Nuclear object pairing device

Abstract

The utility model relates to a nuclear object pairing device which comprises a base, a gantry driving module, a camera module, a detection platform and an industrial personal computer are connected to the base, the gantry driving module is connected with a gripper connecting plate, and a flexible gripper is installed on the gripper connecting plate. The detection platform comprises a transparent platform body and a driving motor for driving the transparent platform body to rotate; the camera module comprises a first camera, a second camera and a third camera, the first camera is fixedly installed on the upper surface of the base and corresponds to the transparent table body, the second camera is fixedly installed in the base and located below the transparent table body, and the third camera is fixedly installed above the transparent table body; the camera module is electrically connected with the industrial personal computer, and the industrial personal computer is used for pairing the to-be-paired objects according to the collected images of the to-be-paired objects. According to the invention, automatic pairing of kernel objects is realized, and the pairing efficiency and precision are ensured.

Description

Technical Field

[0001] This application relates to the technical field of pairing nuclear objects, and in particular to a pairing device for nuclear objects. Background Technology

[0002] In the cultural market, including antique and collectible items, pairing core-type objects requires comprehensive consideration of multiple dimensions such as outline shape, texture, size, and surface quality. Traditional methods rely on human experience, using visual comparison or measurement tools to achieve pairing. However, manual pairing is highly dependent on the accumulated experience of the appraisers. Different appraisers have subjective differences in their criteria for judging "similarity," resulting in poor consistency in pairing results. This is time-consuming and labor-intensive, making it difficult to meet the needs of large-scale transactions, and suffers from technical defects such as strong subjectivity and low efficiency.

[0003] Therefore, there is an urgent need for a technical solution that can automate the pairing of walnut objects to overcome the efficiency and accuracy bottlenecks in judging the contour shape, size and other dimensional features of walnut objects during traditional manual pairing, and promote the standardization and intelligent upgrading of walnut pairing. Utility Model Content

[0004] In order to achieve automated pairing of nuclear objects and ensure pairing efficiency and accuracy, this application provides a pairing device for nuclear objects.

[0005] A pairing device for nuclear objects includes a base, on which a gantry drive module, a camera module, a detection platform, and an industrial control computer are connected. The gantry drive module is connected to a gripper connecting plate, on which a flexible gripper is mounted. The detection platform includes a transparent stage and a drive motor for driving the transparent stage to rotate. The gantry drive module drives the flexible gripper to the position of the object to be paired, and after the flexible gripper completes the gripping of the object, drives the flexible gripper to the position of the transparent stage, so that the flexible gripper places the object to be paired on the transparent stage.

[0006] The camera module includes a first camera, a second camera, and a third camera. The first camera is fixedly installed on the upper surface of the base and correspondingly positioned to the transparent platform. The second camera is fixedly installed inside the base and located below the transparent platform. The third camera is fixedly installed above the transparent platform. The first camera is used to capture a side image of the object to be paired on the transparent platform. The second camera is used to capture a bottom image of the object to be paired on the transparent platform. The third camera is used to capture a top image of the object to be paired on the transparent platform.

[0007] The camera module is electrically connected to the industrial control computer, which is used to pair each object to be paired based on the images of each object collected.

[0008] By adopting the above technical solution, a flexible gripper can automatically grasp the objects to be paired onto a transparent platform. The industrial control computer obtains the size and contour of the objects to be paired and performs pairing based on the size and contour, realizing a fully automated pairing production line. The camera module can collaboratively cover the 360° features of the objects to be paired from three perspectives, reducing the blind spot problem of a single camera, improving pairing accuracy, and achieving automated pairing of core objects, ensuring pairing efficiency and precision.

[0009] Optionally, the base is connected to an upper housing, and the upper housing has a material conveying port on its side; a material conveying guide rail is fixedly connected to the upper surface of the base, the material conveying guide rail extends from the material conveying port toward the side close to the transparent platform, and a tray is slidably connected to the material conveying guide rail, the tray including multiple independent placement slots, each of which stores an object to be paired.

[0010] By adopting the above technical solution, the nuclear object pairing device can form a semi-enclosed protective structure. The upper shell can block external dust and other interference, protecting internal structures such as camera modules and flexible grippers. Objects to be paired are loaded in batches onto a pallet through a feeding port. The pallet transports multiple loaded objects to the target location along a feeding guide rail, facilitating the flexible gripper to sequentially grasp the objects in each slot for pairing and detection.

[0011] Optionally, the transparent stage includes a first area stage and a second area stage, wherein the first area stage is made of transparent material, the second area stage is made of frosted material, and the arithmetic mean deviation of the profile in the surface roughness parameter of the second area stage is 0.4 micrometers to 0.8 micrometers.

[0012] By adopting the above technical solution, the second area stage, with its rough surface, achieves diffuse light reflection, reducing background light interference when capturing top images. The first area stage, made of a transparent material, ensures accurate acquisition of bottom images.

[0013] Optionally, a spring clip is provided between the tray and the base. The spring clip includes a first clip body and a second clip body. The first clip body is fixedly installed on the base, and the second clip body is fixedly installed on the end face of the tray. When the tray slides along the material guide rail to the target position, the first clip body engages with the second clip body. A micro switch sensor is installed on the upper surface of the base. The micro switch sensor is used to detect whether the tray has reached the target position. The micro switch sensor is electrically connected to the industrial control computer.

[0014] By adopting the above technical solution, when the pallet slides along the guide rail to the target position, the elastic arm of the first buckle is compressed and deformed. After the second buckle slides in, the spring returns to its original deformation, achieving self-locking and eliminating pallet displacement deviation. When the pallet reaches the target position, it triggers the microswitch sensor, which sends an electrical signal to the industrial control computer. The industrial control computer can then trigger subsequent actions such as the flexible gripper automatically grasping the object to be paired. Through the real-time feedback of the microswitch sensor, secondary verification of the pallet position is achieved.

