Spatial self-positioning clamping jaw based on vision and using method thereof

Abstract

The invention discloses a vision-based spatial self-positioning clamping jaw and a using method thereof.The vision-based spatial self-positioning clamping jaw comprises a clamping jaw body, a first camera and a second camera are arranged on the clamping jaw body, the focal length of the first camera is larger than that of the second camera, and the field angle of the second camera is larger than that of the first camera; the first camera and the second camera are used for capturing an identification code image on a grabbed object; according to the invention, through the first camera and the second camera with different focal lengths, the identification precision of identification codes at different distances can be effectively compatible and identified; when the distance between the identification code and the camera is too close, the definition of the image acquired by the second camera is high, so that the pose of the grabbed object is judged by adopting the identification code image acquired by the second camera; when the distance between the identification code and the camera is far, the definition of the image acquired by the first camera is high, so that the pose of the grabbed object is judged by adopting the identification code image acquired by the first camera; according to the invention, wider and farther view fields can be considered.

Description

Technical Field

[0001] The present invention relates to the technical field of clamping devices, and in particular to a vision-based spatial self-positioning clamp and a use method thereof. Background Art

[0002] An existing manipulator or gripper is usually equipped with a camera (or camera head), and an identification code (usually an Arcuo code) is set on the grasped object. When the manipulator or gripper grasps the object, the identification code on the grasped object is captured by the camera (or camera head) on the gripper, and the current posture of the grasped object is judged by the posture state of the identification code to assist the gripper in grasping the object.

[0003] However, the existing grippers have the following shortcomings: 1. The current grippers use a single camera (or camera) with a fixed focal length design. When the camera uses the Arcuo code to determine the position, it can only obtain a high-definition identification code image within a certain range of distances. Therefore, it can only have a high recognition accuracy within this specific distance range. Once it leaves this distance range, for example, when the camera is too close or too far from the Arcuo code, the camera (or camera) internal parameters (internal parameters will cause image distortion) and depth of field (depth of field will cause image blur) will cause the image to lose cleanliness and accuracy when the distance is too close or too far, resulting in a serious decrease in recognition accuracy. 2. In the actual working process of the existing grippers, the gripper body will block the camera's field of view, resulting in a large blind spot in the camera's field of view. In some cases, it may cause the camera to be unable to capture the identification code on the grasped object. Summary of the invention

[0004] The purpose of the present invention is to solve the deficiencies in the prior art and provide a vision-based spatial self-positioning clamp and a method of using the same.

[0005] The objective of the present invention is achieved through the following technical solutions: a vision-based spatial self-positioning gripper, comprising a gripper body, a first camera and a second camera are arranged on the gripper body, the focal length of the first camera is greater than the focal length of the second camera, the field of view of the second camera is greater than the field of view of the first camera, and the first camera and the second camera are used to capture identification code images on the gripped object; When only the first camera captures the identification code image, the position and posture of the grasped object are determined by the position and posture of the identification code captured by the first camera; When only the second camera captures the identification code image, the position and posture of the grasped object are determined by the position and posture of the identification code captured by the second camera; When both the first camera and the second camera capture the identification code image, determine whether the identification code image obtained by the first camera is clear; if the identification code image obtained by the first camera is clear, the posture of the grasped object is determined by the posture of the identification code captured by the first camera; if the identification code image obtained by the first camera is blurred, the posture of the grasped object is determined by the posture of the identification code captured by the second camera.

[0006] Preferably, the identification code is an Arcuo code.

[0007] Preferably, the focal length of the first camera is 4 mm, and its field of view angle is 78°; the focal length of the second camera is 2.5 mm, and its field of view angle is 130°.

