Intelligent box grabbing device and method based on 3D visual camera guidance

Abstract

The invention discloses an intelligent box grabbing device and method based on 3D visual camera guidance, and belongs to the technical field of vehicle guarantee. The intelligent box grabbing device comprises a multi-degree-of-freedom driving mechanism, a hoisting framework and a processor. The multi-degree-of-freedom driving mechanism comprises a mechanical arm and a movable joint; a movable joint is arranged between every two adjacent mechanical arms; the hoisting framework is rotationally connected to the moving end, and a locking assembly and a 3D camera are arranged on the hoisting framework; the distribution form of the locking assemblies on the hoisting framework is the same as the distribution form of the butt joint holes in the top end face of the box body. The 3D camera is arranged on one side of the hoisting framework; the processor is in communication connection with the movable joints and the 3D camera and drives the movable joints to move according to the image information. The grabbing device provided by the invention can adapt to boxes at different positions, not only can realize accurate identification of butt joint holes in the tops of the boxes, but also can adjust grabbing of the boxes based on the positions of the boxes, so that the grabbing efficiency of the boxes is remarkably improved.

Description

Technical Field

[0001] This invention belongs to the field of vehicle protection technology, specifically relating to a box-shaped intelligent grasping device and method guided by a 3D vision camera. Background Technology

[0002] Containers are widely used in the transportation sector, and the transshipment of containers plays a particularly crucial role in the container transportation process. Whether the transshipment of containers is accurate not only affects the time required for transshipment but also the transportation efficiency. Moreover, if the containers cannot be placed in a stable position, there is a risk of the containers falling during transportation.

[0003] In related technologies, the vehicle body is mainly grasped manually by hoisting it. However, manual grasping requires a high level of skill from the operator and is quite difficult. If the grasp is inaccurate, the swaying of the body due to gravity will be more pronounced, posing a safety hazard. Summary of the Invention

[0004] In view of one or more of the above-mentioned defects or improvement needs of the prior art, the present invention provides a box intelligent grasping device and method based on 3D vision camera guidance. It can adapt to boxes in different positions, not only accurately identify the docking holes on the top of the box, but also adjust the grasping of the box based on the position of the box, thereby significantly improving the grasping efficiency of the box while ensuring the stability of the grasping.

[0005] To achieve the above objectives, the present invention provides a box-grabbing intelligent grasping device guided by a 3D vision camera, for grasping a box with several docking holes on its top surface, comprising: A multi-degree-of-freedom drive mechanism, comprising several robotic arms and several movable joints; The movable joint is provided between two adjacent robotic arms, and the robotic arms are connected in sequence to form a multi-degree-of-freedom robotic arm; one end of the multi-degree-of-freedom robotic arm is fixed, and the other end of the multi-degree-of-freedom robotic arm is a moving end; A hoisting frame, which is rotatably connected to the moving end, and is provided with several locking components and a 3D camera; The locking components are distributed in the same way as the docking holes on the top surface of the box body, so that when the lifting frame docks with the box body, the locking components can be locked in the docking holes. The 3D camera is set on one side of the hoisting frame and is used to collect image information of the docking holes on both sides after the hoisting frame is rotated horizontally. The processor is communicatively connected to the movable joint and the 3D camera, respectively, for receiving the image information and driving the movable joint to move according to the image information, so as to lock the locking component on the hoisting frame in the docking hole.

[0006] As a further preferred embodiment of the present invention, the multi-degree-of-freedom drive mechanism includes a first robotic arm, a second robotic arm, a third robotic arm, a fourth robotic arm, a fifth robotic arm, and a base; A first motion joint is provided between the first robotic arm and the crane frame; a second motion joint is provided between the first robotic arm and the second robotic arm; a third motion joint is provided between the second robotic arm and the third robotic arm; a fourth motion joint is provided between the third robotic arm and the fourth motion joint; a fifth motion joint is provided between the fourth robotic arm and the fifth robotic arm; and a sixth motion joint is provided between the fifth robotic arm and the base.

[0007] As a further preferred embodiment of the present invention, four docking holes are provided at the four corners of the top surface of the box body, and correspondingly, the hoisting frame is an I-shaped truss, and four locking components are provided at the four corners of the bottom surface of the hoisting frame.

