A cell loading method based on multi-view visual recognition technology

Abstract

The present application relates to a kind of based on multi-view visual identification technology's electric core feeding method, comprising the following steps: step S1: AGV intelligent handling robot according to the feeding instruction of control system places full electric core multi-layer tray in electric core feeding area;Step S2: AGV intelligent handling robot replenishment is completed and sends signal to control system, high-altitude mobile camera group mechanism and 2D camera are photographed to electric core feeding area and image is transmitted to control system;Step S3: control system obtains the remaining electric core quantity of electric core feeding area according to image;Step S4: control system sends instruction to feeding robot and carries out electric core feeding operation, i.e. electric core is moved from electric core feeding area to conveying mechanism, and foam tray is moved to tray placement area;Step S5: conveying mechanism drives electric core to move.The present application can form multi-view visual identification, establish three-dimensional model to the electric core to be grabbed, realize the electric core grabbing of different state, improve the intelligentization, automation and universality of feeding system.

Description

Technical Field

[0001] This invention belongs to the field of battery loading technology, specifically relating to a battery cell loading method based on multi-view vision recognition technology. Background Technology

[0002] With the advancement and implementation of carbon peaking and carbon neutrality strategies, energy storage batteries are ushering in a period of rapid development. Various cell manufacturers and PACK integrators are emerging rapidly. As a core component of battery PACKs, cells need to be loaded, which involves transferring cells stacked in multiple layers on foam trays to cell buffer platforms.

[0003] The following problems exist with incoming battery cells: (1) There are differences in the height of the base support of each battery cell factory; (2) The battery cell stacking methods of each battery cell factory are inconsistent, with alternating stacking and unidirectional stacking coexisting; (3) The orientation of the electrode posts of battery cells from different manufacturers is random; (4) The size and material of the battery cell spacer are different from those of each battery cell factory, and the spacing between adjacent battery cells is different.

[0004] Currently, the common implementation of automated cell loading systems involves an industrial robot paired with a head-up camera. However, this approach suffers from limitations such as a small field of view, inability to detect the distance between adjacent cells, inability to detect cell height, and inability to detect the incoming material deflection angle. Furthermore, it places high demands on the consistency of incoming materials and the precision of loading. How to adapt the robot to materials from different cell suppliers and reduce the requirements for consistency and loading precision has become a pressing issue for many PACK production line manufacturers and battery PACK integrators. Summary of the Invention

[0005] The purpose of this invention is to provide a battery cell loading method based on multi-view vision recognition technology to solve the problem of poor versatility in battery cell loading.

[0006] The battery cell loading method based on multi-view vision recognition technology of the present invention is implemented as follows:

[0007] A method for feeding battery cells based on multi-view vision recognition technology includes the following steps:

[0008] Step S1: The AGV intelligent handling robot places the multi-layered pallet filled with battery cells into the battery cell loading area according to the loading instructions of the control system.

[0009] Step S2: The AGV intelligent handling robot completes the material replenishment and sends a signal to the control system. The high-altitude mobile camera group and 2D camera take pictures of the battery cell loading area and transmit the images to the control system.

[0010] Step S3: The control system obtains the number of remaining cells in the cell feeding area based on the image. If the number of remaining cells in the cell feeding area is lower than the set value, it returns to step S1; otherwise, it executes step S4.

[0011] Step S4: The control system sends instructions to the loading robot to perform the battery cell loading operation, that is, to move the battery cells from the battery cell loading area to the conveying mechanism and move the foam tray to the tray placement area;

[0012] Step S5: The conveyor mechanism drives the battery cell to move, and the OCV testing mechanism is used to test the internal resistance of the battery cell. Unqualified products are removed from the conveyor mechanism by the grippers of the testing station.

[0013] Step S6: Repeat steps S1-S5 above;

[0014] In step S4, the specific steps of the cell loading operation include:

[0015] Step T1: The loading robot moves to the set origin point;

[0016] Step T2: The high-altitude mobile camera group takes pictures of the battery cell loading area;

[0017] Step T3: The loading robot moves, and the 2D camera on its cell clamping fixture takes pictures of the cell loading area;

[0018] Step T4: The loading robot returns to the origin. The control system receives the images obtained by the high-altitude moving camera group and the 2D camera, and determines whether there are any remaining battery cells on the foam tray. If so, proceed to step T5; otherwise, proceed to step T7.

[0019] Step T5: The control system completes multi-view vision based on the images obtained by the high-altitude mobile camera group and the 2D camera, determines the placement of the top layer of battery cells, and establishes a three-dimensional coordinate system for the battery cells to be grasped.

