Cloth feeding method of circular knitting machine based on visual identification

Abstract

The invention discloses a circular knitting machine cloth feeding method based on visual identification, and belongs to the field of textile machinery automation, the method comprises the following steps: S1, identifying a rod core and calculating the diameter of a cloth roll: respectively identifying two ends of the rod core through forward-looking cameras arranged on mechanical clamping jaws arranged on two sides and used for respectively clamping two ends of the rod core, obtaining and recording the height of the rod core, and calculating the diameter of the cloth roll; s2, the rod core is clamped, specifically, the mechanical clamping jaw moves forwards, a downward-looking camera arranged on the mechanical clamping jaw stops after recognizing the rod core, then the mechanical clamping jaw descends to the height of the rod core, and the clamping action is executed; s3, taking out the yardage roll: after the mechanical clamping jaw grabs the rod core, enabling the yardage roll to rotate for n1 turns by a cloth releasing mechanism arranged on the mechanical clamping jaw to release cloth, and then vertically moving upwards to the initial height; and S4, the cloth roll is placed, specifically, the cloth unwinding mechanism continues to rotate the cloth roll for n2 turns for cloth unwinding, the mechanical clamping jaw moves towards the upper portion of the placement area in the horizontal direction, and finally the cloth roll is placed in the placement area.

Description

Technical Field

[0001] The present invention relates to the field of textile machinery automation, and in particular to a method for discharging cloth from a circular knitting machine based on visual recognition. Background Art

[0002] The automation of textile machinery is an important field in the intelligent transformation of the manufacturing industry. Among them, the circular knitting machine, as one of the core equipment in the textile industry, is widely used in the production of high-quality knitted fabrics. After the traditional circular knitting machine completes the processing of the cloth roll, manual operations such as replacing the empty bobbin core, grasping the cloth roll, cutting the cloth, and installing the new bobbin core are required. With the increase in labor costs and the growing demand for intelligent manufacturing, the industry urgently needs to realize unmanned operation of the entire process of cloth roll handling through automation technology. At present, some enterprises have tried to introduce automatic guided vehicles or autonomous mobile robots to assist the circular knitting machine in operation. For example, rail-mounted AGVs are used for material transportation, or robotic arms are controlled by preset programs to perform grasping actions. However, the space in the cloth discharging area of the circular knitting machine is limited. Due to relying on a single sensor and preset programs, the traditional system is difficult to accurately identify the position of the bobbin core and the diameter of the cloth roll, and mis-grasping often occurs. In addition, the existing technology cannot calculate the change in the diameter of the cloth roll in real time, resulting in inaccurate control of the cloth tension, and easily causing cloth tearing or mechanical damage during the process of taking out and placing the cloth roll. At the same time, the lack of an effective visual recognition system to accurately judge the height and initial position of the bobbin core affects the accuracy of cloth roll grasping and placement. In view of the above problems, the existing technology urgently needs to be improved. Summary of the Invention

[0003] The purpose of the present application is to provide a method and system for discharging cloth from a circular knitting machine based on visual recognition, which has the advantages of improving the grasping accuracy of the cloth roll, reducing mis-grasping phenomena, realizing real-time calculation of the cloth roll diameter, and enhancing the stability of cloth tension control.

[0004] The present application provides a method for discharging cloth from a circular knitting machine based on visual recognition, and the technical solution is as follows: A method for discharging cloth from a circular knitting machine based on visual recognition, characterized in that: S1. Identify the bobbin core and calculate the diameter of the cloth roll: The front cameras respectively arranged on the mechanical jaws for clamping the two ends of the bobbin core on both sides are used to identify the two ends of the bobbin core respectively, obtain and record the height of the bobbin core, and calculate the diameter of the cloth roll; S2. Clamp the bobbin core: The mechanical jaws move forward, and stop when the lower camera arranged on the mechanical jaws recognizes the bobbin core, then lower the mechanical jaws to the height of the bobbin core and perform the clamping action; S3. Take out the cloth roll: After the mechanical jaws grasp the bobbin core, the cloth discharging mechanism arranged on the mechanical jaws rotates the cloth roll for n1 turns to discharge the cloth, and then moves vertically upward to the initial height; S4. Place the cloth roll: The cloth discharging mechanism continues to rotate the cloth roll for n2 turns to discharge the cloth, the mechanical jaws move horizontally above the placement area, and finally place the cloth roll in the placement area.