[0015] Optionally, a first light source is provided at the first camera, a second light source is provided at the second camera, and a third light source is provided at the third camera.

[0016] By adopting the above technical solution and setting up a first light source, a second light source, and a third light source, it is convenient to extract surface texture or contour.

[0017] Optionally, a light-shielding plate is provided on the side of the transparent platform away from the first camera. The light-shielding plate is fixedly connected to the base and is positioned facing the first camera.

[0018] By adopting the above technical solution, the light-shielding plate can form a physical isolation barrier. The light-shielding plate is set facing the first camera, blocking stray ambient light from entering the lens field of view of the first camera, thus reducing background interference in the acquired side images.

[0019] Optionally, a fourth camera is installed on the flexible gripper, and the fourth camera is electrically connected to the industrial control computer; the fourth camera is used to collect image data of the object to be paired after the flexible gripper grasps the object to be paired, and transmit the image data to the industrial control computer.

[0020] By adopting the above technical solution, after the flexible gripper completes the grasping of the object to be paired, the image data collected by the fourth camera can be used to adjust the pose of the flexible gripper and the object to be paired, ensuring that the object to be paired enters the transparent platform in a stable and low-damage state for subsequent processes.

[0021] Optionally, a plurality of support rods are inserted into the transparent platform, each support rod extending in a direction perpendicular to the plane of the transparent platform, each support rod being connected to a lifting drive assembly, and each lifting drive assembly being connected to the industrial control computer.

[0022] By adopting the above technical solution, the height-adjustable support rod can dynamically match the shape of the base, enabling the object to be paired to maintain the target posture required for image acquisition, thereby improving the accuracy of side image acquisition. Attached Figure Description

[0023] Figure 1This is a structural block diagram of a nuclear object pairing device according to one embodiment of this application.

[0024] Figure 2 This is one embodiment of the present application. Figure 1 Enlarged diagram of point A in the middle.

[0025] Figure 3 This is a schematic diagram illustrating the second camera mounting structure in one embodiment of this application.

[0026] Figure 4 This is a schematic diagram illustrating the connection relationship between the upper shell and the base in one embodiment of this application.

[0027] Figure 5 This is a top view of the transparent platform in one embodiment of this application.

[0028] Figure 6 This is a structural block diagram illustrating the connection relationship between the industrial control computer and the joint drive component in one embodiment of this application.

[0029] Figure 7 This is a structural block diagram illustrating the connection relationship between the industrial control computer and the joint drive component in another embodiment of this application.

[0030] Figure 8 This is a schematic diagram illustrating the structure of the support rod in one embodiment of this application.

[0031] Figure 9 This is a structural block diagram illustrating the connection relationship between the industrial control computer and the lifting drive assembly in one embodiment of this application.

[0032] Figure 10 This is a schematic flowchart of a method for pairing nuclear objects using a nuclear object pairing device, according to one embodiment of this application.

[0033] In the diagram, 1. Base; 11. First support; 111. First light source; 12. Second support; 13. Third support; 131. Third light source; 14. Second light source; 15. Upper shell; 151. Material inlet; 16. Material guide rail; 17. Pallet; 171. Placement slot; 18. Spring buckle; 19. Light shield; 2. Gantry drive module; 21. Gripper connecting plate; 3. Industrial computer; 31. Piezoresistive sensor network; 32. Six-dimensional force sensor; 33. Lifting drive assembly; 34. Miniature piezoresistive film sensor; 35. Micro switch sensor; 4. Flexible gripper; 41. Joint drive assembly; 42. Fourth camera; 5. Transparent platform; 51. First area platform; 52. Second area platform; 53. Support rod; 6. First camera; 7. Second camera; 8. Third camera. Detailed Implementation

[0034] The principles and features of this utility model are described below. The examples given are only for explaining this utility model and are not intended to limit the scope of this utility model.

[0035] The present application will be further described in detail below with reference to the accompanying drawings.

[0036] like Figures 1 to 3 As shown in the figure, this application provides a nuclear object pairing device, including a base 1. A gantry drive module 2, a camera module, a detection platform, and an industrial control computer 3 are connected to the base 1. The gantry drive module 2 is connected to a gripper connecting plate 21, and a flexible gripper 4 is mounted on the gripper connecting plate 21. The detection platform includes a transparent platform 5 and a drive motor for driving the transparent platform 5 to rotate. The gantry drive module drives the flexible gripper 4 to the position of the object to be paired, and after the flexible gripper 4 completes the gripping of the object, it drives the flexible gripper 4 to the position of the transparent platform 5, so that the flexible gripper 4 places the object to be paired on the transparent platform 5.

[0037] like Figure 1 As shown, in this embodiment, the gantry drive module 2 includes an X-axis drive module, a Y-axis drive module, and a Z-axis drive module. The X-axis drive module may include two parallel X-axis guide rails, a first drive mechanism, and a first moving mechanism. The two X-axis guide rails may be fixed to the upper surface of the base 1, respectively. The first drive mechanism may consist of a first servo motor, a reducer, and a first transmission component. The first transmission component may be a gear rack or a synchronous pulley. The output shaft of the first servo motor may be directly connected to the first transmission component via a coupling.

[0038] The first moving mechanism may include an X-axis moving base plate and a first slider. The X-axis moving base plate is fixedly connected to the first slider, and the first slider is connected to a first transmission component. The X-axis moving base plate can be slidably connected to the X-axis guide rail via the first slider. The X-axis moving base plate can be made of aluminum alloy or a steel frame, and its surface is provided with mounting interfaces to support the Y-axis drive module. The X-axis module serves as the basis for the lateral movement of the gantry drive module 2. It drives the first transmission component to transmit power to the first slider via a first servo motor, thereby achieving high-precision linear motion of the Y-axis drive module along the X-axis.

[0039] The Y-axis drive module may include a Y-axis guide rail, a second drive mechanism, and a second moving mechanism. The Y-axis guide rail spans two X-axis moving base plates and can be made of aluminum alloy or steel beams. The second drive mechanism includes a second servo motor and a second transmission component. The second transmission component can be a ball screw or a linear motor, and the output shaft of the second servo motor is directly connected to the second transmission component.