[0020] Step T6: The feeding robot grabs the corresponding battery cell using its battery cell clamping fixture according to the instructions of the control system and transfers it to the conveying mechanism, thus executing step T8;

[0021] Step T7: The control system controls the feeding robot to move the battery cell clamping fixture to the pallet placement area;

[0022] Step T8: The loading robot returns to its origin, and the battery cell fixture is reset;

[0023] Step T9: Repeat steps T2-T8 above until all multi-layer cells in the cell loading area have been loaded.

[0024] Furthermore, the high-altitude mobile camera assembly includes a mounting bracket, and at least three cameras and a high-brightness projector fixed to the bottom of the mounting bracket;

[0025] The mounting bracket is fixed on two parallel mounting rods, and the two ends of the mounting rods are mounted on two parallel support frames located above the cell loading area via linear module I.

[0026] Furthermore, there are two battery cell loading areas, which are arranged side by side below the support frame.

[0027] Furthermore, the placement of the battery cells includes the stacking direction of the cells, the orientation of the terminals, the spacing between the cells, the rotation phase, and the height of the cells.

[0028] Furthermore, the battery cell clamping fixture is mounted on the robotic arm of the loading robot;

[0029] The battery cell clamping fixture includes a mounting base and multiple battery cell grippers arranged side by side at the bottom of the mounting base with variable spacing, as well as foam grippers located around the mounting base and movable up and down.

[0030] A vertically arranged baffle cylinder is installed on one side of the mounting base, and a baffle plate is installed at the lower end of the piston rod of the baffle cylinder.

[0031] Furthermore, the bottom of the mounting base is provided with a ball screw that is in the same direction as the arrangement of the battery cell grippers. One end of the ball screw is connected to a motor. A diamond-shaped take-up and release frame is provided above the battery cell grippers. The nut of the ball screw is connected to the diamond-shaped take-up and release frame. By tightening or unfolding the diamond-shaped take-up and release frame, the individual battery cell grippers are driven to retract or disperse.

[0032] The inner wall of the cell clamp is covered with a sponge strip.

[0033] Furthermore, a vertically arranged lifting cylinder is mounted on the outer side of the mounting base, and the foam gripper is fixed to the end of the piston rod of the lifting cylinder.

[0034] Furthermore, the 2D camera is mounted on one side of the mounting base parallel to the direction of the cell clamp arrangement, and a supplementary light is installed below the 2D camera to cooperate with it.

[0035] Furthermore, the conveying mechanism includes a tracked roller conveyor and a grooved box mounted on the tracked roller conveyor.

[0036] At least two battery cells can be placed side by side in each recessed box.

[0037] Furthermore, an OCV testing mechanism is installed on the conveying mechanism;

[0038] The OCV testing mechanism includes a mounting frame spanning the conveying mechanism and a lifting seat fixed on the mounting frame and movable up and down. At least two sets of test probes are mounted on one side of the lifting seat.

[0039] After adopting the above technical solution, the beneficial effects of the present invention are as follows:

[0040] (1) This invention enables automatic replenishment of the battery cell loading area by linking the AGV intelligent handling robot with the control system, thereby improving the automation level and efficiency of battery cell loading.

[0041] (2) The present invention uses the combination of a high-altitude mobile camera group and a 2D camera to form a multi-view vision recognition structure, establish a three-dimensional model of the battery cell to be grasped, realize the grasping of battery cells in different directions, spacing, rotation phase and height, and improve the intelligence, automation and versatility of the feeding system. Attached Figure Description

[0042] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0043] Figure 1 This is a structural diagram of a battery cell loading system based on multi-view vision recognition technology according to a preferred embodiment of the present invention;

[0044] Figure 2 yes Figure 1 Enlarged view of section A;

[0045] Figure 3 This is a structural diagram of the high-altitude moving camera group mechanism of the battery cell loading system based on multi-view vision recognition technology according to a preferred embodiment of the present invention;

[0046] Figure 4 This is a structural diagram of the battery cell clamping fixture of the battery cell feeding system based on multi-view vision recognition technology according to a preferred embodiment of the present invention;

[0047] Figure 5 This is a structural diagram of the battery cell clamping fixture of the battery cell feeding system based on multi-view vision recognition technology according to a preferred embodiment of the present invention;

[0048] Figure 6 This is a structural diagram of the battery cell clamping fixture (without battery cell grippers) of the battery cell feeding system based on multi-view vision recognition technology according to a preferred embodiment of the present invention;

[0049] Figure 7 This is a flowchart of a preferred embodiment of the battery cell loading method based on multi-view vision recognition technology of the present invention;

[0050] Figure 8 This is a flowchart illustrating the specific process of battery cell loading based on multi-view vision recognition technology according to a preferred embodiment of the present invention.