[0005] Furthermore, the present application also proposes to provide a first identification mark on the surfaces at both ends of the rod core, and calculate the height of the rod core by identifying the first identification mark through a front camera.

[0006] Furthermore, the present application also proposes to provide a second identification mark in the guiding groove of the circular knitting machine. The height position of the second identification mark is the initial height of the rod core before the cloth is wound. The front camera simultaneously identifies the first identification mark and the second identification mark to obtain the height of the rod core and the initial height of the rod core respectively, and obtains the height difference ∆h = h1 - h0, where h1 is the height of the rod core and h0 is the initial height of the rod core, that is, the cloth roll diameter d = 2 * ∆h.

[0007] Furthermore, the present application also proposes that in S3, n1 = k * (h2 - h1) / πd, where k is the cloth feeding coefficient and h2 is the initial height of the mechanical gripper.

[0008] Furthermore, the present application also proposes that in S4, n2 = k * L / πd, where k is the cloth feeding coefficient and L is the horizontal moving distance of the mechanical gripper.

[0009] Furthermore, the present application also proposes that the cloth feeding coefficient k is greater than 1 to control the tension of the cloth within a certain range.

[0010] Furthermore, the present application also proposes that in S4, the cloth feeding mechanism rotates the cloth roll while the cloth roll is moving for cloth feeding.

[0011] As can be seen from the above, a method and system for unwinding cloth of a circular knitting machine based on visual recognition provided by the present application accurately judge the height and initial position of the rod core through a visual recognition system, and combine the multi-dimensional motion control of a mechanical gripper to realize the real-time calculation of the diameter change and the precise adjustment of the tension during the cloth roll grasping process, having the technical effects of improving the accuracy of automated operations and reducing cloth damage. Description of the Drawings

[0012] Figure 1 is a schematic structural diagram of an unwinding cloth device of a circular knitting machine; Detailed Embodiments

[0013] The technical solutions in the present application will be clearly and completely described below with reference to the accompanying drawings in the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the present application usually described and illustrated in the accompanying drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed present application, but merely represents the selected embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative efforts fall within the scope of protection of the present application. It should be noted that similar reference numerals and letters indicate similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of the present application, the terms "first", "second", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0014] In the prior art, the automation of textile machinery is an important field in the intelligent transformation of the manufacturing industry. As a core equipment in the textile industry, the circular knitting machine traditionally relies on manual labor to complete the processes of replacing empty bobbin cores, grasping fabric rolls, and cutting fabrics. Existing automation systems have problems such as single sensors, inability to accurately identify the position of the bobbin core, and out-of-control fabric tension, resulting in a high mis-grasp rate and easy fabric tearing. For example, when using an orbital AGV to transport materials, due to the inability to dynamically identify the diameter of the fabric roll, the robotic arm fails to grasp or the fabric tension is abnormal.

[0015] To solve the above problems, aiming at the problems of bobbin core positioning deviation and tension control during the process of grasping fabric rolls, a multi-dimensional vision detection is considered to replace the single-sensor solution. First, establish the mathematical relationship between the height of the bobbin core and the diameter of the fabric roll to solve the problem of low efficiency of traditional manual measurement. Second, introduce real-time visual feedback into the movement trajectory of the mechanical gripper to eliminate the positioning error of the robotic arm. Finally, design a staged fabric releasing mechanism to control the tension change through the rotation of the fabric roll at different moving stages.

[0016] Therefore, the present application proposes a method for discharging fabric under a circular knitting machine based on visual recognition, including the steps of: identifying both ends of the bobbin core through the front-view cameras on both sides of the mechanical grippers and calculating the diameter of the fabric roll; the mechanical grippers move to the recognition position with the lower-view camera and then execute the grasping; after grasping, the fabric releasing mechanism rotates the fabric roll during the vertical movement for the first fabric release; when moving horizontally to the placement area, continue to rotate the fabric roll to complete the second fabric release.