[0040] The second moving structure may include a Y-axis moving base plate and a second slider. The Y-axis moving base plate is fixedly connected to the second slider, and the second slider is connected to the second transmission component. The Y-axis moving base plate can be slidably connected to the Y-axis guide rail through the second slider.

[0041] In this embodiment, the second transmission component can adopt a double synchronous pulley or long shaft linkage design to ensure that both ends of the Y-axis guide rail can move synchronously. The Y-axis drive module can realize the longitudinal movement of the gantry structure and ensure the stability of the load under high-speed movement.

[0042] like Figure 1 As shown, the Z-axis drive module may include a Z-axis guide rail, a gripper connecting plate 21, and a third servo motor. The Z-axis guide rail is fixedly connected to the Y-axis guide rail, the gripper connecting plate 21 is connected to the output shaft of the third servo motor, and the gripper connecting plate 21 is slidably connected to the Z-axis guide rail. The flexible gripper 4 is mounted on the gripper connecting plate 21. Driven by the third servo motor, the flexible gripper 4 can move vertically along the Z-axis guide rail via the gripper connecting plate 21.

[0043] In this embodiment, the contact surface between the flexible gripper 4 and the nuclear object can be a flexible layer made of flexible materials such as silicone or elastic polymers (e.g., TPU).

[0044] like Figure 1 and Figure 3 As shown, in this embodiment, the camera module includes a first camera 6, a second camera 7, and a third camera 8. The first camera 6 is fixedly installed on the upper surface of the base 1 and is correspondingly arranged with the transparent platform 5. The second camera 7 is fixedly installed inside the base 1 and located below the transparent platform 5. The third camera 8 is fixedly installed above the transparent platform 5. The first camera 6 is used to capture side images of the object to be paired on the transparent platform 5. The second camera 7 is used to capture bottom images of the object to be paired on the transparent platform 5. The third camera 8 is used to capture top images of the object to be paired on the transparent platform 5.

[0045] The camera module is electrically connected to the industrial control computer 3. The industrial control computer 3 is used to obtain the size and outline of each object to be paired based on the acquired images of each object to be paired, so as to pair each object to be paired based on the size and outline.

[0046] The first camera 6 is a side-mounted imaging unit. A first bracket 11 is fixedly connected to the upper surface of the base 1, and the first camera 6 is mounted on the first bracket 11. The optical axis of the first camera 6 is parallel to the upper surface of the transparent stage 5. The first camera 6 can be an industrial-grade camera that is resistant to high temperatures (e.g., -20℃ to 80℃) and electromagnetic interference.

[0047] In this embodiment, the drive motor can be a servo motor, which can be paired with an encoder to achieve angle control and support 360° stepless rotation. The base 1 has a hollow structure, and the drive motor can be fixedly installed inside the base 1. The output shaft of the drive motor is connected to a driving gear, and the transparent platform 5 is connected to a driven gear. The driving gear and the driven gear mesh, thereby enabling the drive motor to rotate the transparent platform 5. For example, the drive motor can rotate the transparent platform 5 at a preset speed, synchronously triggering the first camera 6 to acquire side images of the object to be paired at time intervals. By driving the transparent platform 5 to rotate, the first camera 6 can acquire side images of the object to be paired from different angles, improving the pairing accuracy of the object.

[0048] like Figure 1 and Figure 3 As shown, the second camera 7 is a bottom imaging unit. The second bracket 12 is fixedly connected inside the base 1. The second camera 7 is installed on the second bracket 12, so that the second camera 7 is embedded inside the base 1 and located directly below the transparent stage 5. The optical axis of the second camera 7 penetrates the transparent stage 5 vertically upward.

[0049] In this embodiment, the second camera 7 can be equipped with a macro lens (e.g., working distance ≤ 50mm), which can clearly capture the tiny structures on the bottom of objects (e.g., grooves, scratches). The second camera 7 can be a camera with a dustproof and sealed structure. The body of the second camera 7 adopts a multi-layer sealing process, and key parts (e.g., interfaces or lens mounts) are sealed with rubber gaskets or silicone sealing rings, reducing the possibility of dust falling into the base 1 and contaminating the lens.

[0050] In this embodiment, the transparent platform 5 facilitates the precise acquisition of the bottom image of the object to be paired by the second camera 7, thereby improving the accuracy of the pairing of the objects.

[0051] like Figure 1 As shown, the third camera 8 is a top imaging unit. A third bracket 13 is fixedly connected to the upper surface of the base 1. The third camera 8 is mounted on the third bracket 13, so that the third camera 8 is suspended above the transparent stage 5, and the optical axis of the third camera 8 is vertically downward and aligned with the center of the transparent stage 5. In this embodiment, the third camera 8 can be equipped with a zoom lens (e.g., optical zoom 5X) to facilitate adaptation to the height differences of the objects to be paired.

[0052] In this embodiment, the first camera 6, the second camera 7, and the third camera 8 can be connected to the industrial control computer 3 via gigabit Ethernet or fiber optic interfaces, supporting synchronous trigger acquisition and direct transmission of RAW data. The nuclear object pairing device can also employ an independent redundant power supply module with built-in overvoltage / overcurrent protection circuits, and reduce signal interference through electromagnetically shielded cables (e.g., twisted-pair shielded cables). The industrial control computer 3 is equipped with a multi-threaded image processing algorithm to achieve real-time registration (based on feature point matching algorithms, such as SIFT) and feature fusion (e.g., color, texture, geometric topology) of three-view images. An object feature library is established using a deep learning model (e.g., convolutional neural network), a similarity threshold is calculated, and the optimal pairing result is output (e.g., a pairing result with a confidence level ≥ 99% can be considered the optimal pairing result). When the first camera 6, the second camera 7, and the third camera 8 are all 3D cameras, the industrial control computer 3 can also perform 3D point cloud reconstruction based on stereo vision algorithms.