[0051] In the diagram: 1. Battery cell loading area; 2. High-altitude mobile camera assembly; 2-1. Mounting bracket; 2-2. Camera; 2-3. High-brightness projector; 2-4. Mounting rod; 2-5. Linear module I; 2-6. Support frame; 3. Loading robot; 4. Battery cell clamping fixture; 4-1. 2D camera; 4-2. Mounting base; 4-3. Battery cell gripper; 4-4. Foam gripper; 4-5. Material blocking cylinder; 4-6. Material blocking plate; 4-7. Ball screw; 4-8. Motor; 4-9. Nut; 4-10. Connecting plate; 4-11. Diamond-shaped take-up and drop-off rack; 4. Base plate. 4-12, Bearing housing 4-13, Protruding column 4-14, Lifting cylinder 4-15, Supplementary light 4-16, Conveying mechanism 5, Tracked roller line 5-1, Groove box 5-2, Pallet unloading area 6, OCV testing mechanism 7, Mounting frame 7-1, Lifting seat 7-2, Test probe 7-3, Linear module II 7-4, Probe bracket 7-5, Column 7-6, Mounting plate 7-7, Scanner 7-8, Clamping block 7-9, Round rod 7-10, Guardrail 8, Window I 8-1, Window II 8-2. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0054] like Figure 1-8 As shown, a cell loading method based on multi-view vision recognition technology includes the following steps:

[0055] Step S1: The AGV intelligent handling robot places the multi-layered pallet filled with battery cells into the battery cell loading area 1 according to the loading instructions of the control system;

[0056] Among them, the AGV intelligent handling robot adopts laser SLAM navigation technology, integrates visual detection, and has the functions of pallet position recognition, vehicle posture correction, real-time route planning, and can automatically avoid and alarm pedestrians and obstacles in the route.

[0057] Furthermore, the AGV intelligent handling robot has a narrow aisle forklift-style structure, with its forks capable of lifting, and an overall lifting height of 1.8m, enabling it to handle multiple layers of materials.

[0058] Step S2: The AGV intelligent handling robot completes the material replenishment and sends a signal to the control system. The high-altitude mobile camera group 2 and the 2D camera 4-1 take pictures of the battery cell loading area 1 and transmit the images to the control system.

[0059] Step S3: The control system obtains the number of remaining cells in cell loading area 1 based on the image. If the number of remaining cells in cell loading area 1 is lower than the set value, it returns to step S1; otherwise, it executes step S4.

[0060] The AGV intelligent handling robot communicates with the control system to achieve intelligent linkage with equipment such as the loading robot 3 and the high-altitude mobile camera group 2. When the multi-view vision recognition structure formed by the high-altitude mobile camera group 2 and the 2D camera 4-1 detects that the number of battery cells is lower than the set value, the control system will send a signal to the AGV intelligent handling robot. The AGV intelligent handling robot will travel along the prescribed route to the warehouse shelf to forklift the battery cells and transfer them to the battery cell loading area 1.

[0061] Step S4: The control system sends instructions to the loading robot 3 to perform the battery cell loading operation, that is, to transfer the battery cells from the battery cell loading area 1 to the conveying mechanism 5, and to transfer the foam pallet to the pallet placement area 6.

[0062] The loading robot 3 picks up the battery cells one by one or in groups and transfers them to the conveying mechanism 5 for further processing. After the battery cells on each layer of foam trays are loaded, the loading robot 3 picks up the foam trays and transfers them to the tray placement area 6 to expose the lower layer of battery cells for subsequent battery cell loading operations.

[0063] Step S5: The conveying mechanism 5 drives the battery cell to move, and the OCV testing mechanism 7 is used to test the internal resistance of the battery cell. Unqualified products are removed from the conveying mechanism 5 by the grippers of the testing station.

[0064] The OCV testing unit 7 can scan each battery cell individually and perform internal resistance testing. If there are unqualified products in the internal resistance test, the grippers at the test station will remove the unqualified products from the conveying mechanism 5.

[0065] Step S6: Repeat steps S1-S5 above;

[0066] In step S4, the specific steps of the cell loading operation include:

[0067] Step T1: The loading robot 3 moves to the set origin.

