Robot vision positioning calibration method

Abstract

The invention discloses a robot vision positioning calibration method. The technical problems that a traditional robot vision calibration method is low in calibration efficiency and poor in precision are solved. By presetting the spatial geometry, the spatial position calibration in the motion range of the robot can be realized, so that errors such as installation and assembly are considered, and adverse effects on calibration accuracy are avoided. In addition, the robot tail end cuspidal point position is determined by adjusting the support, the cuspidal point position capturing precision and reliability can be effectively improved, and the image recognition error is reduced. Therefore, the corresponding relation between the robot motion coordinate system and the camera pixel coordinate system is established through the space geometry, and calibration between the camera and the tail end of the robot is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial robot control, and in particular to a robot vision positioning calibration method. Background Art

[0002] Industrial robots are multi-joint manipulators or multi-degree-of-freedom robots for the industrial field. They use their own power and control capabilities to achieve various functions. Industrial robots can accept human commands or act according to pre-programmed programs or principles based on artificial intelligence technology. They are composed of three basic parts: the main body, the drive system and the control system, and some also include walking mechanisms. Most industrial robots have 3 to 6 degrees of freedom of movement and can perform complex operating tasks.

[0003] Industrial robot vision calibration is an important part of robotics technology, which involves determining the internal and external parameters of the camera and the relative position relationship between the camera and the robot coordinate system. Vision calibration is crucial to improving the performance of the robot vision system, which can significantly improve positioning accuracy and stability, and optimize the robot's motion trajectory and path planning.

[0004] In the process of visual calibration, commonly used calibration techniques include hand-eye calibration. Hand-eye calibration aims to establish the relationship between the visual pixel coordinate system and the robot or axis group base coordinate system. Through calibration, the pixel coordinates of the object captured by the camera can be converted into a three-dimensional spatial position that the robot can understand and operate. Commonly used hand-eye calibration methods include the nine-point method and the checkerboard method. However, the above calibration methods are not directly related to the robot's motion coordinate system, the calibration accuracy is poor, and multiple calibrations are required, resulting in low calibration efficiency.

[0005] The information disclosed in this background technology section is only used to deepen the understanding of the background technology of the present disclosure, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art known to those skilled in the art. Summary of the invention

[0006] In view of at least one of the above technical problems, the present disclosure provides a robot vision positioning calibration method, aiming to solve the technical problems of low calibration efficiency and insufficient accuracy of traditional robot vision calibration methods.

[0007] According to one aspect of the present disclosure, a robot vision positioning calibration method is provided, which mainly includes the following steps: (1) Presetting a virtual space geometry for the tip motion of the robot end and setting the lengths of each side of the virtual space geometry; (2) Guide the robot's end point to move to a vertex of the virtual space geometric body, and record the end coordinates of the robot's end point at that position in the robot's motion coordinate system ( Xn ，Y n ，Z n ); (3) Adjusting the adjustment bracket until the target point on the top of the adjustment bracket coincides with the tip of the robot end; (4) Start the camera to take a picture of the target point and obtain the pixel position coordinates (x n ，y n ，z n ); (5) Repeat steps (2) to (4) to obtain the end coordinates and pixel position coordinates corresponding to the remaining vertices of the spatial geometric body; (6) Establish the correspondence between the pixel position coordinates (x, y, z) and the end coordinates (X, Y, Z), and based on this correspondence, obtain the coordinate value of any pixel position point within the range of the spatial geometric body in the robot motion coordinate system to complete the camera visual calibration.

[0008] In some embodiments of the present disclosure, the spatial geometric body includes at least four vertices, and the spatial range corresponding to the spatial geometric body matches the spatial motion range of the robot.

[0009] In some embodiments of the present disclosure, the spatial geometric body is a cuboid, and the length, width and height of the cuboid are integer multiples of the robot motion scale.

[0010] In some embodiments of the present disclosure, the camera is fixed at a corresponding position of the end of the robot.

[0011] In some embodiments of the present disclosure, in step (4), before taking pictures, the robot is guided to move to the picture-taking position, and the camera field of view is adjusted until the target is within the camera field of view.

[0012] In some embodiments of the present disclosure, the camera is a fixed camera that does not move with the robot end.