[0008] Preferably, the clamp body comprises a main body component and an electric cylinder. The main body component is provided with two rocker arm assemblies, the rocker arm assembly comprises an inner rocker arm, an outer rocker arm and a clamping jaw connecting rod. One end of the inner rocker arm and the outer rocker arm are both rotatably connected to the main body component, and the clamping jaw connecting rod is rotatably connected to the inner rocker arm and the outer rocker arm at the same time. The inner rocker arm, the outer rocker arm, the main body component and the clamping jaw connecting rod together constitute a parallelogram mechanism; a guide rail is provided on the main body component, and a slider connecting block is slidably connected to the guide rail. The sliding connecting block is connected to the inner rocker arm through a slider connecting rod, and the two ends of the slider connecting rod are rotatably connected to the sliding connecting block and the inner rocker arm respectively. The electric cylinder is provided on the main body component, and the electric cylinder is connected to the sliding connecting block and drives the sliding connecting block to move toward the guide rail direction; the first camera and the second camera are both fixedly arranged on the main body component.

[0009] Preferably, the main body component includes an upper cover plate and a lower cover plate, an electric cylinder fixing block and a padding block are provided between the upper cover plate and the lower cover plate, and the electric cylinder is arranged in the electric cylinder fixing block; the first camera is arranged on the padding block, and the second camera is arranged on the upper cover plate.

[0010] Preferably, the electric cylinder fixing block is provided with an electric cylinder front end fixing piece and an electric cylinder rear end limiting plate, the electric cylinder front end fixing piece is made of rubber, the front end of the electric cylinder is connected to the electric cylinder front end fixing piece, and the electric cylinder rear end limiting plate is used to limit the tail of the electric cylinder.

[0011] Preferably, the electric cylinder tail end limit plate is made of nylon material.

[0012] Preferably, the first camera is at the same height as the swing arm assembly, and the first camera is located between the two swing arm assemblies; the second camera is higher than the swing arm assembly.

[0013] A method for using a vision-based spatial self-positioning gripper, the specific method is as follows: Acquire an identification code image on the grasped object through the first camera and the second camera; the identification code is an Arcuo code; When only the first camera captures the identification code image, the position and posture of the grasped object are determined by the position and posture of the identification code captured by the first camera; When only the second camera captures the identification code image, the position and posture of the grasped object are determined by the position and posture of the identification code captured by the second camera; When both the first camera and the second camera capture the identification code image, determine whether the identification code image obtained by the first camera is clear; if the identification code image obtained by the first camera is clear, the posture of the grasped object is determined by the posture of the identification code captured by the first camera; if the identification code image obtained by the first camera is blurred, the posture of the grasped object is determined by the posture of the identification code captured by the second camera.

[0014] Preferably, when judging whether the identification code image acquired by the first camera is clear, the specific method is as follows: Step 1: extracting four peripheral corner points of the code area in the identification code image acquired by the first camera, and cutting out a quadrilateral code area image according to the four peripheral corner points; Step 2: Evaluate the fuzziness of the code area image. The specific method is as follows: S1: Detect the number of corner points in the code area image and set the standard single corner sharpness threshold R threshold ; S2: Calculate the blur score value of the code area image through the image blur score function; wherein the image blur score function is: ; Where F is the fuzziness score of the code area image; α is the first coefficient, the value of α is 0.1, β is the second coefficient, the value of β is 0.9; N max is the total number of corner points in the Arcuo code, N is the number of corner points currently detected; R is the clarity value of each point calculated by the Shi-Tomasi corner point detection algorithm; Step 3: Set the judgment standard value Y; If F>Y, it is determined that the identification code image acquired by the first camera is in a clear state; the position and posture of the grasped object are determined by the position and posture of the identification code captured by the first camera; If F≤Y, it is determined that the identification code image acquired by the first camera is in a blurred state, and the position and posture of the grasped object are determined by the position and posture of the identification code captured by the second camera.

[0015] The beneficial effects of the present invention are: 1. The present invention can effectively and compatiblely identify the recognition accuracy of the identification code at different distances through the first camera and the second camera with different focal lengths; when the distance between the identification code and the camera is too close, the clarity of the image obtained by the second camera is higher, so the identification code image obtained by the second camera is used to determine the posture of the grasped object; when the distance between the identification code and the camera is far, the clarity of the image obtained by the first camera is higher, so the identification code image obtained by the first camera is used to determine the posture of the grasped object; through the coordination of the first camera and the second camera, the posture information of the identification code on the grasped object can be more accurately identified, and a wider and farther field of view can also be taken into account.