[0068] Step T2: The high-altitude mobile camera group 2 takes pictures of the battery cell loading area 1.

[0069] The high-altitude mobile camera assembly 2 includes a mounting bracket 2-1, and at least three cameras 2-2 and a high-brightness projector 2-3 fixed to the bottom of the mounting bracket 2-1.

[0070] In this embodiment, the mounting bracket 2-1 is a rectangular frame structure, and four cameras 2-2 are set, located at the four corners of the mounting bracket 2-1 respectively. The high-brightness projector 2-3 is located at the bottom center of the mounting bracket 2-1. It adopts a blue light projector to provide a light source for the features within the field of view of the camera 2-2, avoiding the influence of strong ambient light on the detection of the camera 2-2.

[0071] Preferably, camera 2-2 can be a 3D camera.

[0072] In order to achieve the lateral movement of the high-altitude mobile camera group mechanism 2, the mounting bracket 2-1 is fixed on two parallel mounting rods 2-4. The two ends of the mounting rods 2-4 are mounted on two parallel support frames 2-6 located above the battery cell loading area 1 via linear modules I 2-5.

[0073] Specifically, linear modules I2-5 are installed on the top rods of the two support frames 2-6. The two ends of the mounting rod 2-4 are assembled on the linear modules I2-5. The linear modules I2-5 can drive the mounting rod 2-4 and the high-altitude mobile camera group mechanism 2 to move laterally above the support frame 2-6, so as to cooperate with the cell loading area 1 below the support frame 2-6 to complete the multi-view vision recognition operation.

[0074] In this embodiment, there are two battery cell loading areas 1, arranged side by side below the support frame 2-6. The AGV intelligent handling robot and the loading robot 3 can work alternately in the two battery cell loading areas 1. That is, when one battery cell loading area 1 is replenished by the AGV intelligent handling robot, the loading robot 3 can move and load the battery cells in the other battery cell loading area 1, thereby further improving the battery cell loading efficiency.

[0075] Step T3: The loading robot 3 moves, and the 2D camera 4-1 on its cell clamping fixture 4 takes pictures of the cell loading area 1.

[0076] The loading robot 3 is a six-axis industrial robot. The battery cell clamping fixture 4 is installed on the robotic arm of the loading robot 3, that is, at the flange position of its sixth axis. Based on the incoming phase (angle) signal fed back by the control system, the loading robot 3 adaptively adjusts the angle of the battery cell clamping fixture 4, which can clamp battery cells with any deflection angle.

[0077] Specifically, the flange position of the sixth axis of the loading robot 3 is a replaceable structure, which can be adapted to different operation requirements by installing corresponding fixtures and tooling.

[0078] The 2D camera 4-1 on the battery cell fixture 4 can work with the high-altitude mobile camera group 2 to achieve multi-view visual recognition of the battery cell loading area 1.

[0079] Step T4: The loading robot 3 returns to the origin. The control system receives the images obtained by the high-altitude moving camera group 2 and the 2D camera 2, and determines whether there are any remaining battery cells on the foam tray. If so, proceed to step T5; otherwise, proceed to step T7.

[0080] Step T5: The control system completes multi-view vision based on the images obtained by the high-altitude mobile camera group 2 and 2D camera 4-1, determines the placement of the top layer of battery cells, and establishes a three-dimensional coordinate system for the battery cells to be grasped.

[0081] The principle of multi-view vision recognition technology is achieved through the cooperation of camera 2-2 in the high-altitude mobile camera group mechanism and 2D camera 4-1 on the battery cell fixture. The high-altitude mobile camera group mechanism uses multiple cameras 2-2 arranged in groups, each responsible for different functions. Different cameras 2-2 are combined and applied to capture images of the left and right sides of the same position from different perspectives through two identical cameras 2-2. At the same time, images are captured by 2D camera 4-1, thereby obtaining multiple pairs of images of the same object from different angles. Then, the depth information of the object is obtained by using the principle of triangulation, and a three-dimensional coordinate system of the battery cell is constructed in all directions. This enables the battery cell fixture 4 to have a blind zone self-adjustment function and is compatible with battery cells of different layers and heights.

[0082] The placement of battery cells includes the stacking direction, terminal orientation, cell spacing, rotation phase, and cell height.