[0013] In some embodiments of the present disclosure, in step (6), a linear relationship between the pixel position coordinates (x, y, z) and the end coordinates (X, Y, Z) is established as follows: ; in, x 0 , y 0 , z 0 is the minimum coordinate value of the vertex of the spatial geometry corresponding to the x, y, and z directions in the camera pixel coordinate system, xm , y m 、z m are the maximum coordinate values ​​of the vertices of the spatial geometric body corresponding to the x, y, and z directions in the camera pixel coordinate system; X 0 , Y 0 , Z 0 is the minimum coordinate value of the vertex of the spatial geometric body corresponding to the x, y, and z directions in the robot motion coordinate system, X m , Y m , Z m It is the maximum coordinate value of the vertex of the spatial geometric body corresponding to the x, y, and z directions in the robot motion coordinate system.

[0014] One or more technical solutions provided in the embodiments of the present application have at least any of the following technical effects or advantages: By presetting the virtual space geometry, the spatial position within the robot's motion range can be calibrated, thereby taking into account errors such as installation and assembly, and avoiding adverse effects on calibration accuracy. In addition, by adjusting the bracket to determine the tip position of the robot end, the tip position capture accuracy and reliability can be effectively improved, and image recognition errors can be reduced. Therefore, the correspondence between the robot motion coordinate system and the camera pixel coordinate system is established through the virtual space geometry, and the calibration between the camera and the robot end is realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a schematic diagram of the alignment of the target point and the sharp point in one embodiment of the present application.

[0016] Figure 2 It is a schematic diagram of a spatial geometric body in an embodiment of the present application.

[0017] In the above figures, 1 is the tip of the end of the robot arm, and 2 is the adjustment bracket. DETAILED DESCRIPTION

[0018] The programs involved or relied upon in the following embodiments are all conventional programs or simple programs in the technical field, and those skilled in the art can make conventional selections or adaptive adjustments according to specific application scenarios. In order to better understand the technical solution of this application, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0019] In order to solve the technical problems of low calibration efficiency and poor accuracy of existing visual calibration methods, this example discloses a robot visual positioning calibration method, which specifically includes the following steps: (1) Preset the spatial geometry of the robot’s end point motion and specify the lengths of each side of the spatial geometry.

[0020] Considering that the end of the robot is in a spatial motion state during actual use, and the robot is designed, assembled and installed, there may be errors between the actual position coordinates of the robot's end point and the robot's motion coordinates. The camera is installed at the end of the robot and due to limitations such as installation accuracy, there are also assembly errors between it and the designed position point, which results in a difference between the pixel position coordinates corresponding to the camera and the end coordinates of the robot's end point, and the two are not numerically equivalent.

[0021] Therefore, in this embodiment, in order to ensure that the robot can be effectively calibrated in all three-dimensional motion directions, a spatial geometric body for the movement of the robot's terminal tip is preset, so that the robot's terminal tip moves to each vertex position of the spatial geometric body, thereby realizing spatial calibration. The number of vertices of the spatial geometric body is not less than four, and in this embodiment, the spatial geometric body is preset as a cuboid with a total of eight vertices. When the robot's terminal tip moves from one point to the remaining points, motion paths along the x-plane, y-plane, z-plane and along the space are constructed respectively, thereby improving the calibration accuracy and reliability. In some other embodiments, spatial geometric bodies with other numbers of vertices such as 16 or 32 are preset to further improve the calibration accuracy.

[0022] (2) Guide the tip of the robot end to move to a vertex of the spatial geometric body, and record the end coordinates of the tip of the robot end at that position in the robot motion coordinate system ( X n ，Y n ，Z n ).

[0023] During calibration, by inputting the side length corresponding to the preset spatial geometric body into the robot's motion control system, the system controls the robot movement accordingly, so that the tip of the robot end moves to a vertex position of the preset spatial geometric body. In this embodiment, considering that the motion space of the robot end is limited by its own structure, that is, the robot movement is limited to a certain space, for this reason, the preset spatial geometric body is within the robot's motion space range, so that the tip of the robot end can move to each vertex position of the spatial geometric body for calibration. In addition, in this example, the side length of the spatial geometric body is set to an integer multiple of the robot's motion scale, thereby ensuring that the tip of the robot end can accurately shift to each vertex position of the spatial geometric body, which helps to reduce the calibration error.

[0024] After the tip of the robot's end moves to the vertex position corresponding to the spatial geometric body, the actual coordinate value corresponding to the vertex under the robot's motion coordinates is recorded, that is, the end coordinate corresponding to the vertex ( X n ，Y n ，Z n ).

[0025] (3) Adjust the adjustment bracket until the target point on the top of the adjustment bracket coincides with the tip of the robot end.