[0017] Among them, the front camera refers to an optical imaging device installed at the front end of the mechanical gripper, which can be specifically implemented by an industrial-grade CMOS camera combined with a barcode recognition module, and is used to capture the recognition marks at the end of the rod core and calculate the three-dimensional coordinates. The downward camera refers to a vision sensor installed at the bottom of the gripper, which can be specifically equipped with a laser ranging module, and is used to monitor the position of the rod core in real time and feedback the descending distance of the gripper. The cloth feeding mechanism refers to a rotary drive device integrated in the mechanical gripper, which controls the number of turns of the cloth roll rotation through a preset algorithm to adjust the cloth tension.

[0018] Specifically, the front cameras on both sides of the mechanical gripper synchronously collect images of both ends of the rod core, calculate the actual height of the rod core through the marker recognition algorithm, and deduce the cloth roll diameter based on the height difference. During the movement of the gripper, the downward camera continuously scans the working area, and triggers a stop signal when the recognition mark on the rod core is recognized, guiding the gripper to accurately descend to the target height to complete the clamping. During the vertical lifting stage, the cloth feeding mechanism actively rotates the cloth roll to release some cloth, eliminating the sudden change in tension under the action of gravity; during the horizontal movement stage, the rotation amount of the cloth roll is secondarily controlled in combination with the path length to ensure that the cloth remains evenly relaxed during the movement.

[0019] Compared with the prior art, the traditional method relies on a fixed program to control the movement trajectory of the robotic arm and cannot adapt to the offset of the rod core position or the change of the cloth roll diameter. This solution dynamically corrects the positioning coordinates of the gripper through the vision system to achieve centimeter-level spatial positioning accuracy. The prior art uses a single cloth feeding action, which is likely to cause cloth accumulation or stretching and breaking. This method forms a dual tension adjustment through a phased cloth feeding mechanism, releasing the basic tension during the vertical movement stage and compensating for the path difference during the horizontal movement stage.

[0020] Through the above technical solutions, this application realizes the real-time dynamic recognition of the rod core position during the cloth roll grasping process, and solves the problem of mis-grasping caused by the visual blind area in the traditional system. Through the cloth feeding method of rotating the cloth roll in stages, the matching degree between the cloth release speed and the movement trajectory of the robotic arm is effectively controlled, avoiding wrinkles or tears in the cloth due to sudden changes in tension. The multi-dimensional visual collaborative positioning of the mechanical gripper controls the calculation error of the cloth roll diameter within the millimeter range, ensuring the accuracy of subsequent placement actions.

[0021] This application further proposes to set first recognition marks on the surfaces of both ends of the rod core, and calculate the height of the rod core by recognizing the first recognition marks through the front camera.

[0022] Among them, the first identification mark refers to the physical marks set at both ends of the core rod that can be recognized by the vision system. Specifically, it can be implemented by using high-contrast QR codes, reflective patches or geometric patterns with specific shapes. The setting position thereof needs to form a fixed spatial relationship with the axis of the core rod. The front-view camera refers to an image acquisition device mounted at the front end of the mechanical gripper and having a three-dimensional spatial positioning function. Specifically, it can be implemented by using a binocular stereo vision system or an industrial camera with a depth sensor. The three-dimensional coordinate calculation is completed by capturing the two-dimensional pixel coordinates and spatial depth information of the identification mark.

[0023] Specifically, the identification marks set on the surfaces at both ends of the core rod establish a spatial mapping relationship with the axis of the core rod through high-precision calibration. During the operation process, the front-view camera synchronously acquires images of the identification marks at both ends of the core rod, calculates the three-dimensional coordinates of the two markers in the camera coordinate system by using three-dimensional vision algorithms, and combines the preset physical size parameters of the core rod to deduce the actual position and the vertical height value of the axis of the core rod in space. This calculation process eliminates the image interference caused by fabric wrinkles and environmental light changes, ensuring that the height measurement error is controlled within the positioning accuracy range allowed by the mechanical gripper.

[0024] Compared with the prior art, the traditional method relies on a single contact sensor to measure the position of the core rod, and is prone to misjudgment due to fabric coverage or metal surface reflection. This solution uses non-contact vision measurement technology to set standardized identification marks on the core rod body, which can penetrate fabric occlusion and accurately identify the physical characteristics of core rods made of different materials, fundamentally avoiding the problem of grasping position deviation caused by material mixing.