[0083] Specifically, regarding the identification of positive and negative terminals of battery cells: A battery cell contains two terminals. To meet the needs of backend PACK integration testing, the loading robot 3 must ensure that the positive terminals of the cells are stacked in a uniform direction on the assembly line. Currently, there are two solutions in the industry: one relies on the cell manufacturer to ensure this during palletizing, but different manufacturers' methods are difficult to standardize; the other involves the PACK integrator manually flipping the packages, which wastes manpower. With multi-view vision, cell position detection is performed by a high-altitude moving camera group 2. The 2D camera 4-1 mounted on the robotic arm of the loading robot 3 can be used independently for positive and negative terminal identification, guiding the robotic arm's posture adjustment to ensure consistent cell loading direction.

[0084] Regarding cell spacing detection and alarm: The cells are picked up and fed by the cell gripper 4-3 on the robotic arm. The gripper has a certain thickness, and a certain spacing must be maintained between adjacent cells during gripping to prevent damage when the gripper 4-3 descends. Currently, this mainly relies on the foam partitions in the cell packaging, without spacing detection and alarm functions. Due to the relatively large height of the cells, when placed among the partitions, some tilting is inevitable, resulting in a smaller relative spacing. With the assistance of a projector, camera 2-2 can identify and model the three-dimensional coordinates of each cell, measuring the cell spacing. The control system compares this to a safety threshold. When the spacing is less than the safety threshold, the loading robot 3 is triggered to stop operation and issue an alarm, preventing cell damage.

[0085] Cell position height detection: The cell position height determines the downward displacement of the cell clamping fixture 4. Currently, the industry standard is laser ranging, where a laser rangefinder is deployed on the robotic arm to detect the height of each layer of cells, guiding the robotic arm to adapt to the gripping action. However, this method requires multiple checks when the incoming pallet is partially deformed, impacting the loading cycle time. The high-altitude moving camera assembly 2 can construct a three-dimensional coordinate system for the cells from all directions. Furthermore, because it moves in parallel with the robotic arm, detection can be performed when the robotic arm turns to load cells, without affecting the loading cycle time, thus improving loading efficiency.

[0086] Cell rotation phase detection: The 2D camera 4-1 lacks rotation phase detection functionality. In this case, the program is generally set for the robotic arm to place the incoming material in a positive, unbiased position. If the material loading angle deviates, the robotic arm will continue to move downwards to grasp the cell, which greatly increases the probability of the gripper colliding with the cell and causing damage. Based on multi-view vision with no blind spots, by registering the subtle features of the target object, the problem of distinguishing the reverse side of the surface and the rotation phase can be solved.

[0087] Step T6: The feeding robot 3 grabs the corresponding battery cell through its battery cell clamping fixture 4 according to the instructions of the control system and transfers it to the conveying mechanism 5, thus executing step T8.

[0088] Step T7: The control system controls the feeding robot 3 to move the battery cell clamping fixture 4 to the foam tray for placement.

[0089] The battery cell clamping fixture 4 also includes a mounting base 4-2 and multiple battery cell grippers 4-3 arranged side by side at the bottom of the mounting base 4-2 with variable spacing, as well as foam grippers 4-4 located around the mounting base 4-2 and movable up and down.

[0090] The variable distance setting of the cell gripper 4-3 can be used to grip cells with different spacing and facilitates the placement of the cells on the conveyor mechanism 5.

[0091] The foam gripper 4-4 can be raised and lowered to grip the foam tray when it needs to be moved, and raised when gripping the battery cell to avoid affecting the gripping of the battery cell.

[0092] Preferably, both the battery cell gripper 4-3 and the foam gripper 4-4 are cylinder grippers, which can realize automatic gripping of the battery cell and the foam tray.

[0093] To prevent the foam tray from moving along with the battery cell when grabbing the battery cell, a vertically arranged baffle cylinder 4-5 is installed on one side of the mounting base 4-2, and a baffle plate 4-6 is installed at the lower end of the piston rod of the baffle cylinder 4-5.

[0094] To achieve variable pitch of the cell grippers 4-3, a ball screw 4-7 is provided at the bottom of the mounting base 4-2, which is in the same direction as the arrangement of the cell grippers 4-3. One end of the ball screw 4-7 is connected to a motor 4-8. A diamond-shaped take-up and release frame 4-11 is provided above the cell grippers 4-3. The nut 4-9 of the ball screw 4-7 is connected to the diamond-shaped take-up and release frame 4-11. By tightening or unfolding the diamond-shaped take-up and release frame 4-11, the individual cell grippers 4-3 are driven to retract or disperse.

