Yarn hanging composite robot and visual control method

Abstract

The invention relates to the technical field of textile equipment, in particular to a yarn hanging composite robot and a visual control method.The yarn hanging composite robot comprises a power base, an electric control box is installed in a box body on the side face of the power base, and a material taking frame with prepared bobbins is placed on the surface of the top of the power base; yarn roll bodies are placed on the surface of the top position of the material taking frame with the prepared spools, a machine frame is installed on the surface of the top position of the electric control box, and a connecting piece is installed on the surface of the top position of the machine frame. The yarn hanging efficiency of yarn is greatly improved when the composite robot is used, the working efficiency of the composite robot is higher when the composite robot is used, and manpower resources are saved.

Description

Technical Field

[0001] The invention relates to the technical field of textile equipment, and in particular to a yarn hanging composite robot and a visual control method. Background Art

[0002] The spinning and weaving industry requires a series of complicated process flows from yarn reel to spinning into cloth, including hanging yarn, winding yarn, pickling, perming and dyeing, weaving yarn, etc. Among them, hanging yarn is still a process that requires a lot of manual labor. Workers need to use ladders to hold the yarn reels and hang them one by one on the yarn rack for yarn drawing. This is repetitive work and has low efficiency. Now there is a need for a yarn hanging composite robot.

[0003] The existing composite robot generally uses a forklift to drive a pallet filled with bobbins to follow the staff to the side of the fixed yarn frame, and then manually puts the bobbins on the upper yarn rod of the fixed yarn frame. The efficiency of the yarn loading operation is low. In addition, the existing composite robot is not efficient enough when hanging yarn on the yarn roll body, which greatly reduces the yarn hanging efficiency of the composite robot when in use, reduces the working efficiency of the composite robot when in use, and wastes human resources. Summary of the invention

[0004] The object of the present invention is to provide a yarn hanging composite robot and a visual control method to solve the problem in the above-mentioned background technology that the composite robot is not efficient enough when hanging yarn on a yarn roll body.

[0005] To achieve the above-mentioned objectives, the present invention provides a yarn hanging composite robot on one hand: the yarn hanging composite robot comprises a power base, an electric control box is installed in the side box body of the power base, a material taking rack with prepared yarn tubes is placed on the surface at the top position of the power base, a yarn roll body is placed on the surface at the top position of the material taking rack with prepared yarn tubes, a frame is installed on the surface at the top position of the electric control box, and a connecting piece is installed on the surface at the top position of the frame.

[0006] Preferably, a first mechanical arm is provided on the surface of the connecting member, and the first mechanical arm and the inner wall of the connecting member are rotatably matched with each other, and a second mechanical arm is provided on the surface of the first mechanical arm, and the second mechanical arm and the surface of the first mechanical arm are rotatably matched with each other.

[0007] Preferably, a material picking rack is provided on the surface of the second robotic arm, and the material picking rack and the surface of the second robotic arm are rotatably matched with each other. A clamp for clamping the yarn roll is installed on the surface of the material picking rack, and the diameter of the clamp is smaller than the diameter of the yarn roll body.

[0008] Preferably, a light-sensing positioning sensor is installed on the surface of the material picking rack, and the input end of the light-sensing positioning sensor is electrically connected to the output end of the electric control box. A laser sensor is installed on the surface of the material picking rack, and the input end of the laser sensor is electrically connected to the output end of the electric control box. A controller is installed on the surface of the second robotic arm, and the input end of the controller is electrically connected to the output end of the electric control box.

[0009] Preferably, an in-position sensor is installed on the surface of the material picking rack, the input end of the in-position sensor is electrically connected to the output end of the electric control box, and the output end of the controller is fixed to the surface of the material clamp.

[0010] Preferably, the interior of the power base is provided with a loading assembly for taking materials from the material taking rack with prepared yarn tubes, the loading assembly and the inner wall of the power base are slidably matched with each other, and the loading assembly and the surface at the bottom position of the material taking rack with prepared yarn tubes are matched with each other.