[0053] In this embodiment, the flexible gripper 4 can automatically grasp the object to be paired onto the transparent platform 5 for size and contour recognition, realizing a fully automated pairing production line. The camera module can cover the 360° features of the object to be paired from three perspectives, reducing the blind spot problem of a single camera, improving the pairing accuracy, realizing automated pairing of core objects, and ensuring pairing efficiency and accuracy.

[0054] like Figure 1 and Figure 4 As shown, in this embodiment, the base 1 is connected to an upper housing 15, and the upper housing 15 has a material conveying port 151 on its side; a material conveying guide rail 16 is fixedly connected to the upper surface of the base 1, and the material conveying guide rail 16 extends from the material conveying port 151 toward the side close to the transparent platform 5. A tray 17 is slidably connected to the material conveying guide rail 16, and the tray 17 includes multiple independent placement slots 171, each of which contains an object to be paired.

[0055] The base 1 is connected to the upper housing 15. The upper housing 15 has a material inlet 151 on its side. The material inlet 151 serves as the entry and exit channel for the object to be paired, so that the nuclear object pairing device can form a semi-enclosed protective structure. The upper housing 15 can block external dust and other interference, and protect internal structures such as camera modules and flexible grippers 4.

[0056] The material conveying guide rail 16 is fixed to the upper surface of the base 1 and extends from the material conveying port 151 to the side of the transparent platform 5. The tray 17 can move on the material conveying guide rail 16 via a linear slide rail. Multiple independent placement slots 171 can be arranged in a honeycomb or matrix pattern, etc. Each placement slot 171 can correspond to a coordinate position in the tray 17. For example, the coordinate position of a certain placement slot 171 can be represented as x row x column.

[0057] The objects to be paired are loaded in batches onto the pallet 17 through the feeding port 151. The pallet 17 transports the loaded objects to the target position along the feeding guide rail 16, so that the flexible gripper 4 can sequentially grab the objects in each placement slot 171 for pairing detection. The target position is the preset position where the pallet 17 slides into place.

[0058] The multi-slot design 171 supports simultaneous inspection of multiple objects. Combined with the rapid response of the gantry drive module 2, the tray 17 of the multi-slot 171 enables continuous feeding, carrying multiple objects to be paired at once (e.g., ornamental walnuts), thus improving operational efficiency. The coordinated movement of the material guide rail 16 and the tray 17 replaces manual handling, reducing non-inspection time. The rigid connection of the material guide rail 16 ensures the stability of the tray 17's movement path, reducing the possibility of the flexible gripper 4 failing to grasp due to tray 17 offset.

[0059] like Figure 1 and Figure 2 As shown, in this embodiment, a spring clip 18 is provided between the tray 17 and the base 1. The spring clip 18 includes a first clip body and a second clip body. The first clip body is fixedly installed on the base 1, and the second clip body is fixedly installed on the end face of the tray 17. When the tray 17 slides along the material guide rail 16 to the target position, the first clip body engages with the second clip body. A micro switch sensor 35 is installed on the upper surface of the base 1. The micro switch sensor 35 is used to detect whether the tray 17 has reached the target position. The micro switch sensor 35 is electrically connected to the industrial control computer 3.

[0060] The first buckle is fixedly connected to the base 1 and can be made of elastic steel sheet or shape memory alloy. A guide slope can be provided on its surface to guide the second buckle to slide in. The second buckle is fixedly connected to the end face of the tray 17 and includes a slot or concave structure, forming an interference fit with the first buckle. When the tray 17 slides along the guide rail to the target position, the elastic arm of the first buckle is compressed and deformed. After the second buckle slides in, the spring returns to its original deformation, achieving self-locking and eliminating displacement deviation of the tray 17.

[0061] The microswitch sensor 35 can be IP67 protected, with built-in bimetallic contacts and a triggering force of no more than 0.5N. The microswitch sensor 35 can be installed on the upper surface of the base 1 near the spring clip 18. The microswitch sensor 35 ensures the positional accuracy of the tray 17. When the tray 17 reaches the target position and triggers the microswitch sensor 35, it sends a voltage signal to the industrial control computer 3. The industrial control computer 3 can then trigger subsequent actions such as the flexible gripper 4 automatically grasping the object to be paired. Through real-time feedback from the microswitch sensor 35, a secondary verification of the tray 17's position is achieved, avoiding the risk of "false locking".

[0062] Through the electrical control linkage between the micro-switch sensor 35 and the industrial control computer 3, "trigger upon arrival" is achieved, establishing a fully automated workflow. The mechanical self-locking of the spring clip 18 provides physical redundancy for the electrical control system, preventing accidental displacement.

[0063] like Figure 5 As shown, in this embodiment, the transparent platform 5 includes a first region platform 51 and a second region platform 52. The first region platform 51 is a platform made of transparent material, and the second region platform 52 is a platform made of frosted material. The arithmetic mean deviation of the profile in the surface roughness parameter of the second region platform 52 is 0.4 micrometers to 0.8 micrometers.

[0064] The transparent stage 5 is divided into sections. The first section stage 51 can be made of quartz glass or reinforced acrylic with high light transmittance (e.g., light transmittance ≥ 95%), which reduces the refraction loss of the detection light. The first section stage 51 is the area that needs to be optically detected, such as the corresponding area for image acquisition of the bottom of a walnut.

[0065] The second stage 52, serving as a diffuse reflection area, reduces the possibility of inaccurate top image acquisition caused by reflections from the object to be paired. Simultaneously, the frosted material of the second stage 52 provides a uniform background lighting environment. The second stage 52 can be precisely shaped through laser engraving or machining, forming regularly arranged micron-level depressions to ensure uniform depth and distribution. The arithmetic mean deviation (Ra) of the second stage 52's profile can be controlled between 0.4 and 0.8 micrometers. When the Ra value is below 0.4 micrometers, diffuse reflection is insufficient, easily leaving reflective spots. When the Ra value is above 0.8 micrometers, the transmittance of the second stage 52 drops sharply, limiting light transmission.