[0095] Specifically, the diamond-shaped take-up and take-down rack 4-11 is formed by multiple pairs of strips connected in a cross manner. The cell clamping claw 4-3 away from the motor 4-8 is fixed on the base plate 4-12 at the bottom of the mounting base 4-2, and the tail intersection of the diamond-shaped take-up and take-down rack 4-11 is fixed above the base plate 4-12. The cell clamping claw 4-3 close to the motor 4-8 is fixed on the connecting plate 4-10, and the connecting plate 4-10 is connected to the front intersection of the diamond-shaped take-up and take-down rack 4-11. The remaining cell clamping claws 4-3 are fixed in sequence at the various intersections in the middle of the diamond-shaped take-up and take-down rack 4-11.

[0096] The tail of the ball screw 4-7 is fixed to the bottom of the mounting base 4-2 by the bearing seat 4-13. The motor 4-8 rotates, driving the ball screw 4-7 to rotate, which in turn drives the nut 4-9 on the ball screw 4-7 to move linearly in its axial direction. Through the movement of the nut 4-9, the connecting plate 4-10 can then drive the diamond-shaped take-up and take-down frame 4-11 to retract inward or expand outward, so as to realize the change of the distance between the individual cell clamps 4-3.

[0097] Preferably, a protruding post 4-14 is provided at the intersection of the diamond-shaped take-up and take-down rack 4-11 to facilitate its connection with the corresponding connecting plate 4-10, base plate 4-12 or cell clamp 4-3.

[0098] The inner wall of the cell gripper 4-3 is covered with a sponge strip to prevent wear on the surface of the cell when gripping it.

[0099] In order to achieve the lifting and lowering of the foam gripper 4-4, a vertically arranged lifting cylinder 4-15 is installed on the outer side of the mounting base 4-2, and the foam gripper 4-4 is fixed to the end of the piston rod of the lifting cylinder 4-15.

[0100] Specifically, the lifting cylinders 4-15 are located on the left and right sides of the mounting base 4-2, that is, on the sides parallel to the arrangement direction of the cell clamps 4-3. The stability of the movement of the foam tray can be ensured by the synchronous clamping of multiple foam clamps 4-4.

[0101] In order to cooperate with the high-altitude mobile camera assembly 2, the 2D camera 4-1 is installed on the side of the mounting base 4-2 parallel to the arrangement direction of the battery cell clamps 4-3, and a supplementary light 4-16 is installed below the 2D camera 4-1 to cooperate with it.

[0102] The 2D camera 4-1 is suspended in the air without obstruction, and the fill light 4-16 ensures the clarity of the images captured by the 2D camera 4-1.

[0103] Specifically, the conveying mechanism 5 includes a tracked roller conveyor 5-1 and grooved boxes 5-2 mounted on the tracked roller conveyor 5-1. The bottom of the grooved boxes 5-2 is fixed to a stepping belt, and the spacing between each grooved box is equal. The belt steps the same distance each time, allowing for the testing of the battery cells inside each grooved box 5-2. Furthermore, the grooved boxes 5-2, fixed to the belt of the tracked roller conveyor 5-1, can achieve periodic cyclical movement with the belt. Compared to the traditional double-layer pallet conveying method, the grooved box combined with belt conveyor mechanism 5 features high conveying efficiency, compact structure, low investment, and low power consumption.

[0104] At least two battery cells can be placed side by side in each groove box 5-2, allowing for simultaneous testing and loading of two battery cells.

[0105] In addition, the conveying mechanism 5 is equipped with an OCV testing mechanism 7, which is used to test the internal resistance of the battery cell.

[0106] The OCV testing mechanism 7 includes a mounting frame 7-1 spanning the conveying mechanism 5, and a lifting seat 7-2 fixed on the mounting frame 7-1 and movable up and down. At least two sets of test probes 7-3 are installed on one side of the lifting seat 7-2.

[0107] Specifically, the mounting frame 7-1 is equipped with a vertically arranged linear module II 7-4, which drives the lifting seat 7-2 to move up and down. The probe brackets 7-5 of the two sets of test probes 7-3 are installed at the bottom of the lifting seat 7-2, and a column 7-6 is installed above the lifting seat 7-2. The top of the column 7-6 is equipped with a mounting plate 7-7 extending away from the mounting frame 7-1. The bottom of the mounting plate 7-7 is equipped with a vertically adjustable scanner 7-8 for scanning QR codes on telecommunications equipment.

[0108] Specifically, the end of the mounting plate 7-7 is provided with a clamping block 7-9. The round rod 7-10 on the scanner 7-8 passes through the clamping block 7-9 and is locked with a bolt. When the bolt is loosened, the round rod 7-10 can move up and down on the mounting plate 7-7. When the bolt is locked, the position of the round rod 7-10 can be fixed.

