Vision-based garment piece feeding mechanism
Through the vision-guided garment piece feeding mechanism and the use of non-contact garment piece anchoring and flattening modules, the automatic, precise positioning and smooth spreading of garment pieces are achieved, which solves the efficiency and accuracy problems of garment piece feeding in garment manufacturing and improves production efficiency and product quality.
Patent Information
- Application Number
- CN202421834608.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-31
- Publication Date
- 2025-09-09
- Estimated Expiration
- 2034-07-31
AI Technical Summary
In existing clothing manufacturing, the loading of garment pieces relies on manual operation, resulting in low production efficiency, large errors, high costs, and inaccurate positioning of robotic arms.
A vision-based garment feeding mechanism is adopted, which includes a non-contact garment anchoring module and a flattening module, combined with visual acquisition and a robotic arm feeding module to achieve automated, precise positioning and smooth spreading of garment pieces.
It improves loading efficiency and accuracy, reduces damage to garment pieces, reduces labor costs, and improves the automation level of the production line and product quality.
Smart Images
Figure CN223316031U_ABST
Abstract
Description
Technical Field
[0001] The utility model relates to a clothing machine, in particular to a garment piece feeding mechanism based on vision. Background Art
[0002] In the garment manufacturing industry, loading plays a crucial role in the efficiency and quality of the entire production process. Especially in automated assembly lines, efficient and precise handling and transfer of garment pieces is crucial for ensuring production continuity and consistency. However, current loading operations largely rely on manual labor, which not only limits production efficiency but also increases operational errors and production costs due to the human factor.
[0003] The traditional method for accurate material loading involves marking the outline of the fabric piece on the machine's workbench. Workers then visually determine whether the piece aligns with the markings to ensure accurate positioning. This method has significant drawbacks: first, it requires high worker vision and experience; second, manual operation is difficult to ensure repeatability and consistency, which directly impacts final product quality and production efficiency.
[0004] While robotic arm technology has made significant progress in automation, enabling precise loading along pre-set paths, ensuring the robotic arm accurately positions itself at the same location each time it grabs a garment piece remains a key challenge in achieving efficient and accurate robotic loading. This involves not only the design and programming of the robotic arm but also innovations in garment positioning technology. Utility Model Content
[0005] In order to solve the above technical problems, the utility model proposes a vision-based garment feeding mechanism to improve the feeding efficiency and accuracy in the garment production process, reduce the dependence on manual operation, and thus improve the stability and reliability of the overall production process. The technical solution adopted by the utility model is:
[0006] A garment feeding mechanism based on vision includes a machine platform, a vision acquisition module and a manipulator feeding module are configured on the machine platform. Different from the existing technology, It also includes a non-contact garment piece anchoring module and a non-contact garment piece flattening module arranged on the machine; the non-contact garment piece anchoring module includes a table fixed on the machine, a negative pressure box is fixed under the table, the upper surface of the negative pressure box is flush with the table and densely covered with air holes, and the negative pressure box is connected to the negative pressure air circuit; the non-contact garment piece flattening module includes two first guide rails fixed on the machine and arranged in the left and right directions, and a slider sliding with the two first guide rails is fixed to the bottom plate; a synchronous belt is also rotated on the machine, and the synchronous belt is driven by a motor; the bottom plate is fixed to the upper or lower layer of the synchronous belt; a first cylinder with an up and down stroke is fixed on the bottom plate, and the piston rod of the first cylinder is fixed to the connecting plate, and the two ends of the connecting plate are respectively fixed to the vertical plates. The gap between the two vertical plates passes through the table and is fixed to the flattening tube. One end of the flattening tube is closed and the other end is connected to the high-pressure air circuit through a quick connector; a row of air jet holes facing the table are opened on the wall of the flattening tube.
[0007] Furthermore, axially vertical guide sleeves are fixed on the bottom plates at the front and rear sides of the first cylinder respectively, and the guide sleeves are slidably matched with the guide rods, and the upper ends of the guide rods are fixedly connected to the connecting plates.
[0008] Furthermore, the robot loading module includes a first linear slide fixed on the machine platform with a left-right stroke, a second linear slide fixed on the slider of the first linear slide with a forward-backward stroke, a fixed vertical plate fixed on the slider of the second linear slide, a second cylinder fixed on the fixed vertical plate with a up-and-down stroke, second guide rails fixed on the fixed vertical plates on the front and rear sides of the second cylinder respectively, a second slider slidingly matched with the second guide rail is fixedly connected to a mounting frame, a servo motor that can accurately rotate to a set angle according to a control signal is fixed on the mounting frame, the motor shaft of the servo motor is fixedly connected to a mounting plate parallel to the table surface, and four pneumatic clamps are installed on the mounting plate.