[0026] After obtaining the terminal coordinates corresponding to the corresponding vertex, it is necessary to determine the coordinate value corresponding to the vertex in the camera coordinate system in order to realize calibration. However, since the tip of the robot's end is not easily captured accurately by the camera image, it is impossible to accurately obtain the position coordinates of the tip of the robot's end by directly taking pictures. For this reason, in this embodiment, an adjustment bracket is provided to assist in determining the coordinates of the tip of the robot's end. Specifically, the adjustment bracket is finely adjustable in the height direction, and can also be adjusted along a plane perpendicular to its height direction. A target plate is provided at the top of the adjustment bracket, and a bull's eye is provided in the center of the target plate. After the tip of the robot's end moves to the corresponding vertex position of the spatial geometric body, refer to Figure 1 , adjust the adjustment bracket so that the target point of the adjustment bracket coincides with the sharp point of the robot end, thereby making the target point and the sharp point of the robot end at the same point in space, and the target point can be quickly and accurately identified in the camera image, thereby obtaining the pixel coordinates of the sharp point in the camera image.

[0027] (4) Start the camera to take a picture of the target and obtain the pixel position coordinates (x n ，y n ， z n ).

[0028] After adjusting the adjustment bracket so that the target point on its top coincides with the tip of the robot end, and ensuring that the adjustment bracket is set stably, move the robot to the photo position, and adjust the camera field of view so that the target point is within the field of view. Use the robot end camera to take a photo of the target point at the adjustment bracket, and identify the position of the target point in the image, that is, the pixel position coordinates of the target point in the camera image (x n ，y n ，z n ). In some other embodiments, the camera is a fixed camera separate from the robot, which takes a picture of the target after the target of the bracket is adjusted to coincide with the tip of the robot end and the end of the robot leaves the target.

[0029] (5) Repeat the above steps (2) to (4) to obtain the corresponding end coordinates and pixel position coordinates of the remaining vertices of the spatial geometric body.

[0030] After steps (2) to (4), the initial vertex coordinates are obtained. However, considering that accurate coordinate correspondence cannot be achieved through only one point, when the robot moves in space, due to errors in assembly and installation, its actual movement deviates from the movement it thinks it should move in its coordinate system, so that the error away from the above coordinate point gradually increases with distance. Therefore, in order to obtain the accurate relationship between the robot motion coordinate system and the camera pixel position coordinate system, in this embodiment, steps (2) to (4) are repeated to obtain the terminal coordinates and pixel position coordinates corresponding to the remaining seven vertices of the spatial cuboid respectively. Since the spatial cuboid in this example is limited to the effective activity space of the robot; therefore, the coordinates of the remaining vertices are used to ensure reliable calibration within the range of movement of the robot terminal. In this example, the spatial geometric body adopts a cuboid, and its edges can be set along the coordinate axis direction respectively, which is convenient for subsequent calibration calculations.

[0031] (6) Establish the correspondence between the pixel position coordinates (x, y, z) and the end coordinates (X, Y, Z), and based on this correspondence, obtain the coordinate value of any pixel position point within the range of the spatial geometric body in the robot motion coordinate system to complete the camera visual calibration.

[0032] Through the above steps, the end coordinates of the eight vertices of the space cuboid are obtained respectively ( X n ，Y n ，Z n ) and the corresponding pixel position coordinates (x n ，y n ，z n ), where n = 1, 2, ... 8. For details, see Figure 2 The cuboid shown in the figure is a preset spatial geometric body, which includes eight vertices A to H, each of which corresponds to the robot's end motion coordinates ( X n ，Y n ，Z n ) and the camera pixel position coordinates (x n ，y n ，z n ), n=1, 2, ...8. For example, the terminal coordinates of point A in the robot motion coordinates are ( X 1 ，Y1 ，Z 1 ), the pixel position coordinates in the camera pixel coordinate system are (x 1 ，y 1 ，z 1 ), the coordinate values ​​under the two coordinate systems are proportional, but only the calibration of this point can be realized based on this proportional relationship, and the calibration of other points in the space range cannot be realized. Therefore, the coordinate values ​​of the remaining points of the spatial cuboid in the two coordinate systems are obtained. Figure 2 In the X direction shown, in this example, A is the initial point. Since the lengths of the sides of the cuboid along the X direction are equal, that is, the X coordinates corresponding to points A, D, F, and E are the same, and the X coordinates corresponding to points B, C, G, and H are the same. In this example, the X coordinates corresponding to points A, D, F, and E in the robot motion coordinate system are set to X 0 , the corresponding X-coordinate in the camera pixel coordinate system is x 0 ; Set the X-axis coordinates of points B, C, G, and H in the robot motion coordinate system to X m , the corresponding X-coordinate in the camera pixel coordinate system is x m ; Therefore, the corresponding relationship between the two coordinate systems in the X direction can be established, that is, the x coordinate in the camera pixel coordinate system corresponds to the x coordinate in the robot motion coordinate system: .