[0109] Preferably, sensor assemblies are respectively provided around the cell loading area 1 and the tray placement area 6. The sensor assemblies include a pole and multiple laser sensors 9 installed vertically on the pole. The laser sensors 9 located around the cell loading area can detect whether there are any remaining cells on each tray during the loading process, thereby cooperating with the multi-view vision recognition structure to detect the remaining status of cells in the cell loading area 1.

[0110] The laser sensor 9 located on the periphery of the pallet feeding area 6 can be used to detect the number of layers of the pallet and can also assist in multi-view vision recognition to determine whether all layers of cells in the cell feeding area 1 have been fed.

[0111] Step T8: The loading robot 3 returns to the origin, and the battery cell fixture 4 is reset.

[0112] Among them, the cell clamping fixture 4 resets either the cell gripper 4-3 or the foam gripper 4-4, which facilitates subsequent cell gripping.

[0113] Step T9: Repeat steps T2-T8 above until all multi-layer cells in cell loading area 1 have been loaded.

[0114] In this embodiment, the loading robot 3 of the entire battery cell loading system is located in the center, with two battery cell loading areas 1 arranged side by side on its left side, a pallet placement area 6 arranged on its rear side, and a conveying mechanism 5 on its right side, so that the loading robot 3 can cooperate with the battery cell loading area 1, the pallet placement area 6 and the conveying mechanism 5.

[0115] In addition, a protective railing 8 is set around the battery cell feeding system, and the conveying mechanism 5 extends from the right side of the protective railing 8 to facilitate the transfer of the battery cell to the next processing position.

[0116] The left side of the protective railing 8 corresponding to the cell loading area 1 has a window Ⅰ8-1 for AGV intelligent handling robots to load the cells, and the rear side of the protective railing 8 corresponding to the pallet placement area 6 also has a window Ⅱ8-2 to facilitate the removal of foam pallets.

[0117] This invention utilizes multi-view vision recognition technology in conjunction with AGV intelligent handling robots and loading robots to achieve intelligent delivery, transportation, and loading of battery cells. It effectively solves the problems of inconsistent packaging size, stacking method, and terminal orientation of battery cells from different manufacturers. At the same time, it reduces the accuracy requirement for the loading position of the battery cell trolley to below ±80mm. It also has the functions of measuring battery cell spacing, rotation phase, and battery cell height, realizing intelligent logistics and a high degree of automation in production. It solves the problems of low gripping efficiency, poor positioning accuracy, and high requirements for the consistency of incoming materials in existing equipment, greatly improving production efficiency and the degree of equipment automation.

[0118] Based on the above-described preferred embodiments of the present invention, and through the foregoing description, those skilled in the art can make various changes and modifications without departing from the inventive concept. The technical scope of this invention is not limited to the contents of the specification, but must be determined according to the scope of the claims.

Claims

1. A cell loading method based on multi-view visual recognition technology, characterized in that, Includes the following steps: Step S1: The AGV intelligent handling robot places the multi-layered pallet filled with battery cells into the battery cell loading area according to the loading instructions of the control system (1); Step S2: The AGV intelligent handling robot completes the material replenishment and sends a signal to the control system. The high-altitude mobile camera group (2) and the 2D camera (4-1) take pictures of the battery cell loading area (1) and transmit the images to the control system. Step S3: The control system obtains the number of remaining cells in the cell feeding area (1) based on the image. If the number of remaining cells in the cell feeding area (1) is lower than the set value, it returns to step S1; otherwise, it executes step S4. Step S4: The control system sends instructions to the loading robot (3) to perform the battery cell loading operation, that is, to transfer the battery cells from the battery cell loading area (1) to the conveying mechanism (5) and to transfer the foam tray to the tray placement area (6); Step S5: The conveying mechanism (5) drives the battery cell to move, and the OCV testing mechanism (7) is used to test the internal resistance of the battery cell. Unqualified products are removed from the conveying mechanism (5) by the gripper of the testing station. Step S6: Repeat steps S1-S5 above; In step S4, the specific steps of the cell loading operation include: Step T1: The loading robot (3) moves to the set origin; Step T2: The high-altitude mobile camera group (2) takes pictures of the cell loading area (1); Step T3: The loading robot (3) moves, and the 2D camera (4-1) on its cell clamping fixture (4) takes pictures of the cell loading area (1); Step T4: The loading robot (3) returns to the origin. The control system receives the images obtained by the high-altitude moving camera group (2) and the 2D camera (4-1) and determines whether there are any remaining battery cells on the foam tray. If so, proceed to step T5; otherwise, proceed to step T7. Step T5: The control system completes multi-view vision based on the images obtained by the high-altitude mobile camera group mechanism (2) and the 2D camera (4-1), determines the placement state of the top layer of battery cells, and establishes a three-dimensional coordinate system for the battery cells to be grasped. Step T6: The loading robot (3) grabs the corresponding battery cell through its battery cell clamping fixture (4) according to the instructions of the control system and transfers it to the conveying mechanism (5), and executes step T8; Step T7: The control system controls the loading robot (3) to drive the battery cell clamping fixture (4) to move the foam tray to the tray placement area; Step T8: The loading robot (3) returns to the origin, and the battery cell fixture (4) is reset; Step T9: Repeat steps T2-T8 above until all multi-layer cells in the cell loading area (1) are loaded.