[0016] 2. In the present invention, the first camera and the second camera are respectively arranged at different positions on the gripper body. The two cameras can respectively obtain the field of view at different positions, effectively reducing the blind spot of the field of view, so that the identification code on the grasped object can be better captured. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a structural schematic diagram of the present invention.

[0018] Figure 2 This is a schematic diagram of the structure of the present invention after removing the upper cover plate and the second camera.

[0019] Figure 3 Schematic diagram of the field of view of the first camera and the second camera in the present invention.

[0020] In the figure: 1. Electric cylinder fixing block; 2. Electric cylinder connecting rod; 3. Slider; 4. Spacer block; 5. Electric cylinder tail end limit plate; 6. Lower cover plate; 7. Upper cover plate; 8. Slider connecting block; 9. Slider connecting rod rotating shaft; 10. Slider connecting rod; 11. Inner swing rod; 12. Outer swing rod; 13. First rotating shaft; 14. Second rotating shaft; 15. Clamp connecting rod; 16. Guide rail; 17. Connecting rod swing rod connecting pin; 18. Electric cylinder; 19. Connecting column; 20. Second bearing; 22. Core male head; 23. First camera; 24. First bearing; 25. Electric cylinder front end fixing piece; 26. Second camera. DETAILED DESCRIPTION

[0021] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present invention.

[0022] Those skilled in the art should understand that, in the disclosure of the present invention, the terms "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the drawings, which are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation. Therefore, the above terms should not be understood as limiting the present invention.

[0023] It is to be understood that the term "one" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of an element may be one, while in another embodiment, the number of the element may be multiple, and the term "one" should not be understood as a limitation on the quantity.

[0024] like Figures 1 to 3 As shown, a vision-based spatial self-positioning gripper includes a gripper body, on which a first camera 23 and a second camera 26 are provided, wherein the focal length of the first camera 23 is greater than that of the second camera 26, and the field of view of the second camera 26 is greater than that of the first camera 23, and the first camera 23 and the second camera 26 are used to capture identification code images on the gripped object; When only the first camera 23 captures the identification code image, the position and posture of the grasped object is determined by the position and posture of the identification code captured by the first camera 23; When only the second camera 26 captures the identification code image, the position and posture of the grasped object is determined by the position and posture of the identification code captured by the second camera 26; When both the first camera 23 and the second camera 26 capture the identification code image, it is determined whether the identification code image acquired by the first camera 23 is clear; if the identification code image acquired by the first camera 23 is clear, the posture of the grasped object is determined by the posture of the identification code captured by the first camera 23; if the identification code image acquired by the first camera 23 is blurred, the posture of the grasped object is determined by the posture of the identification code captured by the second camera 26.

[0025] In the present invention, the identification code adopts Arcuo code.

[0026] In the present invention, two sets of cameras are equipped at the same time, and the two cameras have different focal lengths and field of view (field of view angles). The different focal lengths make the two cameras have different effective recognition distance ranges; the first camera 23 has a longer focal length, and its field of view is relatively small, but it has a longer effective recognition distance; while the second camera 26 has a shorter focal length, and its effective recognition distance is shorter, but its field of view is larger, and the shorter focal length makes its depth of field larger; the present invention can effectively and compatiblely recognize the recognition of identification codes at different distances through the first camera 23 and the second camera 26 with different focal lengths Accuracy; when the distance between the identification code and the camera is too close, the clarity of the image obtained by the second camera 26 is higher, so the identification code image obtained by the second camera 26 is used to determine the posture of the grasped object; when the distance between the identification code and the camera is far, the clarity of the image obtained by the first camera 23 is higher, so the identification code image obtained by the first camera 23 is used to determine the posture of the grasped object; through the cooperation of the first camera 23 and the second camera 26, the posture information of the identification code on the grasped object can be more accurately identified, and a wider and farther field of view can also be taken into account.