[0066] In this embodiment, the second area platform 52 is made with a rough surface to achieve diffuse light reflection, reducing background light interference when capturing top images. The first area platform 51 is made of a transparent material to ensure accurate acquisition of bottom images.

[0067] like Figure 1 and Figure 3 As shown, in this embodiment, a first light source 111 is provided at the first camera 6, a second light source 14 is provided at the second camera 7, and a third light source 131 is provided at the third camera 8.

[0068] The first light source 111 can be fixedly mounted on the first bracket 11. The first light source 111 can be a ring light source and is coaxially arranged with the first camera 6. The second light source 14 can be arranged below the transparent stage 5, and the third light source 14 can be coaxially arranged with the second camera 7. The third light source 131 can be fixedly mounted on the third bracket 13 and is coaxially arranged with the third camera 8. In this embodiment, by setting the first light source 111 and the third light source 131, it is convenient to extract surface texture or contour.

[0069] In this embodiment, a light-shielding plate 19 is provided on the side of the transparent platform 5 away from the first camera 6. The light-shielding plate 19 is fixedly connected to the base 1 and is arranged facing the first camera 6.

[0070] The light-shielding plate 19 is installed on the side of the transparent stage 5 away from the first camera 6 and is fixedly connected to the base 1, forming a physical isolation barrier. The light-shielding plate 19 is set facing the first camera 6, which can block ambient stray light from entering the lens field of view of the first camera 6 and reduce background interference in the acquired side image.

[0071] The light-shielding plate 19 can be made of a dark light-absorbing material. For example, the surface of the light-shielding plate 19 can be a matte black coating, which can reduce the reflectivity of ambient light on the other side of the transparent stage 5, making the background of the side image appear uniformly black. Through the spatial arrangement of the light-shielding plate 19 and the transparent stage 5, unnecessary light interference is reduced.

[0072] like Figure 3 As shown in this embodiment, a cooling fan is also installed inside the base 1, and heat dissipation holes are provided on the side wall of the base 1. The cooling fan can cool down the industrial control computer 3. Multiple lockable casters are also installed on the lower surface of the base 1, which improves the ease of movement of the nuclear object pairing device.

[0073] As an optional implementation method of this embodiment, such as Figure 1 and Figure 6 As shown, in this embodiment, the surface of the flexible gripper 4 is integrated with a piezoresistive sensor network 31, and the flexible gripper 4 includes a joint drive assembly 41. The joint drive assembly 41 and the piezoresistive sensor network 31 are electrically connected to the industrial control computer 3.

[0074] The flexible gripper 4 integrates a high-density piezoresistive sensor network 31 on its surface. This network is a distributed sensing system composed of multiple piezoresistive sensors. When an external force is applied to a semiconductor material (e.g., single-crystal silicon), the deformation of the internal lattice causes a change in resistivity, thereby converting the mechanical signal into an electrical signal. For example, the spacing between each sensing node in the piezoresistive sensor network 31 can be no greater than 2 mm, facilitating coverage of the entire gripping surface. The piezoresistive sensor network 31 can monitor the local pressure distribution in real time in contact with the object to be paired, and each sensing node can independently detect the pressure value.

[0075] The joint drive component 41 is the core module for realizing joint movement in the flexible gripper 4. The joint drive component 41 may include a servo motor and supports multi-degree-of-freedom movements such as opening, closing, rotation, and bending. The joint drive component 41 receives instructions from the industrial control computer 3 and dynamically adjusts the gripping posture of the flexible gripper 4 (for example, adjusting the finger bending angle ±0.1°), which can adapt to objects of different sizes and shapes to be paired.

[0076] The industrial control computer 3 has a pre-programmed trajectory, and the flexible gripper 4 approaches the object to be paired at a preset speed. The piezoresistive sensor network 31 provides real-time feedback on the contact pressure. When the pressure at any sensor node reaches a preset pressure threshold, a closing action pause command is triggered.

[0077] The industrial control computer 3 calculates the pressure center based on the initial contact coordinate set. If the offset is greater than the preset offset threshold, it can be determined that the initial pose is abnormal. The initial contact coordinate set refers to the set of coordinates of the initial contact point detected by the piezoresistive sensor network 31 when the flexible gripper 4 first contacts the surface of the object to be paired (e.g., a walnut). The coordinates in the initial contact coordinate set record the specific spatial location where the contact occurs, which can be represented by a three-dimensional coordinate system (e.g., X, Y, Z axes), that is, the set of spatial locations of all sensor nodes that detect pressure values.

[0078] The first center of gravity offset refers to the spatial deviation between the actual center of gravity position of the contact pressure distribution and the preset center of gravity position. The actual center of gravity position can be obtained by weighted calculation based on the pressure values ​​of the piezoresistive sensor network 31. The first center of gravity offset is obtained by comparing the actual center of gravity position with the preset center of gravity position. The preset center of gravity position is an ideal gripping balance point pre-set according to the physical characteristics of the object to be paired (e.g., a walnut for collecting and playing with), used to guide the flexible gripper 4 in gripping the object to be paired.

[0079] An abnormal initial pose can include initial pose tilting and initial pose offset. If the object to be paired is tilted or offset in the placement slot 171, its geometric center does not coincide with the preset grasping center of gravity, resulting in uneven distribution of the first contact point (such as concentrated contact on one side), and the pressure center of gravity will inevitably deviate from the preset position.

[0080] For example, when the object to be paired is a walnut for collecting and playing with, if the walnut is tilted in the placement groove 171, its protruding parts (e.g., the tip or edge) will preferentially contact the sensor node, causing the node pressure value in this area to be significantly higher than that in other areas. After weighted calculation, the pressure center of gravity is biased towards the tilted side. When the first center of gravity offset is greater than the offset threshold, the initial pose abnormality judgment is triggered. When the walnut is not in the position of the preset grasping center of gravity, for example, it is biased towards one side of the groove wall of the placement groove 171. The first contact point is concentrated on the offset side, and the pressure center of gravity deviates from the preset grasping axis. When the first center of gravity offset is greater than the offset threshold, the initial pose abnormality judgment is triggered.