[0025] Through the above technical solution, this application realizes the precise positioning of the spatial position of the core rod, and solves the problem of height calculation deviation caused by the lack of visual features in the traditional cloth laying process. This solution effectively eliminates the empty grasping or collision accidents caused by the mechanical gripper due to height misjudgment, ensures the reliability of the cloth roll grasping process, and provides accurate height reference data for the subsequent cloth laying operation.

[0026] This application further proposes that a second identification mark is provided in the guide groove of the circular knitting machine. The height position of the second identification mark is the initial height of the core rod before the cloth is wound. The front-view camera simultaneously recognizes the first identification mark and the second identification mark to obtain the height of the core rod and the initial height of the core rod respectively, and obtains the height difference ∆h = h1 - h0, where h1 is the height of the core rod and h0 is the initial height of the core rod, that is, the cloth roll diameter d = 2 * ∆h.

[0027] Among them, the second identification mark refers to a physical mark set at a specific position of the guide groove of the circular knitting machine, which can be specifically implemented by a reflective patch or a QR code label. Its installation height is aligned with the original position reference when the core rod has no wound fabric, and is used to establish an absolute height reference system. The front-view camera refers to an optical sensor assembly integrated on the mechanical gripper, which can be specifically implemented by an industrial-grade CCD or CMOS camera module, and captures and analyzes the spatial coordinates of the first and second identification marks in real time through an image processing algorithm. The calculation model of the height difference ∆h refers to a mathematical relationship implemented through image coordinate transformation and triangulation algorithms, which can be specifically implemented by a linear mapping relationship between the pixel displacement and the calibration parameters, and converts the visual data into a physical height difference.

[0028] Specifically, before clamping the core rod, the front-view camera synchronously acquires the image of the second identification mark of the guide groove and the images of the first identification marks at both ends of the core rod, and respectively extracts the vertical coordinate data of the two through image processing technology. The fixed height of the second identification mark is used as the initial reference value, which forms a dynamic comparison with the current core rod height, and the measurement deviation caused by mechanical vibration or environmental interference is eliminated by using the height difference calculation model. Based on the linear relationship between the cloth roll diameter and the height difference, ∆h is the cloth roll radius, which provides an accurate input for the number of turns control of the subsequent cloth releasing mechanism.

[0029] Compared with the prior art, the traditional method relies on a single sensor to measure the height of the core rod, cannot distinguish the diameter change error caused by the cloth layer superposition, and lacks a fixed reference, resulting in the measurement result being easily affected by the equipment vibration. This solution establishes a dual-mark collaborative recognition mechanism, compares the dynamic measurement value with the fixed reference value in real time, not only eliminates the environmental interference factors, but also realizes the accurate calculation of the cloth roll diameter, and fundamentally solves the problem of tension out-of-control caused by inaccurate diameter measurement.

[0030] Through the above technical solution, this application effectively solves the problems of cloth tension out-of-control and mechanical misoperation caused by cloth roll diameter measurement error in the traditional lower cloth process. By establishing a height difference calculation model through a dual-mark recognition system, it ensures the real-time and accurate acquisition of the diameter parameter, provides reliable data support for the rotation number control of the cloth releasing mechanism, avoids cloth tearing or mechanical damage caused by uneven tension, and at the same time reduces the risk of misgrasping.

[0031] This application further proposes that during the vertical movement stage of the mechanical gripper, the cloth roll rotates n1 turns through the cloth releasing mechanism for cloth releasing. The calculation formula for the number of cloth releasing turns n1 is n1 = k*(h2 - h1) / (πd), where k is the cloth releasing coefficient, h2 is the initial height of the mechanical gripper, h1 is the height of the core rod, and d is the cloth roll diameter.