[0011] Another aspect of the present invention provides a visual control method for a yarn hanging composite robot, comprising the following steps:

[0012] S1. Yarn bobbin detection: Use the yarn hanging robot camera to collect multiple batches of yarn bobbin images under different light and dark conditions, mark the yarn bobbin positions, and use deep learning tools to train a yarn bobbin detection model;

[0013] S2, hook detection, using the yarn hanging robot camera to collect multiple batches of creel hook images under different light and dark conditions, after marking the hook position, use deep learning tools to train a hook detection model;

[0014] S3, initial position setting, through the robot arm teaching pendant, calibrate the angle of the clamping jaws at the initial position so that it is perpendicular to the circular surface of the yarn tube entrance and adjust it to a predetermined distance;

[0015] S4, calibration, by completing the calibration process for the horizontal downward viewing angle, obtaining the image coordinates of the center of the bobbin and the physical coordinate conversion matrix of the XY two-dimensional plane of the robot arm, and by completing the calibration process for the vertical rightward viewing angle, obtaining the image coordinates of the end point of the creel hook and the physical coordinate conversion matrix of the YZ two-dimensional plane of the robot arm;

[0016] S5, yarn bobbin positioning, the robot moves to the predetermined photo point above the yarn bobbin, point P0 (x0, y0), takes a photo to obtain an image, uses the yarn bobbin recognition model to obtain the image coordinates of the yarn bobbin center, and converts the image coordinates of the yarn bobbin center to the physical coordinate offset point P1 (x1, y1) through the conversion matrix of the XY two-dimensional plane physical coordinates of the robot arm. Then, the precise position of the yarn bobbin entrance of the robot arm is P2 (x0+x1, y0+y1);

[0017] S6, hook positioning, the robot moves to the predetermined photo point on the left side of the hook endpoint, point P0 (y0, z0), takes a photo to obtain an image, uses the hook recognition model to obtain the image coordinates of the hook endpoint, and converts them into physical coordinate offset point P1 (y1, z1) through the conversion matrix of the creel hook endpoint image coordinates and the robot YZ two-dimensional plane physical coordinates. Then the precise position of the hook placement entrance of the robot is P2 (y0+y1, z0+z1).

[0018] Preferably, in the calibration process of step S4, a fixed target object is selected, and nine evenly distributed calibration points {P1, P2, ..., P9} are defined on its surface, which cover different positions of the entire working area. The three-dimensional coordinates of each calibration point Pi on the target object are known as (Pxi, Pyi, Pzi), where i = 1, 2, ..., 9. For each calibration point Pi, the robot arm positions the end effector near the point, and adjusts the angle and distance of the camera so that the camera can clearly capture the calibration point Pi, and records the robot arm joint angle θj (j = 1, 2, ..., n) and the two-dimensional projection coordinates (uvi) of Pi in the image taken by the camera, where uvi represents the pixel coordinate.

[0019] Preferably, the coordinate transformation process is as follows:

[0020] Assume that Rc and tc are the rotation matrix and translation vector from the camera coordinate system C to the robot base coordinate system B, respectively. The transformation relationship between any point P in the two coordinate systems can be expressed as:

[0021] P B =R c ·P C +t c

[0022] Where PB is the coordinate of point P in the robot base coordinate system, and PC is the coordinate of the same point in the camera coordinate system.

[0023] Preferably, the projection process from space to image is as follows:

[0024] Considering the camera intrinsic parameter matrix K, the coordinates of point P in the camera coordinate system are mapped to the two-dimensional coordinates (uv) on the image plane through the following perspective projection model:

[0025]

[0026] Here, (X,Y,Z)_C represents the coordinates of point P in the camera coordinate system, and K is a 3x3 matrix containing the focal length f and the principal point offset cx,cy:

[0027]

[0028] Compared with the prior art, the beneficial effects of the present invention are: the yarn hanging composite robot greatly improves the yarn hanging efficiency of the composite robot when in use, makes the composite robot more efficient when in use, and saves human resources;