[0009] Furthermore, the pneumatic clamp includes a clamp cylinder, a fixed jaw is fixed at the end of the clamp cylinder, and the fixed jaw is the static part of the pneumatic clamp. A movable jaw is fixed on the piston rod of the clamp cylinder, and the movable jaw is the movable part of the pneumatic clamp. The movable jaw and the fixed jaw cooperate with each other to form a clamping pair.
[0010] Furthermore, two straight slots with their lengths in the front-to-back direction are respectively opened at the front and rear ends of the right side of the mounting plate. Two screws are passed through the two straight slots and then screwed to the clamping cylinder, so that the clamping positions of the two pneumatic clamps in the front-to-back direction can be adjusted; a small circular through hole and an arc slot with the small circular through hole as the center and a central angle greater than or equal to 90° are respectively opened at the front and rear ends of the left side of the mounting plate. Two screws are passed through the small circular through hole and the arc slot and then screwed to the clamping cylinder, so that the clamping angles of the other two pneumatic clamps in the front-to-back direction can be adjusted.
[0011] Furthermore, a first oil pressure buffer is fixed on the fixed vertical plate on the front side of the second cylinder, and the piston rod of the first oil pressure buffer can abut against the mounting frame; a second oil pressure buffer is fixed on the mounting frame on the rear side of the second cylinder, and the piston rod of the second oil pressure buffer passes through the mounting frame and is fixed to the fixed vertical plate.
[0012] Furthermore, it also includes a control module, which can receive and process the image of the garment piece on the table captured by the visual acquisition module, and identify the outline, feature points and coordinate position of the garment piece relative to the workbench through an image recognition algorithm; based on the identified feature points and a preset reference coordinate system, calculate the deviation between the actual position and the expected position of the garment piece; based on the calculated deviation, the control module generates an execution instruction to control the distance and / or rotation angle required for each actuator of the robot loading module to move in a straight line.
[0013] Compared with the prior art, the vision-based garment feeding mechanism of the present invention has the following beneficial technical effects:
[0014] Before the garment pieces are anchored, the non-contact garment flattening module uses a high-pressure airflow to spread the garments evenly on the work surface before loading, reducing loading errors caused by wrinkles. The airflow from the high-pressure airflow quickly and evenly affects the surface of the garment, effectively eliminating wrinkles and unevenness. The flattening process eliminates the need for physical contact, preventing additional damage or deformation to the garment pieces. The flattening module's design ensures that the garments are uniformly flat before loading, providing standardized image input to the vision system and improving positioning accuracy.
[0015] The non-contact garment anchoring module uses a negative pressure system to secure garment pieces, preventing them from moving or deforming due to contact during the loading process. The negative pressure system creates a partial vacuum beneath the work surface, firmly securing the garment pieces and preventing them from shifting during loading. This non-contact securing method reduces potential damage to the garment pieces from physical clamping or friction, preserving their pristine texture and appearance. The negative pressure anchoring module's design allows the garments to remain stable during loading, ensuring accurate and consistent gripping by the robotic arm.
[0016] The collaborative operation of these two modules enables efficient and accurate loading without damaging the garment parts. Non-contact handling reduces physical deformation of the garment parts during loading, maintaining their integrity and improving the quality of the final product. The automated operation of the flattening and anchoring modules reduces manual intervention, lowers labor costs, and improves the overall automation level of the production line.
[0017] The introduction of this automated loading mechanism improves production efficiency, shortens production cycles, and accelerates time to market. It reduces production errors and rework caused by wrinkled or shifted garment pieces, lowering production costs. It also increases production line flexibility, adapting to a wider variety of garment piece loading requirements and enhancing market adaptability.
[0018] In summary, the non-contact garment piece flattening module and non-contact garment piece anchoring module in the vision-based garment piece feeding mechanism, through their unique design and functions, provide effective technical support for the automated production of the garment manufacturing industry, significantly improving production efficiency and product quality. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] Figure 1 It is a structural diagram of the present utility model.
[0020] Figure 2 It is a structural schematic diagram of the non-contact garment piece anchoring module of the utility model.
[0021] Figure 3 It is a structural schematic diagram of the non-contact garment flattening module of the utility model.
[0022] Figure 4 yes Figure 3 A partial enlarged view of .
[0023] Figure 5 It is a structural diagram of the loading module of the robot arm of the utility model.
