*sin(theta)+x0;

[0076] y=x'*r*sin(theta)+y'*r*cos(theta)+y0;

[0077] Where r is the millimeter-pixel ratio (mm / pixel), that is, how many pixels are there in one millimeter, theta is the angle between the two coordinate systems, and (x0, y0) is the distance from the camera coordinate origin to the robot coordinate origin (that is, it can be determined by the position of the robot in the camera coordinate system).

[0078] Simplify the abstract formula and assume that:

[0079] a=r*cos(theta);

[0080] b = r*sin(theta);

[0081] c = x0;

[0082] d = y0;

[0083] get:

[0084] x=x'*a-y'*b+c;

[0085] y=x'*b+y'*a+d;

[0086] Obviously, to solve this equation, we need two sets of correspondences, that is, two sets of corresponding coordinate points. Suppose the two sets of coordinate points are as follows:

[0087] The first group: the camera coordinate point: (xImage1, yImage1) corresponds to the machine coordinate point: (xMachine1, yMachine1);

[0088] The second group: the camera coordinate point: (xImage2, yImage2) corresponds to the machine coordinate point: (xMachine2, yMachine2);

[0089] Then we can solve a, b, c, d as follows:

[0090] a=((xMachine1-xMachine2)*(xImage1-xImage2)+(yMachine1-yMachine2)*(yImage1-yImage2)) / ((xImage1-xImage2)*(xImage1-xImage2)+(yImage1-yImage2)*(yImage1-yImage2));

[0091] b=((yMachine1-yMachine2)*(xImage1-xImage2)-(xMachine1-xMachine2)*(yImage1-yImage2)) / ((xImage1-xImage2)*(xImage1-xImage2)+(yImage1-yImage2)*(yImage1-yImage2));

[0092] c=xMachine1-a*xImage1+b*yImage1;

[0093] d=yMachine1-b*xImage1-a*yImage1;

[0094] Therefore, the second conversion parameter of converting the pixel coordinates of any point on the image into the robot coordinates is obtained.

[0095] Step 133: Using the second conversion parameter, the real-time reference point cloud is converted into a point cloud based on the depth camera coordinate system.

[0096] It can be understood that the above-mentioned second conversion parameter is generally expressed in the form of a matrix. The initial reference point cloud is multiplied by the conversion matrix to obtain the real-time reference point cloud.

[0097] The method of this embodiment can use the real-time reference point cloud to complete the tire point cloud. On the one hand, the actual points in the tire point cloud can be supplemented, and on the other hand, the tire point cloud can be directly discarded and the real-time reference point cloud can be directly used as the tire point cloud.

[0098] Different from the prior art, the tire point cloud completion method provided in this embodiment includes: determining the tire point cloud of the target tire in real time during the process of the robot grabbing the target tire; determining the real-time reference point cloud using the robot's motion parameters and the initial reference point cloud; wherein the initial reference point cloud is determined when the tire model is assembled to the robot and the robot moves to the initial position; and completing the tire point cloud using the real-time reference point cloud. In the above manner, when the tire point cloud is missing, it can be completed using a predetermined tire model, thereby improving the accuracy of the tire point cloud acquired in real time. Furthermore, due to the improved accuracy of the tire point cloud, the efficiency and accuracy of vehicle tire assembly are further improved, the failure rate is reduced, and the probability of collision between the vehicle and the tire is effectively reduced.

[0099] See also Figure 5 , Figure 5 1 is a flow chart of an embodiment of a vehicle tire assembly method provided by the present application, the method comprising:

[0100] Step 51: Obtain a depth image of the scene in which the robot assembles the target tire onto the vehicle.

[0101] In one embodiment, vehicle assembly mainly refers to the process in which the vehicle moves on a predetermined track, in which the tire is clamped by a robot and assembled into the trunk of the vehicle (as a spare tire of the vehicle).

[0102] A conveying mechanism (large board) is arranged in the predetermined track, and the vehicle moves on the carrier along with the conveying mechanism. A depth camera is arranged above the large board, and the photographing range of the depth camera at least covers the vehicle, the manipulator and the tire.