[0029] By providing an efficient yarn hanging mechanism, the yarn roll bodies are placed on the surfaces of the material taking racks with the yarn bobbins prepared, and the yarn roll bodies are stored under the action of the material taking racks with the yarn bobbins prepared, and the power base automatically moves to one side of the material taking racks with the yarn bobbins prepared. At this time, the power base pushes the feeding components out of the interior of the power base respectively, so that one end of the feeding components moves to the bottom of the material taking racks with the yarn bobbins prepared. At this time, the feeding components are automatically clamped on the bottom of the material taking racks with the yarn bobbins prepared, so that the feeding components can place the material taking racks with the yarn bobbins prepared on the surface of the power base. Subsequently, the first mechanical arm rotates on the surface of the connecting piece, and the second mechanical arm is on the surface of the first mechanical arm. The user operates the electric control box so that the electric control box controls the light positioning sensor, the laser sensor and the in-position sensor to work respectively, and the position of the yarn roll body is optically positioned under the action of the light positioning sensor, so that the material taking rack can automatically identify the position of the yarn roll body. The auxiliary clamp senses the position of the yarn roll body under the action of the auxiliary clamp, so that the clamp can accurately clamp the yarn roll body. At this time, the clamp is close to the inner wall of the yarn roll body to prevent the yarn roll body from sliding when hanging the yarn. The yarn roll body can be stably hung on the surface of the material rack with the prepared yarn tube. The user then operates the electric control box to make the electric control box control the controller to work. Under the action of the in-place sensor, the position of the power base can be monitored. When the yarn roll body on the surface of the material rack with the prepared yarn tube is hung with yarn, the power base can be automatically controlled to stop working, and the clamp is driven to move to the inside of the yarn roll body under the action of the controller. Under the joint action of the controller and the clamp, the yarn roll body can be hung with yarn, thereby realizing the function of the composite robot to efficiently hang yarn on the yarn roll body, thereby greatly improving the yarn hanging efficiency of the composite robot when in use, making the composite robot more efficient when in use, and saving human resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a schematic diagram of the three-dimensional structure of the present invention;

[0031] Figure 2 It is a schematic diagram of the three-dimensional structure of the present invention;

[0032] Figure 3 It is a schematic diagram of the main cross-sectional structure of the present invention;

[0033] Figure 4 It is a rear cross-sectional structural schematic diagram of the present invention;

[0034] Figure 5It is a schematic diagram of an enlarged side cross-sectional structure of the present invention;

[0035] Figure 6 It is a schematic diagram of an enlarged side cross-sectional structure of the present invention;

[0036] Figure 7 It is a schematic diagram of an enlarged top view of the cross section of the present invention.

[0037] In the figure: 1. Power base; 101. Electric control box; 102. Frame; 103. First mechanical arm; 104. Second mechanical arm; 105. Yarn roll body; 106. Material taking rack; 107. Connector; 108. Material clamp; 109. Material taking rack with yarn tubes prepared; 110. Laser sensor; 111. Light positioning sensor; 112. Feeding assembly; 113. In-position sensor; 114. Controller. DETAILED DESCRIPTION

[0038] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. In addition, the terms "first", "second", "third", "upper, lower, left, right", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. At the same time, in the description of the present invention, unless otherwise clearly specified and limited, the terms "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0039] The structure of the yarn hanging composite robot provided by the present invention is as follows Figure 3 and Figure 4As shown, it includes a power base 1, an electric control box 101 is installed in the side box of the power base 1, and the model of the electric control box 101 can be selected from the LA series. A material rack 109 with prepared yarn bobbins is placed on the surface at the top position of the power base 1, and a yarn roll body 105 is placed on the surface at the top position of the material rack 109 with prepared yarn bobbins. A frame 102 is installed on the surface at the top position of the electric control box 101, and a connecting piece 107 is installed on the surface at the top position of the frame 102. A first mechanical arm 103 is arranged on the surface of the connecting piece 107. The first mechanical arm 103 is arranged on the surface of the first mechanical arm 103. The arm 103 and the inner wall of the connecting piece 107 are rotatably matched with each other. The surface of the first mechanical arm 103 is provided with the second mechanical arm 104. The second mechanical arm 104 and the surface of the first mechanical arm 103 are rotatably matched with each other. The surface of the second mechanical arm 104 is provided with a material picking rack 106. The material picking rack 106 and the surface of the second mechanical arm 104 are rotatably matched with each other. The surface of the material picking rack 106 is installed with a clamp 108 for clamping the yarn roll. The diameter of the clamp 108 is smaller than the diameter of the yarn roll body 105. The surface of the material picking rack 106 is installed with a light-sensing fixed A position sensor 111, the model of which can be selected from the HAD series, the input end of which is electrically connected to the output end of the electric control box 101, a laser sensor 110 is mounted on the surface of the material taking rack 106, the model of which can be selected from the RS series, the input end of which is electrically connected to the output end of the electric control box 101, a controller 114 is mounted on the surface of the second robot arm 104, the input end of which is electrically connected to the output end of the electric control box 101, the material taking rack 106 06 is installed with an in-position sensor 113 on its surface, and the model of the in-position sensor 113 may be selected from the ZDI series. The input end of the in-position sensor 113 is electrically connected to the output end of the electric control box 101, and the output end of the controller 114 is fixed to the surface of the clamp 108. The interior of the power base 1 is provided with a loading assembly 112 for taking materials from the taking rack 109 with prepared yarn tubes. The loading assembly 112 and the inner wall of the power base 1 slide in cooperation with each other, and the loading assembly 112 and the surface at the bottom of the taking rack 109 with prepared yarn tubes cooperate with each other.