[0027] In the present invention, the first camera 23 and the second camera 26 are respectively arranged at different positions on the gripper body, and the two cameras can respectively obtain the field of view at different positions, effectively reducing the blind spot of the field of view, so that the identification code on the grasped object can be better captured.

[0028] The focal length of the first camera 23 is 4 mm, and its field of view is 78°, and the focal length of the second camera 26 is 2.5 mm, and its field of view is 130°. The first camera 23 and the second camera 26 in the present invention are both RGB cameras.

[0029] Specifically, the clamp body includes a main body component and an electric cylinder 18. Two rocker arm assemblies are arranged on the main body component. The rocker arm assemblies include an inner rocker arm 11, an outer rocker arm 12, and a clamp connecting rod 15. One end of the inner rocker arm 11 and the outer rocker arm 12 are both rotatably connected to the main body component. The clamp connecting rod 15 is also rotatably connected to the inner rocker arm 11 and the outer rocker arm 12. The inner rocker arm 11, the outer rocker arm 12, the main body component, and the clamp connecting rod 15 together constitute a parallelogram mechanism; a guide rail 16 is provided on the main body component. A slider connecting block 8 is slidably connected to the guide rail 16. The sliding connecting block is connected to the inner rocker arm 11 through a slider connecting rod 10. The two ends of the slider connecting rod 10 are rotatably connected to the sliding connecting block and the inner rocker arm 11 respectively. The electric cylinder 18 is arranged on the main body component. The electric cylinder 18 is connected to the sliding connecting block and drives the sliding connecting block to move toward the guide rail 16; the first camera 23 and the second camera 26 are both fixedly arranged on the main body component.

[0030] Among them, a slider 3 is provided on the guide rail 16, and the slider 3 is slidably matched with the guide rail 16. The slider connecting block 8 is provided on the slider 3, and the slider connecting block 8 is connected to the guide rail 16 through the slider 3. One end of the inner swing rod 11 and the outer swing rod 12 are respectively connected to the main body through the first rotating shaft 13, and a second bearing 20 is provided between the two ends of the first rotating shaft 13 and the main body; the other ends of the inner swing rod 11 and the outer swing rod 12 are respectively connected to the clamping claw connecting rod 15 through the second rotating shaft 14, and a second bearing 20 is provided between the two ends of the second rotating shaft 14 and the clamping claw connecting rod 15. One end of the slider connecting rod 10 is rotatably connected to the inner swing rod 11 through the connecting rod swing rod connecting pin 17, and the other end of the slider connecting rod 10 is rotatably connected to the slider connecting block 8 through the slider connecting rod rotating shaft 9. The electric cylinder 18 is connected to the slider connecting block 8 through the electric cylinder connecting rod 2.

[0031] When the clamp body grasps an object, the electric cylinder 18 drives the slider connecting block 8 to move along the guide rail 16, thereby realizing the opening or closing of the rocker arm assembly; when the electric cylinder 18 drives the electric cylinder connecting rod 2 to extend, the slider connecting block 8 moves forward, and the slider connecting block 8 drives the slider connecting rod 10 to move when it moves, and the rocker arm assemblies on both sides are driven to open through the slider connecting rod 10; when the electric cylinder 18 drives the electric cylinder connecting rod 2 to retract, the slider connecting block 8 moves backward, and the slider connecting block 8 drives the slider connecting rod 10 to move when it moves, and the rocker arm assemblies on both sides are driven to close through the slider connecting rod 10.

[0032] The clamping jaws of the present invention can realize self-locking of the breakpoint when the power is off. The electric cylinder 18 adopts a micro servo electric cylinder 18, and the electric cylinder 18 integrates a micro motor, a reducer, a screw mechanism, a sensor and a drive control system, which can realize accurate servo control at any position within the travel range, and the electric cylinder 18 has a built-in absolute position sensor, and the position information will not be lost after power failure, and there is no need for zeroing operation. The clamping jaw body integrated with the first camera 23 and the second camera 26 can obtain the precise position and characteristics of the target object through the camera vision system, thereby improving the gripping accuracy of the clamping jaws and the reliability of operation.