[0008] As a further preferred embodiment of the present invention, the locking assembly includes a rotary lock head and a locking rotation unit; The locking rotation unit is fixedly installed on the hoisting frame, and the rotating end of the locking rotation unit is fixedly connected to one end of the rotating lock head, which is used to drive the rotating lock head to rotate around the axis; The other end of the rotary lock has a T-shaped block, and the mating hole is a square hole, which is used to rotate and embed the T-shaped block into the square hole during locking and gripping, so as to form a lock between the hoisting frame and the box.

[0009] This invention also discloses a box-grabbing intelligent grasping method based on a 3D vision camera, which uses the aforementioned box-grabbing intelligent grasping device based on a 3D vision camera to grasp the box. The grasping method includes the following steps: S1. Collect image information of the four docking holes located at the four corners of the top surface of the box; S2. Obtain the center coordinates of the four docking holes based on the image information of the four docking holes; The coordinates of the center positions of the four docking holes are respectively (x a y a , z a ), (x b y b , z b ), (x c y c , z c ) and (xd y d , z d ); S3. Construct a position matrix of the center of the docking hole relative to the center point of the top surface of the box body based on the coordinates of the center positions of the four docking holes; S4. Calculate the box pose T based on the position matrix; Where, box pose T=[x_temp T y_temp T n T O_temp T ]; S5. Obtain the pose of the lifting frame based on the pose of the end effector of the six-axis robotic arm; S6. Calculate the control amount of the lifting frame based on the position of the lifting frame and the position of the box T, and control the movement of the six-axis robotic arm based on the control amount of the lifting frame until the lifting frame locks the docking hole.

[0010] As a further preferred embodiment of the present invention, S1 includes the following steps: S11. Adjust the posture of the hoisting frame to a horizontal state and transport the hoisting frame to the top of the box; S12. Acquire image information of the first and second docking holes on one side; S13. Control the hoisting frame to rotate 180 degrees and collect image information of the third and fourth docking holes on one side.

[0011] As a further preferred embodiment of the present invention, the image information includes multiple sets of graphic data from different perspectives.

[0012] As a further preferred embodiment of the present invention, S2 includes the following steps: S21. Obtain the spatial coordinates and projection ray direction when the 3D camera captures each set of graphic data; S22. Extract each set of graphic data and find the feature points of each of the said docking holes; S23. Based on the spatial coordinates of the 3D camera corresponding to each docking hole, the direction of the projection ray, and the feature points, the three-dimensional spatial coordinates of each docking hole are calculated sequentially using a triangulation algorithm.

[0013] As a further preferred embodiment of the present invention, S3 includes the following steps: S31. Calculate the coordinates of the center position O_temp on the top surface of the box; S32. Calculate the vector of the center position of each of the said docking holes relative to the center position O of the top surface of the box; S33. Construct the position matrix of the center of the docking hole relative to the center point of the top surface of the box; As a further preferred embodiment of the present invention, the position matrix A of the center of the mating hole relative to the center point of the top surface of the housing is:

[0014] In the formula, the center point of the first mating hole is a, and its coordinates are (x, y, y). a y a , z a The center point of the second mating hole is b, and its coordinates are (x, y). b y b , z b The center point of the third mating hole is c, and its coordinates are x. c y c , z c The center point of the fourth mating hole is d, and its coordinates are (x, y). d y d , z d ).

[0015] In summary, the beneficial effects of the above-described technical solutions conceived by this invention compared with the prior art include: (1) The intelligent box-grabbing device based on 3D vision camera guidance of the present invention includes a multi-degree-of-freedom drive mechanism, a hoisting frame, and a processor. The multi-degree-of-freedom drive mechanism includes several robotic arms and several movable joints; movable joints are provided between adjacent robotic arms, and each robotic arm is connected in sequence to form a multi-degree-of-freedom robotic arm; one end of the multi-degree-of-freedom robotic arm is fixed, and the other end of the multi-degree-of-freedom robotic arm is a moving end; the hoisting frame is rotatably connected to the moving end, and several locking components and a 3D camera are provided on the hoisting frame; the distribution of each locking component on the hoisting frame is the same as the distribution of the docking holes on the top surface of the box, so that the locking components can be locked in the docking holes when the hoisting frame docks with the box; the 3D camera is set on one side of the hoisting frame, so as to collect image information of the docking holes on both sides after the hoisting frame is rotated horizontally; the processor is communicatively connected to the movable joints and the 3D camera respectively, so as to receive image information and drive each movable joint to move according to the image information, so as to lock the locking components on the hoisting frame in the docking holes. This invention provides a gripping method that can adapt to boxes in different positions. It can not only accurately identify the docking holes on the top of the box, but also adjust the gripping of the box based on the box position, thereby significantly improving the gripping efficiency of the box while ensuring the gripping stability.