[0033] Similarly, the corresponding relationship between the Y direction and the Z direction is established, and the robot motion coordinate value corresponding to any point in the space in the camera pixel coordinate system can be obtained: ; in, x 0 , y 0 , z 0 is the minimum coordinate value of the vertex of the spatial geometry corresponding to the x, y, and z directions in the camera pixel coordinate system, x m , y m 、z m are the maximum coordinate values ​​of the vertices of the spatial geometric body corresponding to the x, y, and z directions in the camera pixel coordinate system; X 0 , Y 0 , Z 0is the minimum coordinate value of the vertex of the spatial geometric body corresponding to the x, y, and z directions in the robot motion coordinate system, X m , Y m , Z m It is the maximum coordinate value of the vertex of the spatial geometric body corresponding to the x, y, and z directions in the robot motion coordinate system.

[0034] Therefore, the robot end coordinates corresponding to the point under the robot motion coordinates can be obtained based on the pixel coordinates in the camera photo, thereby overcoming the image system recognition error caused by the camera installation position accuracy and posture error, improving the calibration accuracy and improving the calibration efficiency.

[0035] Although some preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application.

[0036] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the inventive concept. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A robot visual positioning calibration method, characterized in that: The steps include: (1) Preset the spatial geometry of the robot's end point motion and specify the lengths of each side of the spatial geometry; (2) Guide the tip of the robot end to move to a vertex of the spatial geometric body and record the end coordinates of the tip of the robot end at that position in the robot motion coordinate system ( X n ，Y n ，Z n ); (3) Adjusting the adjustment bracket until the target point on the top of the adjustment bracket coincides with the tip of the robot end; (4) Start the camera to take a picture of the target point and obtain the pixel position coordinates (x n ， y n ，z n ); (5) Repeat steps (2) to (4) to obtain the corresponding end coordinates and pixel position coordinates of the remaining vertices of the spatial geometric body; (6) Establish the correspondence between the pixel position coordinates (x, y, z) and the end coordinates (X, Y, Z), and based on this correspondence, obtain the coordinate value of any pixel position point within the range of the spatial geometric body in the robot motion coordinate system to complete the camera visual calibration.

2. The robot vision positioning calibration method according to claim 1, characterized in that: The spatial geometric body includes at least four vertices, and the spatial range corresponding to the spatial geometric body matches the spatial motion range of the robot.

3. The robot vision positioning calibration method according to claim 1, characterized in that: The spatial geometric body is a cuboid, and the length, width and height of the cuboid are integer multiples of the robot's motion scale.

4. The robot vision positioning calibration method according to claim 1, characterized in that: The camera is fixed at a corresponding position at the end of the robot.

5. The robot vision positioning calibration method according to claim 4, characterized in that: In step (4), before taking pictures, the robot is guided to move to the picture-taking position, and the camera field of view is adjusted until the target point is within the camera field of view.

6. The robot vision positioning calibration method according to claim 1, characterized in that: The camera is a fixed camera that does not move with the robot end.

7. The robot vision positioning calibration method according to claim 1, characterized in that: In step (6), the linear relationship between the pixel position coordinates (x, y, z) and the end coordinates (X, Y, Z) is established as: ； In the formula, x 0 , y 0 , z 0 is the minimum coordinate value of the vertex of the spatial geometric body corresponding to the x, y, and z directions in the camera pixel coordinate system; x m , y m 、z m are the maximum coordinate values ​​of the vertices of the spatial geometric body corresponding to the x, y, and z directions in the camera pixel coordinate system; X 0 , Y 0 , Z 0 are the minimum coordinate values ​​of the vertices of the spatial geometric body corresponding to the x, y, and z directions in the robot motion coordinate system; X m , Y m , Z m It is the maximum coordinate value of the vertex of the spatial geometric body corresponding to the x, y, and z directions in the robot motion coordinate system.

Images