[0033] The main body includes an upper cover plate 7 and a lower cover plate 6. An electric cylinder fixing block 1 and a spacer block 4 are provided between the upper cover plate 7 and the lower cover plate 6. The electric cylinder 18 is provided in the electric cylinder fixing block 1. The first camera 23 is provided on the spacer block 4, and the second camera 26 is provided on the upper cover plate 7. A connecting column 19 is provided between the upper cover plate 7 and the lower cover plate 6. A core male connector 22 is provided on the electric cylinder fixing block 1, and the core male connector 22 is used to realize the electrical connection between the electric cylinder 18 and the external control circuit.

[0034] Furthermore, the electric cylinder fixing block 1 is provided with an electric cylinder front end fixing part 25 and an electric cylinder rear end limiting plate 5. The electric cylinder front end fixing part 25 is made of rubber. The front end of the electric cylinder 18 is connected to the electric cylinder front end fixing part 25. The electric cylinder rear end limiting plate 5 is used to limit the tail of the electric cylinder 18.

[0035] The electric cylinder fixing block 1 is provided with a slot for connecting the electric cylinder front end fixing member 25, and the electric cylinder front end fixing member 25 is inserted into the slot on the electric cylinder fixing block 1. The electric cylinder 18 is floatingly connected with the electric cylinder fixing block 1 through the electric cylinder front end fixing member 25. The material of the electric cylinder front end fixing member 25 is rubber. Rubber is a flexible material with a certain deformation ability, which can reduce the requirements for precise alignment during the assembly process of the electric cylinder 18 and improve the convenience and flexibility of assembly.

[0036] When the swing arm assembly on the clamp body is opened (releasing the object), the telescopic rod on the electric cylinder 18 extends forward. At this time, the tail of the electric cylinder 18 presses against the limit plate 5 at the tail end of the electric cylinder, and the electric cylinder 18 is limited by the limit plate 5 at the tail end of the electric cylinder; when the swing arm assembly on the clamp body is closed (grasping the object), the telescopic rod on the electric cylinder 18 retracts. At this time, the electric cylinder 18 will apply a reaction force toward the front of the clamp to the front end fixing piece 25 of the electric cylinder. If the clamping force of the clamp is too large (that is, the driving force of the electric cylinder 18 is too large), the front end fixing piece 25 of the electric cylinder made of rubber can be deformed to a certain extent. The deformation of the front end fixing piece 25 of the electric cylinder can prevent the clamping force from increasing to a certain extent, thereby reducing the probability of the object being damaged by the excessive clamping force, and having a certain protective effect on the grasped object.

[0037] In this embodiment, the electric cylinder tail end limit plate 5 is made of nylon material. Nylon material has good insulation performance, ensuring good insulation between the electric cylinder 18 tail and the electric cylinder fixing block 1 to prevent current from passing directly.

[0038] The first camera 23 is at the same height as the swing bar assembly, and the first camera 23 is located between the two swing bar assemblies; the second camera 26 is higher than the swing bar assembly. The first camera 23 and the second camera 26 are respectively set at different heights, so that the first camera 23 and the second camera 26 can respectively obtain the field of view at different heights, and the two cameras cooperate to effectively reduce the blind area of ​​the camera's field of view.

[0039] The Arcuo code used in the present invention is a black and white image, which has 31 corner points in total. The corner points on the Arcuo code refer to the vertex positions at the right-angle turns between the black area and the white area.