[0016] (2) The intelligent box-grabbing device and method based on 3D vision camera guidance of the present invention has good adaptability, accurate recognition, and high efficiency. By rotating the hoisting frame, a single 3D camera can acquire images of all docking holes, which not only ensures the clarity of image acquisition but also reduces the equipment cost of setting up multiple 3D cameras. Furthermore, when the 3D camera captures the position of the docking hole, the camera field of view of the 3D camera can simultaneously cover two adjacent docking holes while minimizing the distance between the 3D camera and the box, thereby further improving the clarity of the image. At the same time, it accurately calculates the box pose based on the spatial position matrix of the docking hole components, and then accurately controls the movement of each joint based on the box pose and the real-time position of the six-axis robotic arm, thereby enabling the six-axis robotic arm to drive the rotating hoist to achieve stable grasping of the box. It has high promotional value and application prospects. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the overall structure of the intelligent box-grabbing device guided by a 3D vision camera during shooting in an embodiment of the present invention. Figure 2 This is a schematic diagram of the overall structure of the intelligent box-grabbing device guided by a 3D vision camera during shooting from the other side in an embodiment of the present invention. Figure 3 This is a cross-sectional view of the box structure in the grasping state of the intelligent box grasping device guided by a 3D vision camera in an embodiment of the present invention. Figure 4 This is a flowchart of a box-gripping intelligent grasping method based on a 3D vision camera in an embodiment of the present invention.

[0018] In all the accompanying drawings, the same reference numerals denote the same technical features, specifically: 1. First motion joint; 2. Second motion joint; 3. Third motion joint; 4. Fourth motion joint; 5. Fifth motion joint; 6. Sixth motion joint; 7. First docking hole; 8. Second docking hole; 9. Third docking hole; 10. Fourth docking hole; 11. Housing; 12. Camera field of view; 13. Lifting frame; 14. 3D camera; 15. First robotic arm; 16. Second robotic arm; 17. Third robotic arm; 18. Fourth robotic arm; 19. Fifth robotic arm; 20. Base; 21. Locking assembly. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0020] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0021] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0022] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0023] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "over," and "on top" of the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0024] Example: Please see Figures 1-4 The intelligent box-grabbing device and method based on 3D vision camera guidance in the preferred embodiment of the present invention can adapt to boxes 11 in different positions. It can not only accurately identify the docking hole on the top of the box 11, but also adjust the gripping of the box 11 based on the position of the box 11, thereby significantly improving the gripping efficiency of the box 11 while ensuring the gripping stability of the box 11.

[0025] Specifically, in the preferred embodiments of this application, such as Figures 1-3 As shown, the intelligent box-grabbing device based on 3D vision camera guidance includes a multi-degree-of-freedom drive mechanism, a hoisting frame 13, and a processor.

[0026] The multi-degree-of-freedom drive mechanism includes several robotic arms and several movable joints. Movable joints are provided between adjacent robotic arms, and the robotic arms are connected in sequence to form a multi-degree-of-freedom robotic arm. One end of the multi-degree-of-freedom robotic arm is fixed, and the other end is the moving end.

[0027] Furthermore, the hoisting frame 13 is rotatably connected to the moving end, and a number of locking components 21 and a 3D camera 14 are provided on the hoisting frame 13. The distribution of each locking component 21 on the hoisting frame 13 is the same as the distribution of the docking holes on the top surface of the box 11, so that the locking components 21 can be locked in the docking holes when the hoisting frame 13 docks with the box 11. The 3D camera 14 is located on one side of the hoisting frame 13, and is used to collect image information of the docking holes on both sides after the hoisting frame 13 is rotated horizontally.

[0028] Furthermore, the processor is communicatively connected to the movable joints and the 3D camera 14 respectively, to receive image information and drive the movement of each movable joint according to the image information, so as to lock the locking component 21 on the hoisting frame 13 into the docking hole.

[0029] Furthermore, in a preferred embodiment of this application, the multi-degree-of-freedom drive mechanism includes a first robotic arm 15, a second robotic arm 16, a third robotic arm 17, a fourth robotic arm 18, a fifth robotic arm 19, and a base 20.