[0081] The industrial control computer 3 stores the correspondence between the first center of gravity offset and the first adjustment signal. The first adjustment signal is a command received by the joint drive component 41, which can adjust the finger opening and closing angle (e.g., unilateral contraction of 0.5mm) or rotation angle (±3°). By calculating the first center of gravity offset, such anomalies can be identified in the early stages of grasping, and the finger posture can be adjusted in real time by the joint drive component 41 (e.g., unilateral contraction of 0.5mm or rotation of 3°) to ensure that the object to be paired enters the subsequent process in a stable and low-damage state.

[0082] When the object to be paired is a walnut for collecting and playing with, the texture features (such as deep grooves and shallow lines) of the walnut can also be identified by the pressure distribution differences corresponding to multiple sensor nodes in the piezoresistive sensor network 31, which complements the image recognition data collected by the camera module.

[0083] Specifically, when the contact pressure is low, or even zero, and the sensor node has no or weak response, the area corresponding to that sensor node can be a deep groove region. The width of the groove can be estimated based on the pressure loss range. When the contact pressure is concentrated and the sensor node outputs a high value, the area corresponding to that sensor node can be a raised texture, such as "ant patterns" or "starry skies." By using the coordinates of the high-pressure sensor node, the principal direction vector of this raised texture can be fitted. For example, the principal direction vector can be radial or mesh-like. The groove width can be calculated based on the pressure loss range, solving the problem of difficulty in quantifying the three-dimensional shape of the image. The fitted principal direction vector of the raised texture reduces the impact of poor image acquisition quality due to lighting or occlusion on texture recognition.

[0084] In this embodiment, a six-dimensional force sensor 32 is installed on the wrist of the flexible gripper 4. The six-dimensional force sensor 32 is used to monitor the global mechanical state of the flexible gripper 4. The six-dimensional force sensor 32 is a sensor that can simultaneously measure three-dimensional spatial forces (Fx, Fy, Fz) and torques (Mx, My, Mz).

[0085] The industrial control computer 3 can calculate the tilt angle of an object based on the torque components and three-dimensional spatial forces. The specific calculation method is existing technology and will not be elaborated here. The torque components, three-dimensional spatial forces, and the current joint angles and displacements are used as inputs to the inverse dynamics model. The inverse dynamics model outputs joint compensation amounts, which can include joint angle or displacement compensation amounts, to counteract the tilting torque.

[0086] By directly mapping six-dimensional force information to joint motion compensation, coupled control of force and motion is achieved. Through a technical chain of six-dimensional force sensors 32, tilt angle threshold determination, inverse dynamics calculation, and closed-loop control, high-precision and low-damage grasping operations are realized.

[0087] As another optional implementation method of this embodiment, such as Figure 1 and Figure 7 As shown, in this embodiment, a fourth camera 42 is installed on the flexible gripper 4, and the fourth camera 42 is electrically connected to the industrial control computer 3; the fourth camera 42 is used to collect image data of the object to be paired after the flexible gripper 4 grasps the object to be paired, and transmit the image data to the industrial control computer 3.

[0088] The flexible gripper 4 includes a joint drive assembly 41, which is electrically connected to the industrial computer 3. The fourth camera 42 can be a dual wide-angle RGB camera mounted on the wrist or other positions of the flexible gripper 4, and its field of view can cover the gripping area. For example, the resolution of the fourth camera 42 is not less than 1080p, and the frame rate is not less than 60fps.

[0089] The industrial control computer 3 can display the images captured by the fourth camera 42 through a display screen or other devices. The user can adjust the joint drive component 41 according to the displayed images, thereby adjusting the position and posture of the flexible gripper 4 and the object to be grasped.

[0090] As another optional implementation method of this embodiment, such as Figure 1 and Figure 7 As shown, in this embodiment, the flexible gripper 4 includes a joint drive assembly 41, and a fourth camera 42 is mounted on the flexible gripper 4. The joint drive assembly 41 and the fourth camera 42 are electrically connected to the industrial control computer 3. The fourth camera 42 is used to acquire image data of the object to be paired after the flexible gripper 4 grasps the object to be paired, and transmit the image data to the industrial control computer 3. Based on the image data, the industrial control computer 3 obtains a third adjustment signal corresponding to the translational deviation and rotational deviation of the object to be paired, so as to trigger the adjustment of the joint drive assembly 41.

[0091] After the flexible gripper 4 grasps the object to be paired, the fourth camera 42 acquires image data of the object. The fourth camera 42 transmits the image data to the industrial control computer 3. The industrial control computer 3 stores a trained YOLOv8s model, HRNet (High-Resolution Network), and PnP algorithm. The YOLOv8s model can identify the bounding box corresponding to the object to be paired, and the bounding box is used to locate the spatial position of the object to be paired in the image. HRNet is a deep convolutional network that maintains high resolution throughout.

[0092] The 2D image data cropped from bounding boxes by the YOLOv8s model can be used as input to HRNet. HRNet can output a heatmap of multiple pre-defined feature points (e.g., apex vertices) for high-precision localization of key parts of the object to be paired. Each pixel value in the heatmap represents the probability of the key part appearing, and the peak position is the coordinate of the key point. For example, apex vertices can identify the orientation of the object to be paired.

[0093] The PnP algorithm matches 2D keypoints (i.e., peak values ​​in the heatmap) with 3D CAD model points, outputting the translational deviations (Δx, Δy, Δz) and rotational deviations (ΔRx, ΔRy, ΔRz) corresponding to the objects to be paired. The 3D CAD model points are the set of coordinates of high-precision 3D feature points of the object to be grasped within its own coordinate system, used to provide a spatial reference for the PnP algorithm. These 3D CAD model points can be preset standard points.