[0040] During implementation, the yarn roll body 105 is placed on the surface of the material rack 109 with the yarn bobbins prepared, and the yarn roll body 105 is stored under the action of the material rack 109 with the yarn bobbins prepared. The power base 1 automatically moves to one side of the material rack 109 with the yarn bobbins prepared. At this time, the power base 1 pushes the feeding assembly 112 out of the power base 1, so that one end of the feeding assembly 112 moves to the bottom of the material rack 109 with the yarn bobbins prepared. At this time, the feeding assembly 112 is automatically clamped at the bottom of the material rack 109 with the yarn bobbins prepared. , so that the loading assembly 112 can place the material taking rack 109 with the prepared yarn tube on the surface of the power base 1, then the first mechanical arm 103 rotates on the surface of the connecting piece 107, and the second mechanical arm 104 is on the surface of the first mechanical arm 103. The user operates the electric control box 101, so that the electric control box 101 controls the light sensing positioning sensor 111, the laser sensor 110 and the in-position sensor 113 to work respectively, and the position of the yarn roll body 105 is optically positioned under the action of the light sensing positioning sensor 111, so that the material taking The rack 106 can automatically identify the position of the yarn roll body 105, and the auxiliary clamp 108 can sense the position of the yarn roll body 105 under the action of the laser sensor 110, so that the clamp 108 can accurately clamp the yarn roll body 105. At this time, the clamp 108 is close to the inner wall of the yarn roll body 105 to prevent the yarn roll body 105 from sliding when hanging the yarn. The yarn roll body 105 can be stably hung on the surface of the material rack with the yarn tube prepared. The user then operates the electric control box 101, so that the electric control box 101 controls The controller 114 works and can monitor the position of the power base 1 under the action of the in-position sensor 113. When the yarn roll body 105 on the surface of the ready-made yarn tube feeding rack 109 is hung with yarn, the power base 1 can be automatically controlled to stop working. Under the action of the controller 114, the clamp 108 is driven to move to the inside of the yarn roll body 105. Under the joint action of the controller 114 and the clamp 108, the yarn roll body 105 can be hung with yarn, so as to realize the function of the composite robot to efficiently hang yarn on the yarn roll body 105.

[0041] This embodiment also provides a visual control method for a yarn hanging composite robot, comprising the following steps:

[0042] S1. Yarn bobbin detection: Use the yarn hanging robot camera to collect multiple batches of yarn bobbin images under different light and dark conditions. After marking the yarn bobbin positions, use deep learning tools to train a yarn bobbin detection model. Since the placement of the yarn bobbin is also a circular pit, it is necessary to specially collect the interference image data set to enhance the model's anti-interference ability and avoid misidentifying the placement pit as a yarn bobbin;

[0043] S2. Hook detection: Use the yarn hanging robot camera to collect multiple batches of creel hook images under different light and dark conditions. After marking the hook position, use the deep learning tool to train the hook detection model. Since there may be a yarn tube or yarn tube core on the hook, it is necessary to specially collect the interference data to enhance the anti-interference ability of the model and avoid putting yarn on non-empty hooks.