[0030] A first motion joint 1 is provided between the first robotic arm 15 and the crane frame. A second motion joint 2 is provided between the first robotic arm 15 and the second robotic arm 16. A third motion joint 3 is provided between the second robotic arm 16 and the third robotic arm 17. A fourth motion joint 4 is provided between the third robotic arm 17 and the fourth motion joint 4. A fifth motion joint 5 is provided between the fourth robotic arm 18 and the fifth robotic arm 19. A sixth motion joint 6 is provided between the fifth robotic arm 19 and the base 20.

[0031] Further preferably, in the preferred embodiment of this application, the first movable joint, the third movable joint, and the sixth movable joint are all rotational joints about an axis. Correspondingly, the second movable joint, the fourth movable joint, and the fifth movable joint are all pitch-rotational joints.

[0032] Furthermore, in a preferred embodiment of this application, the end of the six-axis robotic arm is rotatably connected to a hoisting frame 13, and a 3D camera 14 is provided on one side of the hoisting frame 13. The field of view 12 of the 3D camera 14 can simultaneously cover two adjacent docking holes.

[0033] In practical use, by rotating the hoisting frame 13, a single 3D camera 14 can acquire images of all the docking holes. This not only ensures the clarity of the acquired images but also reduces the equipment cost of setting up multiple 3D cameras 14.

[0034] More preferably, when the 3D camera 14 captures the position of the docking hole, the camera field of view 12 of the 3D camera 14 can simultaneously cover two adjacent docking holes while the distance between the 3D camera 14 and the housing 11 is as small as possible, so as to further improve the clarity of the image.

[0035] More preferably, in the preferred embodiment of this application, four docking holes are provided at the four corners of the top surface of the box 11. Correspondingly, the hoisting frame 13 is an I-shaped truss, and four locking components 21 are provided at the four corners of the bottom surface of the hoisting frame 13.

[0036] More specifically, in a preferred embodiment of this application, the locking component 21 includes a rotary lock head and a locking rotation unit.

[0037] The locking rotation unit is fixedly installed on the hoisting frame 13. The rotating end of the locking rotation unit is fixedly connected to one end of the rotating lock head, which is used to drive the rotating lock head to rotate around the axis. The other end of the rotating lock head has a T-shaped block with a square hole. When locking and gripping, the T-shaped block is rotated and embedded in the square hole to form a lock between the hoisting frame 13 and the box 11.

[0038] Furthermore, this application also discloses a method for intelligent grasping of a box 11 guided by a 3D vision camera, comprising the following steps: S1. Collect image information of the four docking holes located at the four corners of the top surface of the housing 11; S2. Obtain the center coordinates of the four docking holes based on the image information of the four docking holes; Preferably, the coordinates of the center positions of the four mating holes are (x, y ... a y a , z a ), (x b y b , zb ), (x c y c , z c ) and (x d y d , z d ); S3. Construct a position matrix of the center of the four docking holes relative to the center point of the top surface of the box 11 based on the coordinates of the center positions of the four docking holes. S4. Calculate the 11th pose T of the box based on the position matrix; Preferably, the pose of the box 11 is T=[x_temp] T y_temp T O_temp T ]; S5. Obtain the pose of the lifting frame based on the pose of the end effector of the six-axis robotic arm; S6. Calculate the control quantity of the spreader frame based on the position and orientation of the box 11, and control the movement of the six-axis robotic arm based on the control quantity of the spreader frame until the spreader frame locks the docking hole.

[0039] Furthermore, in a preferred embodiment of this application, step S1 further includes the following steps: S11. Adjust the attitude of the hoisting frame 13 to a horizontal state, and transport the hoisting frame 13 to the top of the box 11; S12. Acquire image information of the first docking hole 7 and the second docking hole 8 on one side; S13. Control the hoisting frame 13 to rotate 180 degrees and collect image information of the third docking hole 9 and the fourth docking hole 10 on one side.

[0040] More specifically, in a preferred embodiment of this application, the image information includes multiple sets of graphic data from different perspectives, so as to accurately calculate the spatial coordinates of the docking hole using the graphic data of the same docking hole at different angles.

[0041] Furthermore, in a preferred embodiment of this application, S2 includes the following steps: S21. Acquire the spatial coordinates and projection ray direction when the 3D camera 14 captures each set of graphic data; S22. Extract the feature points of each docking hole from each set of graphic data; S23. Based on the spatial coordinates, projection ray direction, and feature points of the 3D camera 14 corresponding to each docking hole, the three-dimensional spatial coordinates of each docking hole are calculated sequentially using a triangulation algorithm.