[0094] Based on the translational and rotational deviations, the industrial computer 3 generates a third adjustment signal to drive the joint drive component 41 to make adjustments. For example, when the object to be paired is a walnut, if the current walnut corresponds to ΔRx=-10°, the third adjustment signal includes controlling the flexible gripper 4 to rotate +10° around the X-axis based on the joint drive component 41 to achieve reverse compensation.

[0095] like Figure 8 and Figure 9 As shown, as an optional implementation in this embodiment, a plurality of support rods 53 are inserted into the transparent platform 5. Each support rod 53 extends in a direction perpendicular to the plane of the transparent platform. Each support rod 53 is connected to a lifting drive assembly 33. Each lifting drive assembly 33 is electrically connected to the industrial control computer 3.

[0096] The lifting drive assembly 33 includes a stepper motor, which enables the support rod 53 to perform only vertical reciprocating linear motion. Each support rod 53 is connected to an independent miniature stepper motor, and the industrial control computer 3 controls the lifting commands via a CAN bus. In this embodiment, the field of view of the second camera 7 is located at the center of the transparent stage 5, and multiple support rods 53 can be distributed along the edge of the second camera 7. The support rods 53 can be made of a high-transmittance material. For example, the support rods 53 can be made of quartz glass or reinforced acrylic material with a light transmittance of ≥95%, and can all be set within the first area stage 51.

[0097] A micro-groove can be machined at the top of each support rod 53. A micro-piezoresistive thin-film sensor 34 is embedded in the corresponding micro-groove, exposing the surface of the sensor and allowing it to directly contact the bottom of the object to be paired. The electrodes of the micro-piezoresistive thin-film sensor 34 are led out through metal wires pre-embedded inside the support rod 53, and a transparent insulating layer can be plated on the outside to encapsulate the internal micro-wires. An indium tin oxide (ITO) film can be sputtered onto the surface of the micro-piezoresistive thin-film sensor 34, balancing conductivity and light transmittance to achieve a transparent conductive layer coverage. This ensures accurate acquisition of pressure signals while maintaining high light transmittance and avoids optical interference with the bottom image.

[0098] Once the objects to be paired are placed on the transparent platform 5, the miniature piezoresistive film sensor 34 can directly detect the local pressure value at the bottom of the objects. The industrial control computer 3 receives the discrete local pressure values ​​and pressure coordinates of all support rods 53 and generates an original pressure matrix. Based on the original pressure matrix, the industrial control computer 3 marks the high-pressure area and deep groove, and calculates the second center of gravity offset of the objects to be paired. The calculation process for the second center of gravity offset is the same as that for the first center of gravity offset, and will not be elaborated here.

[0099] When the object to be paired is a collectible walnut, a preset support mode can be obtained based on the current variety of collectible walnuts. The industrial control computer 3 stores the mapping relationship between the support mode, the second center of gravity offset, and the extension and retraction distance of the support rod 53. For example, multiple support rods 53 can be divided into an inner ring and an outer ring, that is, the support rods 53 of the outer ring are located outside the support rods 53 of the inner ring. When the collectible walnut variety is a concave-bottomed official's hat, the support mode is a ring support mode. The support rods 53 of the outer ring are raised to the first target height, and the support rods 53 of the inner ring are lowered to the second target height, which is lower than the first target height.

[0100] As another optional implementation of this embodiment, the lifting drive assembly 33 can be manually controlled by the industrial control computer 3 according to the shape of the object to be paired, so as to adjust the extension distance of each support rod 53.

[0101] The height-adjustable support rods 53 allow multiple support rods 53 to dynamically match the shape of the base 1, enabling the object to be paired to maintain the target posture required for image acquisition, thus improving the accuracy of side image acquisition. The target posture can be the posture in which the actual center of gravity of the object to be paired after being placed on the transparent platform 5 coincides with the preset center of gravity when placed on the transparent platform 5.

[0102] like Figure 10 As shown, based on the same technical concept, this application also provides a pairing method using the above-mentioned nuclear object pairing device, the main process of which is described below (S1-S5):

[0103] S1, based on the gantry drive module, a flexible gripper is driven to grasp the object to be paired to the detection platform, the detection platform including a transparent stage and a drive motor for driving the transparent stage to rotate;

[0104] S2, the industrial control computer controls the third power supply to turn on and controls the third camera to start. The third camera acquires the top image of the object to be paired and transmits the top image to the industrial control computer so that the industrial control computer can identify whether the object to be paired in the top image has reached the target pose.

[0105] S3, when the object to be paired reaches the target pose, control the first power supply and the first camera to turn on, and use the first camera to acquire a side image of the object to be paired in the current position. The first camera transmits the side image of the object in the current position to the industrial control computer. Control the drive motor to start, and use the transparent platform to drive the object to be paired to rotate to the next position. Use the first camera to acquire a side image of the next position until the industrial control computer obtains the side images of the object to be paired in each preset position.

[0106] S4, the industrial control computer controls the third power supply to turn off, controls the second power supply to turn on, and controls the second camera to capture the bottom image of the object to be paired, and the second camera transmits the bottom image to the industrial control computer;

[0107] S5, the industrial control computer obtains the size and outline of the objects to be paired based on the top, side and bottom images of each object.

[0108] As an optional implementation of this embodiment, before the flexible gripper grasps the object to be paired onto the detection platform, the following is also included:

[0109] The flexible gripper approaches the walnut with a pre-programmed trajectory and grasps the object to be paired;

[0110] The fourth camera acquires image data of the object to be paired and transmits the image data to the industrial control computer; the fourth camera is mounted on the flexible gripper, and the flexible gripper includes a joint drive assembly.

[0111] Based on the image data, the industrial control computer obtains a third adjustment signal corresponding to the translational and rotational deviations of the object to be paired, thereby triggering the adjustment of the joint drive component.