[0032] Among them, the cloth feeding coefficient k is a compensation parameter used to adjust the tension control during the cloth release process. Specifically, it can be achieved through a preset parameter library or real-time sensor feedback. Its value is dynamically adjusted according to the cloth material, thickness, and process requirements, and is used to form a tension buffer mechanism during the cloth feeding stage. The initial height h2 of the mechanical gripper refers to the reference position height when the mechanical gripper starts to move vertically. Specifically, it can be obtained in real time through an encoder or a laser ranging device, and is used to calculate the number of rotation turns corresponding to the length of the cloth to be released from the cloth roll. Specifically, during the vertical movement stage of the mechanical gripper, the cloth feeding number of turns n1 is dynamically generated by the formula n1 = k*(h2 - h1) / (πd). When the mechanical gripper moves down from the initial height h2 to the core height h1, the vertical movement distance is h2 - h1, and the cloth release length corresponding to this distance is the vertical movement distance itself. By dividing the cloth release length by the cloth length per turn πd, the theoretical number of rotation turns is obtained. On this basis, the cloth feeding coefficient k is introduced as a tension compensation factor. When k is greater than 1, the cloth feeding number of turns n1 increases, causing a slack in the cloth during the release process, avoiding tension overload caused by the asynchronous speed between the vertical movement speed of the mechanical gripper and the cloth release speed of the cloth roll. For example, when k is set to 1.2, the actual cloth feeding number of turns increases by 20% compared to the theoretical value, providing an additional release margin for the cloth, thus maintaining the stability of the cloth tension during the lifting process of the mechanical gripper.

[0033] Compared with the prior art, the traditional method only controls the cloth feeding number of turns based on a fixed program and cannot be dynamically adjusted according to the actual cloth roll diameter and the displacement of the mechanical gripper, resulting in uncontrollable cloth tension. In this solution, by establishing a mathematical model including the height parameter of the mechanical gripper and combining real-time visual measurement data, an adaptive matching between the cloth feeding number of turns and the actual working conditions is achieved. Especially the introduction of the cloth feeding coefficient k makes the tension control adjustable, and can optimize the parameters according to the physical characteristics of different cloths, solving the problem of cloth stretching caused by rigid cloth feeding.

[0034] Through the above technical solution, this application effectively solves the risk of cloth tearing caused by uncontrollable tension during the vertical movement stage of the cloth roll. By calculating the cloth feeding number of turns in real time and superimposing the tension compensation coefficient, it ensures that the cloth release speed is synchronized with the displacement of the mechanical gripper, forming a buffer area during the lifting process of the cloth roll, and avoiding instantaneous tension peaks caused by speed differences. At the same time, the adjustable nature of the cloth feeding coefficient k enables the system to adapt to the tension sensitivity differences of different materials such as silk and chemical fibers, improving the process compatibility of the automated cloth dropping system.

[0035] This application further proposes that in S4, n2 = k*L / πd, where k is the cloth feeding coefficient and L is the horizontal movement distance of the mechanical gripper.

[0036] Among them, the cloth release coefficient k refers to the proportional factor between the cloth release length and the moving distance of the mechanical gripper. Specifically, it can be achieved through preset empirical values or real-time tension feedback adjustment, and is used to balance the influence of different cloth material characteristics and cloth roll density on the cloth release amount. The horizontal moving distance L of the mechanical gripper refers to the horizontal displacement between the placement area and the grasping position, and can be specifically achieved through real-time measurement by an encoder or a laser rangefinder, and is used to accurately calculate the length of the cloth that the cloth roll needs to release. πd represents the circumference of the cloth roll, where d is the diameter of the cloth roll, and can be specifically obtained by detecting and calculating the change in the height of the core rod by a vision recognition system, and is used to establish the conversion relationship between the number of turns and the cloth release length.

[0037] Specifically, when the mechanical gripper carries the cloth roll and moves horizontally towards the placement area, the cloth release mechanism dynamically calculates the required number of rotation turns n2 based on the real-time obtained moving distance L, in combination with the preset cloth release coefficient k and the current cloth roll circumference πd. The cloth roll is driven to rotate n2 turns by a servo motor, so that the released cloth length completely matches the moving path of the mechanical gripper. In this process, the cloth release coefficient k, as an adjustment variable, can be adaptively adjusted according to physical characteristics such as the elastic modulus and friction coefficient of the cloth, ensuring that the cloth always maintains a constant tension during movement.

[0038] Compared with the prior art, the traditional method uses a fixed number of turns for cloth release and cannot adjust the cloth release amount in real time according to the moving distance, resulting in cloth accumulation or excessive stretching. This solution realizes the closed-loop control of the cloth release length by establishing the mathematical relationship between the moving distance and the cloth release number of turns, and solves the problem of tension fluctuation caused by path change.