[0042] Furthermore, in a preferred embodiment of this application, S3 includes the following steps: S31. Calculate the coordinates of the center position O_temp on the top surface of box 11; S32. Calculate the vector of the center position of each docking hole relative to the center position O of the top surface of the box 11; S33. Construct the position matrix of the center of the docking hole relative to the center point of the top surface of the box 11; Furthermore, in a preferred embodiment of this application, the center position O_temp of the top surface of the box 11 is (xo, yo, zo) = ((xa+xb+xc+xd) / 4, (ya+yb+yc+yd) / 4, (za+zb+zc+zd) / 4).

[0043] More preferably, in the preferred embodiment of this application, the position matrix A of the center of the mating hole relative to the center point of the top surface of the housing 11 is:

[0044] In the formula, the center point of the first mating hole 7 is a, and its coordinates are (x, y, y). a y a , z a The center point of the second mating hole 8 is b, and its coordinates are (x, y). b y b , z b The center point of the third mating hole 9 is c, and its coordinates are x. c y c , z c The center point of the fourth mating hole 10 is d, and its coordinates are (x, y). d y d , z d ).

[0045] Furthermore, in a preferred embodiment of this application, S4 includes the following steps: S41. Perform singular value decomposition on the position matrix A: A=USVT In the formula, U and V are orthogonal matrices; S is a diagonal matrix; and the normal vector n corresponds to the last column of V. S42. Let the normal vector n be the z-axis of the upper plane of box 11; S43, Center position of the first mating hole 7a (x) a y a , z a ) and the center position of the mating hole 4b (x b y b , z b The midpoint O of ) ab For ((x) a +x b ) / 2, (y a +y b ) / 2, (z a +zb ) / 2); S44. Select O Oab as the x-axis of the upper plane of box 11, and its vector is x_temp; S45, the pose of box 11 can be represented as T=[x_temp T y_temp T O_temp T ].

[0046] Furthermore, in a preferred embodiment of this application, S6 includes the following steps: S61. Control the movement of the third motion joint 3 according to [x_tempT, y_tempT]; S62. Based on the position coordinates O_tempT, adjust the movements of all joints except the first joint 1, the second joint 2, the fourth joint 4, the fifth joint 5, and the sixth joint 6 to lower and lock the lifting frame to each docking hole.

[0047] The intelligent box-grabbing device and method based on 3D vision camera guidance in this invention has good adaptability, accurate recognition, and high efficiency. By rotating the hoisting frame 13, a single 3D camera 14 can acquire images of all docking holes, ensuring image clarity and reducing the equipment cost of setting up multiple 3D cameras 14. Furthermore, when the 3D camera 14 captures the position of the docking holes, its field of view 12 can simultaneously cover two adjacent docking holes while minimizing the distance between the 3D camera 14 and the box 11, further improving image clarity. Simultaneously, it accurately calculates the pose of the box 11 based on the spatial position matrix of the docking holes, and then accurately controls the movement of each joint based on the pose of the box 11 and the real-time position of the six-axis robotic arm. This enables the six-axis robotic arm to drive the rotating hoist to stably grasp the box 11, demonstrating high promotional value and application prospects.

[0048] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A box-grabbing intelligent grasping device based on 3D vision camera guidance, used to grasp a box with several docking holes on its top surface, characterized in that, include: A multi-degree-of-freedom drive mechanism, comprising several robotic arms and several movable joints; The movable joint is provided between two adjacent robotic arms, and the robotic arms are connected in sequence to form a multi-degree-of-freedom robotic arm; one end of the multi-degree-of-freedom robotic arm is fixed, and the other end of the multi-degree-of-freedom robotic arm is a moving end; A hoisting frame, which is rotatably connected to the moving end, and is provided with several locking components and a 3D camera; The locking components are distributed in the same way as the docking holes on the top surface of the box body, so that when the lifting frame docks with the box body, the locking components can be locked in the docking holes. The 3D camera is set on one side of the hoisting frame and is used to collect image information of the docking holes on both sides after the hoisting frame is rotated horizontally. The processor is communicatively connected to the movable joint and the 3D camera, respectively, for receiving the image information and driving the movable joint to move according to the image information, so as to lock the locking component on the hoisting frame in the docking hole.