[0112] As another optional implementation of this embodiment, before the flexible gripper grasps the object to be paired onto the detection platform, it further includes:

[0113] The flexible gripper approaches the walnut with a pre-programmed trajectory. Based on the piezoresistive sensor network integrated on the surface of the flexible gripper, it scans the contact pressure distribution and transmits it to the industrial control computer. The industrial control computer also measures axial force and torque based on a six-dimensional force sensor and calculates the tilt angle of the object. The six-dimensional force sensor is installed on the wrist of the flexible gripper for global mechanical state monitoring.

[0114] When the pressure of any sensor node in the piezoresistive sensor network reaches a preset pressure threshold, the industrial control computer controls the flexible gripper to pause its closing action and records the first contact coordinate set. Based on the first contact coordinate set, the industrial control computer calculates a first center of gravity offset. If the first center of gravity offset is greater than the offset threshold, the flexible gripper is determined to have an abnormal initial posture, and a first adjustment signal is sent to the joint drive component of the flexible gripper to trigger the posture adjustment of the middle finger of the flexible gripper.

[0115] When the tilt angle of the object is greater than a preset angle threshold, the flexible gripper is determined to have an abnormal gripping posture. The industrial control computer calculates the joint compensation amount through the inverse dynamics model and sends a second adjustment signal to the joint drive component to trigger the adjustment of the joint angle or displacement of the flexible gripper.

[0116] In this embodiment, after the flexible gripper grasps the object to be paired onto the detection platform, the following steps are also included:

[0117] Each support rod has its own corresponding miniature piezoresistive film sensor, which directly detects the local pressure value at the bottom of the object to be paired.

[0118] The industrial control computer receives discrete local pressure values ​​and pressure coordinates from all support rods and generates the original pressure matrix;

[0119] The industrial control computer calculates the second center of gravity offset of the object to be paired based on the original pressure matrix. Based on the second center of gravity offset, the industrial control computer adjusts the extension distance of each support rod through the corresponding lifting drive components of each support rod so that the actual center of gravity of the object to be paired placed on the transparent platform coincides with the preset center of gravity placed on the transparent platform.

[0120] The terms “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0121] 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 application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0122] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0123] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A pairing device for nuclear-like objects, characterized in that, The system includes a base (1), on which a gantry drive module (2), a camera module, a detection platform, and an industrial control computer (3) are connected. The gantry drive module (2) is connected to a gripper connecting plate (21), on which a flexible gripper (4) is installed. The detection platform includes a transparent platform (5) and a drive motor for driving the transparent platform (5) to rotate. The gantry drive module is used to drive the flexible gripper (4) to the position of the object to be paired, and after the flexible gripper (4) has completed the gripping of the object to be paired, it drives the flexible gripper (4) to the position of the transparent platform (5) so that the flexible gripper (4) places the object to be paired on the transparent platform (5). The camera module includes a first camera (6), a second camera (7), and a third camera (8). The first camera (6) is fixedly installed on the upper surface of the base (1) and is correspondingly arranged with the transparent platform (5). The second camera (7) is fixedly installed inside the base (1) and located below the transparent platform (5). The third camera (8) is fixedly installed above the transparent platform (5). The first camera (6) is used to capture the side image of the object to be paired on the transparent platform (5). The second camera (7) is used to capture the bottom image of the object to be paired on the transparent platform (5). The third camera (8) is used to capture the top image of the object to be paired on the transparent platform (5). The camera module is electrically connected to the industrial control computer (3), which is used to pair each object to be paired according to the collected images of each object to be paired.

2. The nuclear object pairing device according to claim 1, characterized in that, The base (1) is connected to the upper shell (15), and the upper shell (15) has a material conveying port (151) on its side. The upper surface of the base (1) is fixedly connected to a material conveying guide rail (16), which extends from the material conveying port (151) toward the side close to the transparent platform (5). A tray (17) is slidably connected to the material conveying guide rail (16), and the tray (17) includes multiple independent placement slots (171), each of which contains an object to be paired.

3. The nuclear object pairing device according to claim 2, characterized in that, The transparent platform (5) includes a first regional platform (51) and a second regional platform (52). The first regional platform (51) is a platform made of transparent material, and the second regional platform (52) is a platform made of frosted material. The arithmetic mean deviation of the profile in the surface roughness parameter of the second regional platform (52) is 0.4 micrometers to 0.8 micrometers.

4. The nuclear object pairing device according to claim 2, characterized in that, A spring clip (18) is provided between the tray (17) and the base (1). The spring clip (18) includes a first clip body and a second clip body. The first clip body is fixedly installed on the base (1), and the second clip body is fixedly installed on the end face of the tray (17). When the tray (17) slides along the material guide rail (16) to the target position, the first clip body engages with the second clip body. A micro switch sensor (35) is installed on the upper surface of the base (1). The micro switch sensor (35) is used to detect whether the tray (17) has reached the target position. The micro switch sensor (35) is electrically connected to the industrial control computer (3).

5. The nuclear object pairing device according to claim 1, characterized in that, A first light source (111) is provided at the first camera (6), a second light source (14) is provided at the second camera (7), and a third light source (131) is provided at the third camera (8).

6. The nuclear object pairing device according to claim 5, characterized in that, The transparent platform (5) has a light shield (19) on the side away from the first camera (6). The light shield (19) is fixedly connected to the base (1) and is arranged facing the first camera (6).

7. The nuclear object pairing device according to claim 1, characterized in that, A fourth camera (42) is installed on the flexible gripper (4), and the fourth camera (42) is electrically connected to the industrial control computer (3). The fourth camera (42) is used to collect image data of the object to be paired after the flexible gripper (4) grabs the object to be paired, and transmit the image data to the industrial control computer (3).

8. The nuclear object pairing device according to claim 1, characterized in that, Multiple support rods (53) are inserted into the transparent platform (5). Each support rod (53) extends in a direction perpendicular to the plane of the transparent platform (5). Each support rod (53) is connected to a lifting drive assembly (33). Each lifting drive assembly (33) is electrically connected to the industrial computer (3).

Images