[0039] Through the above technical solution, this application can accurately match the moving distance of the mechanical gripper with the cloth release length, effectively preventing the cloth from being pulled and broken due to insufficient cloth release amount, or the cloth from being wrinkled due to excessive cloth release amount. For example, when processing spandex fabric with large elasticity, the k value can be increased to compensate for the cloth retraction amount; when handling high-density cloth rolls, the k value is reduced to avoid excessive cloth release, so as to keep the cloth tension within the safety threshold under various working conditions.

[0040] This application further proposes that the cloth release coefficient k is greater than 1, so as to control the cloth tension within a certain range.

[0041] Among them, the cloth release coefficient k refers to the proportional parameter of the actual rotation turns of the cloth roll to the theoretical rotation turns, which can be specifically realized by the servo motor speed control module. This module dynamically adjusts the rotation rate of the cloth roll according to the moving speed of the mechanical gripper. The setting that the cloth release coefficient k is greater than 1 makes the released length of the cloth greater than the required length of the cloth generated by the movement of the mechanical gripper, thus forming a controllable redundant release amount. Among them, controlling the tension within a certain range means real-time monitoring of the force state of the cloth through a tension sensor, which can be specifically realized by a strain gauge sensor or an optoelectronic tension detection device. When it is detected that the tension exceeds the threshold, the value of the cloth release coefficient k is adjusted through a closed-loop feedback system.

[0042] Specifically, during the cloth release process of the cloth roll rotation, a pulling force is generated on the cloth when the mechanical gripper moves. When the cloth release coefficient k is set to be greater than 1, the length of the cloth released per rotation of the cloth roll is equal to the theoretical release length multiplied by the value of k. For example, when the mechanical gripper moves horizontally by a distance L, the theoretical required number of cloth roll rotation turns is L / (πd), and the actual rotation turns are set to k*L / (πd), where d is the diameter of the cloth roll. By actively increasing the cloth release amount, the cloth remains in a relaxed state during movement, avoiding the instantaneous tension peak caused by rigid pulling. At the same time, this redundant release amount provides a buffer space for the relative movement between the cloth and mechanical components, so that when the tension sensor detects that the tension exceeds the set threshold, the value of k can be immediately reduced to decrease the cloth release amount, thereby forming a dynamically balanced tension control mechanism.

[0043] In some specific embodiments, the value range of the cloth release coefficient k can be set to 1.2 - 1.5. This range can achieve a balance between the cloth relaxation degree and the moving efficiency through experimental verification. The cloth release mechanism uses a servo motor to drive the rotation of the cloth roll, and its rotation speed is synchronously linked with the moving speed of the mechanical gripper through a motion controller.

[0044] Compared with the prior art, the traditional method only controls the rotation turns of the cloth roll based on a fixed program and cannot compensate for the tension change caused by mechanical movement in real time. However, this solution introduces a dynamic adjustment mechanism of the cloth release coefficient k to actively offset the pulling effect caused by mechanical movement during the cloth release process, and realizes closed-loop control through sensor feedback, solving the problem of cloth fiber breakage caused by sudden tension changes.

[0045] Through the above technical solutions, this application effectively avoids the longitudinal tearing phenomenon of the cloth caused by excessive instantaneous tension during the cloth release process, and at the same time prevents the slippage accident of the rod core caused by excessive pulling. The cloth always maintains a uniform stress state during movement, and the integrity of its fiber structure is maintained, and there are no creases or deformations formed by local stress concentration on the cloth surface.

[0046] This application further proposes a technical solution for the cloth release mechanism to rotate the cloth roll while the cloth roll is moving for cloth release.

[0047] Among them, the cloth feeding mechanism refers to a mechanical component with the function of driving the cloth roll to rotate. Specifically, a servo motor can be used in combination with a gear transmission structure to achieve this, and the dynamic matching is formed by controlling the rotation speed of the motor and the moving speed of the mechanical gripper. Rotating the cloth roll to feed the cloth means continuously releasing the cloth during the horizontal movement. Specifically, the number of rotations of the cloth roll can be monitored in real time through an encoder, and a linear relationship is established with the displacement of the mechanical gripper, so as to ensure the synchronization of the released length of the cloth and the moving distance.