[0040] A method for using a vision-based spatial self-positioning gripper, the specific method is as follows: The identification code image on the grasped object is obtained by the first camera and the second camera; the identification code is an Arcuo code; considering that there may be multiple Arcuo codes in the real environment, it is necessary to determine in advance which Arcuo code is used as the identification object, and at the same time, the size of the Arcuo code is preset in the control system; When only the first camera captures the identification code image, the position and posture of the grasped object are determined by the position and posture of the identification code captured by the first camera; When only the second camera captures the identification code image, the position and posture of the grasped object are determined by the position and posture of the identification code captured by the second camera; When both the first camera and the second camera capture the identification code image, determine whether the identification code image obtained by the first camera is clear; if the identification code image obtained by the first camera is clear, the posture of the grasped object is determined by the posture of the identification code captured by the first camera; if the identification code image obtained by the first camera is blurred, the posture of the grasped object is determined by the posture of the identification code captured by the second camera.

[0041] When determining whether the identification code image acquired by the first camera is clear, the specific method is as follows: Step 1: extract the four outer corner points of the code area in the identification code image acquired by the first camera, and cut out a quadrilateral code area image according to the four outer corner points; since the Arcuo code used in the present invention is a black and white image, the processing step of converting the color image to a grayscale image is omitted.

[0042] Step 2: Evaluate the fuzziness of the code area image. The specific method is as follows: S1: Detect the number of corner points in the code area image and set the standard single corner sharpness threshold R threshold The present invention adopts Shi-Tomasi corner detection algorithm to calculate the sharpness of black and white intersection points. When the detected sharpness is greater than the maximum corner sharpness threshold R threshold When , it is determined that there is a corner point at this location; in this embodiment, the maximum threshold value of the corner point sharpness R threshold The default is 255.

[0043] S2: Calculate the blur score value of the code area image through the image blur score function; wherein the image blur score function is: ; Where F is the fuzziness score of the code area image; α is the first coefficient, the value of α is 0.1, β is the second coefficient, the value of β is 0.9; N max is the total number of corner points in the Arcuo code, N is the number of corner points currently detected; R is the clarity value of each point calculated by the Shi-Tomasi corner point detection algorithm, and its value range is 0 to 255.

[0044] Step 3: Set the judgment standard value Y; If F>Y, it is determined that the identification code image acquired by the first camera is in a clear state; the position and posture of the grasped object are determined by the position and posture of the identification code captured by the first camera; If F≤Y, it is determined that the identification code image acquired by the first camera is in a blurred state, and the position and posture of the grasped object are determined by the position and posture of the identification code captured by the second camera.

[0045] The judgment standard value Y in the present invention is set to 0.55.

[0046] The present invention is not limited to the above-mentioned optimal implementation mode. Anyone can derive other various forms of products under the inspiration of the present invention. However, no matter what changes are made in the shape or structure, all technical solutions that are the same or similar to those of the present application fall within the protection scope of the present invention.

Claims

1. A vision-based spatial self-positioning gripper, characterized in that: The clamp body comprises a first camera and a second camera, the first camera has a focal length greater than the second camera, the second camera has a field of view greater than the first camera, and the first camera and the second camera are used to capture identification code images on the object being grasped; When only the first camera captures the identification code image, the position and posture of the grasped object are determined by the position and posture of the identification code captured by the first camera; When only the second camera captures the identification code image, the position and posture of the grasped object are determined by the position and posture of the identification code captured by the second camera; When both the first camera and the second camera capture the identification code image, determine whether the identification code image obtained by the first camera is clear; if the identification code image obtained by the first camera is clear, the posture of the grasped object is determined by the posture of the identification code captured by the first camera; if the identification code image obtained by the first camera is blurred, the posture of the grasped object is determined by the posture of the identification code captured by the second camera.

2. The vision-based spatial self-positioning gripper according to claim 1, characterized in that: The identification code is an Arcuo code.

3. The vision-based spatial self-positioning gripper according to claim 1, characterized in that: The focal length of the first camera is 4 mm, and its field of view angle is 78°; the focal length of the second camera is 2.5 mm, and its field of view angle is 130°.