[0103] Step 52: construct a vehicle point cloud model of the vehicle and a tire point cloud model of the target tire using the depth image.

[0104] Among them, the vehicle point cloud and the tire point cloud can be converted into two different point clouds respectively by performing contour extraction on the depth image.

[0105] Step 53: Complete the tire point cloud model.

[0106] The method for completing the vehicle point cloud may refer to the above-mentioned embodiment and will not be described in detail here.

[0107] Step 54: Perform corresponding processing according to the spatial positions of the vehicle point cloud model and the completed tire point cloud model.

[0108] In one embodiment, since the vehicle (moving with the carrier) and the tire (moving with the manipulator) are both in motion, it is necessary to obtain the vehicle point cloud and the tire point cloud in real time and perform tracking and judgment during the entire assembly process.

[0109] Optionally, the spatial distance between the vehicle point cloud and the tire point cloud can be obtained in real time, and the two point clouds can be collided. For example, when the spatial distance between the two point clouds is less than a set distance threshold, a response process is performed. The processing method may include: pausing the movement of the vehicle and the tire, or keeping the movement of the vehicle unchanged, adjusting the running track of the tire to prevent the tire from colliding with the vehicle, or issuing an alarm.

[0110] Different from the prior art, the vehicle tire assembly method provided in this embodiment includes: obtaining a depth image of a scene in which a robot assembles a target tire to a vehicle; using the depth image to construct a vehicle point cloud model of the vehicle and a tire point cloud model of the target tire; completing the tire point cloud model; and performing corresponding processing according to the spatial position of the vehicle point cloud model and the completed tire point cloud model. In the above manner, when the tire point cloud is missing, it can be completed using a predetermined tire model, thereby improving the accuracy of the tire point cloud acquired in real time. Furthermore, due to the improved accuracy of the vehicle point cloud, the efficiency and accuracy of vehicle tire assembly are further improved, the failure rate is reduced, and the probability of collision between the vehicle and the tire is effectively reduced.

[0111] See also Figure 6 , Figure 6 is a schematic structural diagram of an embodiment of a vehicle assembly control device provided in the present application. The vehicle assembly control device 60 includes a depth camera 61 , a memory 62 and a controller 63 .

[0112] The depth camera 61 is installed on the vehicle motion track and is mainly used to collect the depth image of the scene where the robot assembles the tire to the vehicle; the memory 62 is used to store program data; and the controller 63 is used to execute the program data to implement the following method:

[0113] When the robot grabs the target tire, the tire point cloud of the target tire is determined in real time; the real-time reference point cloud is determined using the robot's motion parameters and the initial reference point cloud; the initial reference point cloud is determined when the tire model is assembled to the robot and the robot moves to the initial position; the tire point cloud is completed using the real-time reference point cloud.

[0114] Acquire a depth image of a scene in which the robot assembles a target tire onto a vehicle; construct a vehicle point cloud model of the vehicle and a tire point cloud model of the target tire using the depth image; complete the tire point cloud model; and perform corresponding processing according to the spatial positions of the vehicle point cloud model and the completed tire point cloud model.

[0115] See also Figure 7 , Figure 7 1 is a schematic diagram of a computer-readable storage medium according to an embodiment of the present application. The computer-readable storage medium 70 stores program data 71. When the program data 71 is executed by the controller, it is used to implement the following method:

[0116] When the robot grabs the target tire, the tire point cloud of the target tire is determined in real time; the real-time reference point cloud is determined using the robot's motion parameters and the initial reference point cloud; the initial reference point cloud is determined when the tire model is assembled to the robot and the robot moves to the initial position; the tire point cloud is completed using the real-time reference point cloud.

[0117] Acquire a depth image of a scene in which the robot assembles a target tire onto a vehicle; construct a vehicle point cloud model of the vehicle and a tire point cloud model of the target tire using the depth image; complete the tire point cloud model; and perform corresponding processing according to the spatial positions of the vehicle point cloud model and the completed tire point cloud model.