2. The intelligent box-grabbing device and method based on 3D vision camera guidance according to claim 1, characterized in that, The multi-degree-of-freedom drive mechanism includes a first robotic arm, a second robotic arm, a third robotic arm, a fourth robotic arm, a fifth robotic arm, and a base; A first motion joint is provided between the first robotic arm and the crane frame; a second motion joint is provided between the first robotic arm and the second robotic arm; A third motion joint is provided between the second and third robotic arms; a fourth motion joint is provided between the third robotic arm and the fourth motion joint; a fifth motion joint is provided between the fourth and fifth robotic arms; and a sixth motion joint is provided between the fifth robotic arm and the base.

3. The intelligent box-grabbing device and method based on 3D vision camera guidance according to claim 1, characterized in that, The top surface of the box body has four docking holes at the four corners. Correspondingly, the hoisting frame is an I-shaped truss, and the bottom surface of the hoisting frame has four locking components at the four corners.

4. The intelligent box-grabbing device and method based on 3D vision camera guidance according to claim 1, characterized in that, The locking assembly includes a rotary lock head and a locking rotary unit; The locking rotation unit is fixedly installed on the hoisting frame, and the rotating end of the locking rotation unit is fixedly connected to one end of the rotating lock head, which is used to drive the rotating lock head to rotate around the axis; The other end of the rotary lock has a T-shaped block, and the mating hole is a square hole, which is used to rotate and embed the T-shaped block into the square hole during locking and gripping, so as to form a lock between the hoisting frame and the box.

5. A method for intelligent box grasping based on 3D vision camera guidance, characterized in that, The box-grabbing intelligent grasping device based on a 3D vision camera, as described in any one of claims 1 to 4, is used to grasp the box. The grasping method includes the following steps: S1. Collect image information of the four docking holes located at the four corners of the top surface of the box; S2. Obtain the center coordinates of the four docking holes based on the image information of the four docking holes; The coordinates of the center positions of the four docking holes are respectively (x a y a , z a ), (x b y b , z b ), (x c y c , z c ) and (x d y d , z d ); S3. Construct a position matrix of the center of the docking hole relative to the center point of the top surface of the box body based on the coordinates of the center positions of the four docking holes; S4. Calculate the box pose T based on the position matrix; Where, box pose T=[x_temp T y_temp T O_temp T ]; S5. Obtain the pose of the lifting frame based on the pose of the end effector of the six-axis robotic arm; S6. Calculate the control amount of the lifting frame based on the position of the lifting frame and the position of the box T, and control the movement of the six-axis robotic arm based on the control amount of the lifting frame until the lifting frame locks the docking hole.

6. The intelligent box-grabbing device and method based on 3D vision camera guidance according to claim 5, characterized in that, S1 includes the following steps: S11. Adjust the posture of the hoisting frame to a horizontal state and transport the hoisting frame to the top of the box; S12. Acquire image information of the first and second docking holes on one side; S13. Control the hoisting frame to rotate 180 degrees and collect image information of the third and fourth docking holes on one side.

7. The intelligent box-grabbing device and method based on 3D vision camera guidance according to claim 6, characterized in that, The image information includes multiple sets of graphic data from different perspectives.

8. The intelligent box-grabbing device and method based on 3D vision camera guidance according to claim 7, characterized in that, S2 includes the following steps: S21. Obtain the spatial coordinates and projection ray direction when the 3D camera captures each set of graphic data; S22. Extract each set of graphic data and find the feature points of each of the said docking holes; S23. Based on the spatial coordinates of the 3D camera corresponding to each docking hole, the direction of the projection ray, and the feature points, the three-dimensional spatial coordinates of each docking hole are calculated sequentially using a triangulation algorithm.

9. The intelligent box-grabbing device and method based on 3D vision camera guidance according to claim 8, characterized in that, S3 includes the following steps: S31. Calculate the coordinates of the center position O_temp on the top surface of the box; S32. Calculate the vector of the center position of each of the said docking holes relative to the center position O of the top surface of the box; S33. Construct the position matrix of the center of the docking hole relative to the center point of the top surface of the box.

10. The intelligent box-grabbing device and method based on 3D vision camera guidance according to claim 5, characterized in that, The position matrix A of the center of the docking hole relative to the center point of the top surface of the box is: In the formula, the center point of the first mating hole is a, and its coordinates are (x, y, y). a y a , z a The center point of the second mating hole is b, and its coordinates are (x, y). b y b , z b The center point of the third mating hole is c, and its coordinates are x. c y c , z c The center point of the fourth mating hole is d, and its coordinates are (x, y). d y d , z d ).

Images