4. The vision-based spatial self-positioning gripper according to claim 1, characterized in that: The clamp body includes a main body component and an electric cylinder. Two rocker arm assemblies are arranged on the main body component. The rocker arm assembly includes an inner rocker arm, an outer rocker arm and a clamping jaw connecting rod. One end of the inner rocker arm and the outer rocker arm are both rotatably connected to the main body component. The clamping jaw connecting rod is also rotatably connected to the inner rocker arm and the outer rocker arm. The inner rocker arm, the outer rocker arm, the main body component and the clamping jaw connecting rod together constitute a parallelogram mechanism. A guide rail is provided on the main body component. A slider connecting block is slidably connected to the guide rail. The sliding connecting block is connected to the inner rocker arm through a slider connecting rod. The two ends of the slider connecting rod are rotatably connected to the sliding connecting block and the inner rocker arm respectively. The electric cylinder is arranged on the main body component. The electric cylinder is connected to the sliding connecting block and drives the sliding connecting block to move toward the guide rail direction. The first camera and the second camera are both fixedly arranged on the main body component.

5. The vision-based spatial self-positioning gripper according to claim 4, characterized in that: The main body component includes an upper cover plate and a lower cover plate. An electric cylinder fixing block and a padding block are arranged between the upper cover plate and the lower cover plate. The electric cylinder is arranged in the electric cylinder fixing block. The first camera is arranged on the padding block, and the second camera is arranged on the upper cover plate.

6. The vision-based spatial self-positioning gripper according to claim 5, characterized in that: The electric cylinder fixing block is provided with an electric cylinder front end fixing piece and an electric cylinder rear end limiting plate. The electric cylinder front end fixing piece is made of rubber. The front end of the electric cylinder is connected to the electric cylinder front end fixing piece. The electric cylinder rear end limiting plate is used to limit the tail of the electric cylinder.

7. The vision-based spatial self-positioning gripper according to claim 6, characterized in that: The electric cylinder tail end limit plate is made of nylon material.

8. The vision-based spatial self-positioning gripper according to claim 1, characterized in that: The first camera is at the same height as the swing arm assembly, and the first camera is located between the two swing arm assemblies; the second camera is higher than the swing arm assembly.

9. A method for using a vision-based space self-positioning gripper, based on the vision-based space self-positioning gripper according to any one of claims 1 to 8, characterized in that: The specific method is as follows: Acquire an identification code image on the grasped object through the first camera and the second camera; the identification code is an Arcuo code; When only the first camera captures the identification code image, the position and posture of the grasped object are determined by the position and posture of the identification code captured by the first camera; When only the second camera captures the identification code image, the position and posture of the grasped object are determined by the position and posture of the identification code captured by the second camera; When both the first camera and the second camera capture the identification code image, determine whether the identification code image obtained by the first camera is clear; if the identification code image obtained by the first camera is clear, the posture of the grasped object is determined by the posture of the identification code captured by the first camera; if the identification code image obtained by the first camera is blurred, the posture of the grasped object is determined by the posture of the identification code captured by the second camera.

10. The method for using the vision-based spatial self-positioning gripper according to claim 9, characterized in that: When judging whether the identification code image acquired by the first camera is clear, the specific method is as follows: Step 1: extracting four peripheral corner points of the code area in the identification code image acquired by the first camera, and cutting out a quadrilateral code area image according to the four peripheral corner points; Step 2: Evaluate the fuzziness of the code area image. The specific method is as follows: S1: Detect the number of corner points in the code area image and set the standard single corner sharpness threshold R threshold ; S2: Calculate the blur score value of the code area image through the image blur score function; wherein the image blur score function is: ； Where F is the fuzziness score of the code area image; α is the first coefficient, the value of α is 0.1, β is the second coefficient, the value of β is 0.9; N max is the total number of corner points in the Arcuo code, N is the number of corner points currently detected; R is the clarity value of each point calculated by the Shi-Tomasi corner detection algorithm; Step 3: Set the judgment standard value Y; If F>Y, it is determined that the identification code image acquired by the first camera is in a clear state; the position and posture of the grasped object are determined by the position and posture of the identification code captured by the first camera; If F≤Y, it is determined that the identification code image acquired by the first camera is in a blurred state, and the position and posture of the grasped object are determined by the position and posture of the identification code captured by the second camera.

Images