[0048] Specifically, when the mechanical gripper carries the cloth roll and moves horizontally towards the placement area, the cloth feeding mechanism drives the cloth roll to rotate at a specific angular velocity, so that the length of the cloth pulled out forms a corresponding relationship with the horizontal displacement of the mechanical gripper. During this process, the release rate of the cloth is adjusted in real time through the control system. For example, the rotation speed of the motor is dynamically corrected according to the change in the moving speed, so that the cloth always maintains a uniform tension in the moving path. This synchronous action avoids the over-stretching of the cloth caused by one-way movement or the accumulation and relaxation of the cloth caused by one-way rotation, thus achieving tension balance in two dimensions of spatial displacement and cloth roll rotation.

[0049] Compared with the prior art, traditional methods usually only release the cloth when the mechanical gripper is in a static state, or only rely on a fixed rotation speed to feed the cloth during the movement, resulting in a mismatch between the released length of the cloth and the displacement. This solution enables the cloth to continuously maintain stable tension during the dynamic displacement process through the coordinated control of movement and rotation, eliminating the instantaneous tension mutation caused by one-way actions.

[0050] Through the above technical solution, the present application realizes the continuous tension control of the cloth during the movement, effectively preventing the cloth from being torn due to local overload, and at the same time avoiding the wear of the mechanical structure caused by abnormal stress. The real-time matching of the cloth release rate and the moving speed makes the cloth between the cloth roll and the placement area always in a taut state, but the tension value is strictly limited within the material yield strength range.

[0051] The above are only the embodiments of the present application and are not used to limit the protection scope of the present application. For those skilled in the art, the present application can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for discharging fabric from a circular knitting machine based on visual recognition, characterized in that: S1. Identify the core rod and calculate the fabric roll diameter: Use front cameras installed on mechanical jaws on both sides for respectively clamping the two ends of the core rod to identify the two ends of the core rod, obtain and record the height of the core rod, and calculate the fabric roll diameter; S2. Clamp the core rod: The mechanical jaws move forward and stop after the lower camera installed on the mechanical jaws recognizes the core rod. Then, the mechanical jaws are lowered to the height of the core rod and a clamping action is performed; S3. Take out the fabric roll: After the mechanical jaws grab the core rod, the fabric discharging mechanism installed on the mechanical jaws rotates the fabric roll for n1 turns to discharge the fabric, and then moves vertically upward to the initial height; S4. Place the fabric roll: The fabric discharging mechanism continues to rotate the fabric roll for n2 turns to discharge the fabric. The mechanical jaws move horizontally above the placement area, and finally place the fabric roll in the placement area.

2. The method for discharging fabric from a circular knitting machine based on visual recognition according to claim 1, characterized in that: A first recognition mark is provided on the surfaces of both ends of the core rod, and the height of the core rod is calculated by the front camera recognizing the first recognition mark.

3. The method for discharging fabric from a circular knitting machine based on visual recognition according to claim 1, characterized in that: A second recognition mark is provided in the guiding groove of the circular knitting machine, and the height position of the second recognition mark is the initial height of the core rod before fabric is wound; The front camera simultaneously recognizes the first recognition mark and the second recognition mark to obtain the height of the core rod and the initial height of the core rod respectively, and obtains the height difference ∆h = h1 - h0, where h1 is the height of the core rod and h0 is the initial height of the core rod, that is, the fabric roll diameter d = 2 * ∆h.

4. The method for discharging fabric from a circular knitting machine based on visual recognition according to claim 1, characterized in that: n1 in S3 = k * (h2 - h1) / πd, where k is the fabric discharging coefficient and h2 is the initial height of the mechanical jaws.

5. The method for discharging fabric from a circular knitting machine based on visual recognition according to claim 1, characterized in that: n2 in S4 = k * L / πd, where k is the fabric discharging coefficient and L is the horizontal moving distance of the mechanical jaws.

6. The method for discharging fabric from a circular knitting machine based on visual recognition according to claim 3, characterized in that: The fabric discharging coefficient k is greater than 1 to control the tension of the fabric within a certain range.

7. The method for discharging fabric from a circular knitting machine based on visual recognition according to claim 1, characterized in that: In S4, the fabric discharging mechanism rotates the fabric roll while the fabric roll is moving to discharge the fabric.

Images