2. The cell loading method based on multi-view vision recognition technology according to claim 1, characterized in that, The high-altitude mobile camera group mechanism (2) includes a mounting bracket (2-1), and at least three cameras (2-2) and a high-brightness projector (2-3) fixed to the bottom of the mounting bracket (2-1); The mounting bracket (2-1) is fixed on two parallel mounting rods (2-4). The two ends of the mounting rods (2-4) are mounted on two parallel support frames (2-6) located above the cell loading area (1) via a linear module I (2-5).

3. The cell loading method based on multi-view vision recognition technology according to claim 2, characterized in that, There are two battery cell loading areas (1), which are arranged side by side below the support frame (2-6).

4. The cell loading method based on multi-view vision recognition technology according to claim 1, characterized in that, The placement of the battery cells includes the stacking direction, electrode orientation, cell spacing, rotation phase, and cell height.

5. The cell loading method based on multi-view vision recognition technology according to claim 1, characterized in that, The battery cell clamping fixture (4) is mounted on the robotic arm of the loading robot (3); The battery cell clamping fixture (4) includes a mounting base (4-2) and multiple battery cell grippers (4-3) arranged side by side at the bottom of the mounting base (4-2) with variable distance, as well as foam grippers (4-4) located around the mounting base (4-2) and movable up and down; A vertically arranged baffle cylinder (4-5) is installed on one side of the mounting base (4-2), and a baffle plate (4-6) is installed at the lower end of the piston rod of the baffle cylinder (4-5).

6. The cell loading method based on multi-view vision recognition technology according to claim 5, characterized in that, The bottom of the mounting base (4-2) is provided with a ball screw (4-7) that is in the same direction as the arrangement of the battery cell grippers (4-3). One end of the ball screw (4-7) is connected to a motor (4-8). A diamond-shaped take-up and release frame (4-11) is provided above the battery cell grippers (4-3). The nut (4-9) of the ball screw (4-7) is connected to the diamond-shaped take-up and release frame (4-11). The tightening or unfolding of the diamond-shaped take-up and release frame (4-11) drives each battery cell gripper (4-3) to retract or unfold. The inner wall of the cell clamp (4-3) is covered with a sponge strip.

7. The cell loading method based on multi-view vision recognition technology according to claim 5, characterized in that, A vertically arranged lifting cylinder (4-15) is mounted on the outer side of the mounting base (4-2), and the foam gripper (4-4) is fixed to the end of the piston rod of the lifting cylinder (4-15).

8. The cell loading method based on multi-view vision recognition technology according to claim 5, characterized in that, The 2D camera (4-1) is installed on the side of the mounting base (4-2) parallel to the arrangement direction of the cell clamps (4-3), and a supplementary light (4-16) is installed below the 2D camera (4-1) to cooperate with it.

9. The cell loading method based on multi-view vision recognition technology according to claim 1, characterized in that, The conveying mechanism (5) includes a tracked roller line (5-1) and a grooved box (5-2) mounted on the tracked roller line (5-1); At least two battery cells can be placed side by side in each recessed box (5-2).

10. The cell loading method based on multi-view vision recognition technology according to claim 1, characterized in that, The conveying mechanism (5) is equipped with an OCV testing mechanism (7); The OCV testing mechanism (7) includes a mounting frame (7-1) spanning the conveying mechanism (5) and a lifting seat (7-2) fixed on the mounting frame (7-1) and movable up and down. At least two sets of test probes (7-3) are mounted on one side of the lifting seat (7-2).

Images