[0024] Figure 6 This is a structural diagram of the loading module of the robot arm of the utility model from another perspective.
[0025] Figure 7 yes Figure 6 A partial enlarged view of . DETAILED DESCRIPTION
[0026] The present invention will be further described below with reference to the accompanying drawings.
[0027] As shown in the figure, a vision-based garment feeding mechanism includes a machine 100, which is equipped with a vision acquisition module 500, a robot feeding module 600, a non-contact garment piece anchoring module 700 and a non-contact garment piece flattening module 800 arranged on the machine 100; the non-contact garment piece anchoring module 700 includes a table 701 fixed on the machine 100, a negative pressure box 702 is fixed below the table 701, the upper surface of the negative pressure box 702 is flush with the table 701 and is densely covered with air holes, and the negative pressure box 702 is connected to the negative pressure air path; the non-contact garment piece flattening module 800 includes two first guide rails 801 fixed on the machine 100 and arranged in the left and right directions, and Two first guide rails 801 slide together with a slider 802 fixed to a base plate 803; a synchronous belt 804 is also mounted on the machine 100, driven by a motor 805; the base plate 803 is fixed to the upper or lower layer of the synchronous belt 804; a first cylinder 806 with an up-and-down stroke is fixed to the base plate 803, the piston rod of the first cylinder 806 being fixed to a connecting plate 807, the ends of which are respectively fixed to vertical plates 808. The gap between the two vertical plates 808 passes through the table 701 and is fixed to a flattening tube 809. One end of the flattening tube 809 is sealed, and the other end is connected to a high-pressure air circuit via a quick connector 811; a row of air injection holes 810 are provided on the wall of the flattening tube 809, facing the table 701. "Non-contact" here means that there is no mechanical contact, such as directly touching or grasping the object without using a clamp, fingers, or other physical obstacles.
[0028] The visual acquisition module 500 is responsible for capturing images of garment pieces placed on the machine table 100. This module includes a camera and corresponding lighting system to obtain clear images of the garment pieces. The control module receives and processes images of the garment pieces on the table 701 captured by the visual acquisition module 500. Using an image recognition algorithm, it identifies the garment piece's outline, feature points, and coordinate position relative to the worktable. Based on the identified feature points and a preset reference coordinate system, it calculates the deviation between the garment piece's actual position and the desired position. Based on the calculated deviation, the control module generates execution instructions to control the required linear movement distance and / or rotation angle of the various actuators in the robot loading module 600. The garment pieces are placed on the table 701 of the machine table 100 via the upstream mechanism. The non-contact garment piece flattening module 800 begins operation. The synchronous belt 804, driven by the motor 805, moves the base plate 803 and the slider 802 on it. The first cylinder 806 drives the connecting plate 807 and the vertical plate 808, bringing the flattening tube 809 closer to the table 701. The flattening tube 809 is connected to the high-pressure air circuit via a quick connector 811. When the high-pressure gas is ejected through the air jet 810, it blows back and forth across the garment pieces to eliminate wrinkles and spread them evenly on the table 701. Once the garment pieces are flattened, the non-contact garment anchoring module 700 begins operation. The negative pressure box 702 generates negative pressure through the air holes on its upper surface, firmly adsorbing the garment pieces to the table 701 without any physical clamps or contact. At this time, the robotic arm loading module 600 begins operation. The visual acquisition module 500 transmits the image of the garment pieces to the control module, which calculates the movement path that the robotic arm needs to execute. Based on the calculation results, the robotic arm loading module 600 accurately moves above the garment pieces and performs the grasping action. After the robotic arm grasps the garment pieces, it moves the pieces to the next production link according to the preset path or real-time updated instructions. After completing one loading operation, the robotic loading module 600 returns to its starting point to prepare for the next loading operation, while the flattening module 800 and anchoring module 700 prepare for the next garment piece. This entire process automates the loading of garment pieces, reducing manual intervention and improving efficiency and accuracy. The non-contact design minimizes damage to garment pieces and ensures a smooth loading process and high quality.
[0029] In another preferred embodiment, axially vertical guide sleeves 812 are fixed to the base plate 803 on the front and rear sides of the first cylinder 806. The guide sleeves 812 slidably engage guide rods 813, the upper ends of which are fixedly connected to the connecting plate 807. By fixing the guide sleeves to the base plate and slidingly engaging with the guide rods, this design improves the precision, stability, and reliability of the cylinder's movement, playing a significant role in ensuring the efficient and accurate operation of the entire garment piece feeding mechanism.