[0044] S3, initial position setting, through the robot arm teaching pendant, calibrate the angle of the clamping jaws at the initial position so that it is perpendicular to the circular surface of the yarn tube entrance and adjust it to a predetermined distance;

[0045] S4, calibration, by completing the calibration process for the horizontal downward viewing angle, obtaining the image coordinates of the center of the bobbin and the physical coordinate conversion matrix of the XY two-dimensional plane of the robot arm, and by completing the calibration process for the vertical rightward viewing angle, obtaining the image coordinates of the end point of the creel hook and the physical coordinate conversion matrix of the YZ two-dimensional plane of the robot arm;

[0046] S5, yarn bobbin positioning, the robot moves to the predetermined photo point above the yarn bobbin, point P0 (x0, y0), takes a photo to obtain an image, uses the yarn bobbin recognition model to obtain the image coordinates of the yarn bobbin center, and converts the image coordinates of the yarn bobbin center to the physical coordinate offset point P1 (x1, y1) through the conversion matrix of the XY two-dimensional plane physical coordinates of the robot arm. Then, the precise position of the yarn bobbin entrance of the robot arm is P2 (x0+x1, y0+y1);

[0047] S6, hook positioning, the robot moves to the predetermined photo point on the left side of the hook endpoint, point P0 (y0, z0), takes a photo to obtain an image, uses the hook recognition model to obtain the image coordinates of the hook endpoint, and converts them into physical coordinate offset point P1 (y1, z1) through the conversion matrix of the creel hook endpoint image coordinates and the robot YZ two-dimensional plane physical coordinates. Then the precise position of the hook placement entrance of the robot is P2 (y0+y1, z0+z1).

[0048] This embodiment uses a nine-point calibration method for the robot arm and camera system, which allows the hand-eye (i.e., the robot's end effector and camera) to move in coordination while the target object remains stationary. This method is intended to ensure the precise conversion relationship between the camera coordinate system and the robot base coordinate system to improve the accuracy of robot operation. During the calibration process, a fixed target object is selected and nine uniformly distributed calibration points {P1, P2, ..., P9} are defined on its surface. These points cover different positions of the entire working area. The three-dimensional coordinates of each calibration point Pi on the target object are known as (Pxi, Pyi, Pzi), where i = 1, 2, ..., 9. For each calibration point Pi, the robot arm positions the end effector near the point and adjusts the angle and distance of the camera so that the camera can clearly capture the calibration point Pi, and records the robot arm joint angle θj (j = 1, 2, ..., n) and the two-dimensional projection coordinates (uvi) of Pi in the image captured by the camera, where uvi represents the pixel coordinate.

[0049] The coordinate transformation process is as follows:

[0050] Assume that Rc and tc are the rotation matrix and translation vector from the camera coordinate system C to the robot base coordinate system B, respectively. The transformation relationship between any point P in the two coordinate systems can be expressed as:

[0051] P B =R c ·P C +t c

[0052] Where PB is the coordinate of point P in the robot base coordinate system, and PC is the coordinate of the same point in the camera coordinate system.

[0053] The projection process from space to image is as follows:

[0054] Considering the camera intrinsic parameter matrix K, the coordinates of point P in the camera coordinate system are mapped to the two-dimensional coordinates (uv) on the image plane through the following perspective projection model:

[0055]

[0056] Here, (X,Y,Z)_C represents the coordinates of point P in the camera coordinate system, and K is a 3x3 matrix containing the focal length f and the principal point offset cx,cy:

[0057]

[0058] The visual hardware of the yarn hanging robot of this embodiment is matched with the yarn tube clamp, and the central optical axis of the camera is parallel to the direction in which the clamp reaches down to clamp the yarn; the camera has a variety of fill-light LED lights, which can cope with various situations such as weak light, strong light, and reflection in the center of the calibration plate according to the needs of the use link; by adjusting the focal length of the camera to change the depth of field, it can cover the target object distance range from 35cm to 100cm.