[0118] In the several embodiments provided in this application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device implementation described above is only illustrative, for example, the division of the modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0119] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0120] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0121] The above description is only an implementation method of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made according to the description and drawings of the present application, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A tire point cloud completion method, characterized in that: The method comprises: During the process of the robot grabbing the target tire, determining the tire point cloud of the target tire in real time; The real-time reference point cloud is determined by using the motion parameters of the robot and the initial reference point cloud; wherein the step of obtaining the initial reference point cloud comprises: obtaining a first point cloud of a tire model, and obtaining a second point cloud of a fixture; wherein the robot clamps the target tire by using the fixture; combining the first point cloud and the second point cloud to obtain a third point cloud; after the fixture is fixedly connected to the robot, determining the position and posture of the third point cloud to obtain the initial reference point cloud; The tire point cloud is completed using the real-time reference point cloud.

2. The method according to claim 1, characterized in that The first point cloud and the second point cloud are point clouds based on a spatial coordinate system; The determining the pose of the third point cloud to obtain the initial reference point cloud includes: Acquire the motion parameters of the robot and the first conversion parameters of the end position of the robot; The initial reference point cloud is determined by the third point cloud, the initial motion parameters of the robot, and the first conversion parameters.

3. The method according to claim 2, characterized in that The method of using the motion parameters of the robot and the initial reference point cloud to determine the real-time reference point cloud comprises: The real-time reference point cloud is determined based on the third point cloud, the initial motion parameters of the robot, and the first conversion parameters.

4. The method according to claim 1, characterized in that: The combining the first point cloud and the second point cloud to obtain a third point cloud includes: The first point cloud and the second point cloud are combined according to the clamping positions of the clamp and the tire model to obtain a third point cloud.

5. The method according to claim 1, characterized in that The real-time reference point cloud is a point cloud based on the robot coordinate system, and the tire point cloud is a point cloud based on the depth camera coordinate system; Before the tire point cloud is completed by using the real-time reference point cloud, the method further includes: The real-time reference point cloud is converted into a point cloud based on the depth camera coordinate system.

6. The method according to claim 5, characterized in that The converting the real-time reference point cloud into a point cloud based on the depth camera coordinate system comprises: Obtaining position information of a reference point of the robot in the depth camera coordinate system; Determine a second conversion parameter of the robot coordinate system and the depth camera coordinate system according to the position information; The real-time reference point cloud is converted into a point cloud based on the depth camera coordinate system using the second conversion parameters.

7. A vehicle tire assembly method, characterized in that: The method comprises: Acquire a depth image of a scene in which the robot assembles a target tire onto a vehicle; constructing a vehicle point cloud model of the vehicle and a tire point cloud model of the target tire using the depth image; Completing the tire point cloud model using the method according to any one of claims 1 to 6; Corresponding processing is performed according to the spatial positions of the vehicle point cloud model and the completed tire point cloud model.

8. The method according to claim 7, characterized in that The corresponding processing according to the spatial position of the vehicle point cloud model and the completed tire point cloud model includes: According to the spatial positions of the vehicle point cloud model and the completed tire point cloud model, the running track of the tire is adjusted to assemble the tire to the vehicle.

9. A vehicle assembly control device, characterized in that: The control device comprises: A depth camera, used to collect depth images of the scene where the robot assembles the target tire to the vehicle; A memory for storing program data; A controller, connected to the depth camera and the memory, configured to execute the program data to implement the method according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program data, and when the stored data is executed by the controller, it is used to implement the method according to any one of claims 1 to 8.

Citation Information

Patent Citations

  • Joint training method for object recognition model, server and electrical equipment

    CN110991512A

  • Laser radar 3D real-time target detection method fusing multi-frame time sequence point cloud

    CN111429514A

  • Tire point cloud construction method, assembly method, control device and storage medium

    CN112700537A

  • Method and device for predictive encoding / decoding of a point cloud

    US20200258262A1