[0030] In this embodiment, the robot loading module 600 includes a first linear slide 601 fixed on the machine platform 100 with a left-right stroke, a second linear slide 602 with a front-to-back stroke is fixed on the slider of the first linear slide 601, a fixed vertical plate 603 is fixed on the slider of the second linear slide 602, a second cylinder 604 with an up-down stroke is fixed on the fixed vertical plate 603, second guide rails 605 are respectively fixed on the fixed vertical plates 603 on the front and back sides of the second cylinder 604, a second slider 606 that slides with the second guide rail 605 is fixed to a mounting frame 607, a servo motor 608 that can accurately rotate to a set angle according to a control signal is fixed on the mounting frame 607, the motor shaft of the servo motor 608 is fixed to a mounting plate 609 parallel to the table surface 701, and four pneumatic clamps are installed on the mounting plate 609. The pneumatic clamp includes a clamp cylinder 610, a fixed jaw 611 is fixed at the end of the clamp cylinder 610, and the fixed jaw 611 is the static part of the pneumatic clamp. A movable jaw 612 is fixed on the piston rod of the clamp cylinder 610, and the movable jaw 612 is the movable part of the pneumatic clamp. The movable jaw 612 and the fixed jaw 611 cooperate with each other to form a clamping pair. Two straight slots 613 with their lengths in the front-to-back direction are respectively opened at the front and rear ends of the right side of the mounting plate 609. Two screws are passed through the two straight slots 613 and then screwed to the clamping jaw cylinder 610, so that the clamping positions of the two pneumatic clamps in the front-to-back direction can be adjusted; a small circular through hole and an arc slot 614 with the small circular through hole as the center and a central angle greater than or equal to 90° are respectively opened at the front and rear ends of the left side of the mounting plate 609. Two screws are passed through the small circular through hole and the arc slot 614 and then screwed to the clamping jaw cylinder 610, so that the clamping angles of the other two pneumatic clamps in the front-to-back direction can be adjusted. A first oil pressure buffer 615 is fixed on the fixed vertical plate 603 on the front side of the second cylinder 604, and the piston rod of the first oil pressure buffer 615 can abut against the mounting frame 607; a second oil pressure buffer 616 is fixed on the mounting frame 607 on the rear side of the second cylinder 604, and the piston rod of the second oil pressure buffer 616 passes through the mounting frame 607 and is fixed to the fixed vertical plate 603.
[0031] By using a servo motor 608 and pneumatic grippers, the robot arm can precisely rotate to a set angle and perform a stable gripping action, improving the accuracy of the loading process. The first and second linear slides 601 and 602 provide left-right and front-back movement, expanding the robot arm's operating range. The second pneumatic cylinder 604 provides up-and-down movement, allowing the robot arm to adapt to operations at varying heights. The design of the straight slots 613 and the small round through-holes and arc-shaped slots 614 allows for adjustment of the pneumatic gripper's front-to-back position and gripping angle, increasing the robot arm's flexibility and adaptability. The first and second hydraulic buffers 615 and 616 absorb the shock and vibration generated by the robot arm's rapid movement, protecting the robot arm and the garment being manipulated. The use of hydraulic buffers ensures that the robot arm stops smoothly at the end of its stroke, preventing damage caused by sudden stops. The design of the fixed vertical plate 603 and mounting bracket 607 provides stable support, ensuring the stability of the robot arm during operation. The modular design makes the various components of the robot arm easy to maintain and replace. This design can adapt to garment pieces of different sizes and weights, increasing the versatility of the robot loading module.
[0032] The present invention may be summarized in other specific forms that do not violate the spirit or main features of the present invention. Therefore, no matter from which point of view, the above embodiments of the present invention can only be regarded as an illustration of the present invention and not as a limitation of the present invention. The claims indicate the scope of the present invention, while the above description does not indicate the scope of the present invention. Therefore, any changes within the meaning and scope equivalent to the claims of the present invention should be considered as included within the scope of the claims of the present invention.