[0059] Working principle: When in use, first place the power base 1 in the yarn production workshop, and place the yarn roll body 105 on the surface of the material rack 109 with the yarn tubes prepared. Under the action of the material rack 109 with the yarn tubes prepared, the yarn roll body 105 is stored, and the power base 1 automatically moves to one side of the material rack 109 with the yarn tubes prepared. At this time, the power base 1 pushes the feeding assembly 112 out of the power base 1, so that one end of the feeding assembly 112 moves to the bottom of the material rack 109 with the yarn tubes prepared. At this time, the feeding assembly 112 is automatically connected to the bottom of the material rack 109 with the yarn tubes prepared. The loading assembly 112 enables the material taking rack 109 with the prepared yarn tube to be placed on the surface of the power base 1. Subsequently, the first mechanical arm 103 rotates on the surface of the connecting piece 107, and the second mechanical arm 104 is on the surface of the first mechanical arm 103. The user operates the electric control box 101, so that the electric control box 101 controls the light sensing positioning sensor 111, the laser sensor 110 and the in-position sensor 113 to work respectively. Under the action of the light sensing positioning sensor 111, the position of the yarn roll body 105 is optically positioned, so that the material taking rack 106 can automatically identify the position of the yarn roll body 105. Under the action of the laser sensor 110, the auxiliary clamp 108 senses the position of the yarn roll body 105, so that the clamp 108 can accurately clamp the yarn roll body 105. At this time, the clamp 108 is tightly attached to the inner wall of the yarn roll body 105 to prevent the yarn roll body 105 from sliding when hanging the yarn. The yarn roll body 105 can be stably hung on the surface of the material rack with the prepared yarn tube. The user then operates the electric control box 101, so that the electric control box 101 controls the controller 114 to work, and under the action of the in-position sensor 113, the position of the power base 1 can be monitored. When the material rack 114 with the prepared yarn tube is in place, After the yarn roll body 105 on the surface of 09 is hung with yarn, the power base 1 can be automatically controlled to stop working, and the clamp 108 is driven to move to the inside of the yarn roll body 105 under the action of the controller 114. Under the joint action of the controller 114 and the clamp 108, the yarn roll body 105 can be hung with yarn to realize the function of the composite robot to efficiently hang yarn on the yarn roll body 105, thereby greatly improving the yarn hanging efficiency of the composite robot when in use, making the composite robot more efficient when in use, saving human resources, and finally completing the use of the composite robot.

[0060] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A yarn hanging composite robot, comprising a power base (1), characterized in that: An electric control box (101) is installed in the side box of the power base (1); a material taking rack (109) with ready yarn bobbins is placed on the surface at the top position of the power base (1); a yarn roll body (105) is placed on the surface at the top position of the material taking rack (109) with ready yarn bobbins; a frame (102) is installed on the surface at the top position of the electric control box (101); and a connecting piece (107) is installed on the surface at the top position of the frame (102).

2. The yarn hanging composite robot according to claim 1, characterized in that: A first mechanical arm (103) is arranged on the surface of the connecting member (107), and the first mechanical arm (103) and the inner wall of the connecting member (107) are rotatably matched with each other. A second mechanical arm (104) is arranged on the surface of the first mechanical arm (103), and the second mechanical arm (104) and the surface of the first mechanical arm (103) are rotatably matched with each other.

3. The yarn hanging composite robot according to claim 2, characterized in that: The surface of the second robotic arm (104) is provided with a material picking rack (106), and the material picking rack (106) and the surface of the second robotic arm (104) are rotatably matched with each other. The surface of the material picking rack (106) is installed with a clamp (108) for clamping the yarn roll, and the diameter of the clamp (108) is smaller than the diameter of the yarn roll body (105).

4. The yarn hanging composite robot according to claim 3, characterized in that: A light-sensing positioning sensor (111) is installed on the surface of the material-retrieving rack (106), and the input end of the light-sensing positioning sensor (111) is electrically connected to the output end of the electric control box (101). A laser sensor (110) is installed on the surface of the material-retrieving rack (106), and the input end of the laser sensor (110) is electrically connected to the output end of the electric control box (101). A controller (114) is installed on the surface of the second robot arm (104), and the input end of the controller (114) is electrically connected to the output end of the electric control box (101).

5. The yarn hanging composite robot according to claim 4, characterized in that: A position sensor (113) is installed on the surface of the material picking rack (106), the input end of the position sensor (113) is electrically connected to the output end of the electric control box (101), and the output end of the controller (114) is fixed to the surface of the material clamp (108).