Claims
1. A garment piece feeding mechanism based on vision, comprising a machine (100), a vision acquisition module (500) and a manipulator feeding module (600) configured on the machine (100), characterized in that: The non-contact garment piece anchoring module (700) and the non-contact garment piece flattening module (800) are arranged on the machine (100); the non-contact garment piece anchoring module (700) includes a table (701) fixed on the machine (100), a negative pressure box (702) is fixed below the table (701), the upper surface of the negative pressure box (702) is flush with the table (701) and is densely covered with air holes, and the negative pressure box (702) is connected to the negative pressure air path; the non-contact garment piece flattening module (800) includes two first guide rails (801) fixed on the machine (100) and arranged in the left and right directions, and a slider (802) slidingly matched with the two first guide rails (801) is fixed to the bottom plate (803); the machine A synchronous belt (804) is also provided on (100), and the synchronous belt (804) is driven by a motor (805); the bottom plate (803) is fixedly connected to the upper layer or the lower layer of the synchronous belt (804); a first cylinder (806) having a stroke in the up-down direction is fixedly provided on the bottom plate (803), the piston rod of the first cylinder (806) is fixedly connected to a connecting plate (807), and both ends of the connecting plate (807) are respectively fixedly connected to vertical plates (808), and a gap between the two vertical plates (808) passes through the table (701) and is fixedly connected to a flattening tube (809), one end of the flattening tube (809) is closed, and the other end is connected to a high-pressure gas circuit through a quick connector (811); a row of air injection holes (810) facing the table (701) is opened on the wall of the flattening tube (809).
2. A vision-based garment feeding mechanism according to claim 1, characterized in that: Axially vertical guide sleeves (812) are fixedly provided on the bottom plates (803) at the front and rear sides of the first cylinder (806), respectively. The guide sleeves (812) are slidably fitted with the guide rods (813), and the upper ends of the guide rods (813) are fixedly connected to the connecting plates (807).
3. The vision-based garment feeding mechanism according to claim 1, characterized in that: The manipulator loading module (600) includes a first linear slide (601) fixed on the machine platform (100) with a left-right stroke, a second linear slide (602) fixed on the slider of the first linear slide (601) with a front-back stroke, a fixed vertical plate (603) fixed on the slider of the second linear slide (602), a second cylinder (604) fixed on the fixed vertical plate (603) with a top-bottom stroke, second guide rails (605) fixed on the fixed vertical plates (603) on both sides of the front and back of the second cylinder (604), a second slide (606) slidingly engaged with the second guide rail (605) fixedly connected to a mounting frame (607), a servo motor (608) fixedly connected to the mounting frame (607) that can accurately rotate to a set angle according to a control signal, a motor shaft of the servo motor (608) fixedly connected to a mounting plate (609) parallel to the table surface (701), and four pneumatic grippers installed on the mounting plate (609).
4. A vision-based garment feeding mechanism according to claim 3, characterized in that: The pneumatic clamp comprises a clamp cylinder (610), a fixed jaw (611) is fixed at the end of the clamp cylinder (610), the fixed jaw (611) is the static part of the pneumatic clamp, a movable jaw (612) is fixed on the piston rod of the clamp cylinder (610), the movable jaw (612) is the movable part of the pneumatic clamp, and the movable jaw (612) and the fixed jaw (611) cooperate with each other to form a clamping pair.
5. A vision-based garment feeding mechanism according to claim 3 or 4, characterized in that: Two straight slots (613) with their lengths in the front-to-back direction are respectively opened at the front and rear ends of the right side of the mounting plate (609), and two screws are passed through the two straight slots (613) and then screwed to the clamping jaw cylinder (610), so that the clamping positions of the two pneumatic clamps in the front-to-back direction can be adjusted; a small round through hole and a circular arc slot (614) with the small round through hole as the center and a central angle greater than or equal to 90° are respectively opened at the front and rear ends of the left side of the mounting plate (609), and two screws are passed through the small round through hole and the circular arc slot (614) and then screwed to the clamping jaw cylinder (610), so that the clamping angles of the other two pneumatic clamps in the front-to-back direction can be adjusted.
6. A vision-based garment feeding mechanism according to claim 3 or 4, characterized in that: A first oil pressure buffer (615) is fixedly mounted on the fixed vertical plate (603) on the front side of the second cylinder (604), and the piston rod of the first oil pressure buffer (615) is capable of abutting against the mounting frame (607); a second oil pressure buffer (616) is fixedly mounted on the mounting frame (607) on the rear side of the second cylinder (604), and the piston rod of the second oil pressure buffer (616) passes through the mounting frame (607) and is fixedly mounted on the fixed vertical plate (603).
7. A vision-based garment feeding mechanism according to any one of claims 1 to 4, characterized in that: The system further comprises a control module, wherein the control module is capable of receiving and processing an image of a garment piece on the table (701) acquired by the visual acquisition module (500), and identifying the contour, feature points and coordinate position of the garment piece relative to the workbench through an image recognition algorithm; calculating the deviation between the actual position and the expected position of the garment piece based on the identified feature points and a preset reference coordinate system; and generating an execution instruction based on the calculated deviation to control the linear movement distance and / or rotation angle of each actuator of the manipulator loading module (600).