6. The yarn hanging composite robot according to claim 1, characterized in that: The interior of the power base (1) is provided with a loading assembly (112) for taking material from a material taking rack (109) with prepared yarn bobbins. The loading assembly (112) and the inner wall of the power base (1) are slidably matched with each other, and the loading assembly (112) and the surface at the bottom position of the material taking rack (109) with prepared yarn bobbins are matched with each other.

7. A visual control method for a yarn hanging composite robot, characterized in that: The following steps are involved: S1. Yarn bobbin detection: Use the yarn hanging robot camera to collect multiple batches of yarn bobbin images under different light and dark conditions, mark the yarn bobbin positions, and use deep learning tools to train a yarn bobbin detection model; S2, hook detection, using the yarn hanging robot camera to collect multiple batches of creel hook images under different light and dark conditions, after marking the hook position, use deep learning tools to train a hook detection model; S3, initial position setting, through the robot arm teaching pendant, calibrate the angle of the clamping jaws at the initial position so that it is perpendicular to the circular surface of the yarn tube entrance and adjust it to a predetermined distance; S4, calibration, by completing the calibration process for the horizontal downward viewing angle, obtaining the image coordinates of the center of the bobbin and the physical coordinate conversion matrix of the XY two-dimensional plane of the robot arm, and by completing the calibration process for the vertical rightward viewing angle, obtaining the image coordinates of the end point of the creel hook and the physical coordinate conversion matrix of the YZ two-dimensional plane of the robot arm; S5, yarn bobbin positioning, the robot moves to the predetermined photo point above the yarn bobbin, point P0 (x0, y0), takes a photo to obtain an image, uses the yarn bobbin recognition model to obtain the image coordinates of the yarn bobbin center, and converts the image coordinates of the yarn bobbin center to the physical coordinate offset point P1 (x1, y1) through the conversion matrix of the XY two-dimensional plane physical coordinates of the robot arm. Then, the precise position of the yarn bobbin entrance of the robot arm is P2 (x0+x1, y0+y1); S6, hook positioning, the robot moves to the predetermined photo point on the left side of the hook endpoint, point P0 (y0, z0), takes a photo to obtain an image, uses the hook recognition model to obtain the image coordinates of the hook endpoint, and converts them into physical coordinate offset point P1 (y1, z1) through the conversion matrix of the creel hook endpoint image coordinates and the robot YZ two-dimensional plane physical coordinates. Then the precise position of the hook placement entrance of the robot is P2 (y0+y1, z0+z1).

8. The visual control method of the yarn hanging composite robot according to claim 7 is characterized in that: In the calibration process of step S4, a fixed target object is selected and nine uniformly distributed calibration points {P1, P2, ..., P9} are defined on its surface. These points cover different positions of the entire working area. The three-dimensional coordinates of each calibration point Pi on the target object are known to be (Pxi, Pyi, Pzi), where i = 1, 2, ..., 9. For each calibration point Pi, the robot arm positions the end effector near the point and adjusts the angle and distance of the camera so that the camera can clearly capture the calibration point Pi. The robot arm joint angle θj (j = 1, 2, ..., n) and the two-dimensional projection coordinates (uvi) of Pi in the image taken by the camera are recorded, where uvi represents the pixel coordinates.

9. The visual control method of the yarn hanging composite robot according to claim 7, characterized in that: The coordinate transformation process is as follows: Assume that Rc and tc are the rotation matrix and translation vector from the camera coordinate system C to the robot base coordinate system B, respectively. The transformation relationship between any point P in the two coordinate systems can be expressed as: P B =R c ·P C +t c Where PB is the coordinate of point P in the robot base coordinate system, and PC is the coordinate of the same point in the camera coordinate system.

10. The visual control method of the yarn hanging composite robot according to claim 7, characterized in that: The projection process from space to image is as follows: Considering the camera intrinsic parameter matrix K, the coordinates of point P in the camera coordinate system are mapped to the two-dimensional coordinates (uv) on the image plane through the following perspective projection model: Here, (X,Y,Z)_C represents the coordinates of point P in the camera coordinate system, and K is a 3x3 matrix containing the focal length f and the principal point offset cx,cy:

Images