A method for pulling a film and double bagging

Abstract

The application discloses a kind of pull film double bag carrying methods, belong to bag carrying method field, including the following steps: judge which side mechanical arm reaches position first, and the other side mechanical arm waits in waiting position;Mechanical arm of first reaching position carries bag, and another side mechanical arm carries bag after bag carrying work is completed;Error generated in bag carrying work process is detected and optimized, and error includes camera positioning error, servo motor error, mechanical arm deformation error, coordinate system conversion error.The specific performance process of bag carrying work is as follows: obtain target bag position information by camera, then obtain corresponding global coordinates by coordinate system conversion;Servo motor angle is obtained by mechanical arm motion control;Control mechanical arm to complete grabbing.The application can accurately detect various errors occurring in the bag carrying process and optimize them, thereby avoiding the final packaging effect from being affected.

Description

TECHNICAL FIELD

[0001] The application relates to a bag moving method, in particular to a film pulling double bag moving method. BACKGROUND

[0002] There are various types of packaging machines, and there are many classification methods. From different perspectives, there are various classifications, according to product states, there are liquid, block and bulk packaging machines; according to packaging functions, there are inner packaging and outer packaging machines; according to packaging industries, there are food, daily chemical, textile and other packaging machines; according to packaging stations, there are single-station and multi-station packaging machines; according to automation degree, there are semi-automatic and fully-automatic packaging machines. With the improvement of people's living standards, various corresponding packaging equipment is needed on the market to meet people's consumer demand, therefore, the functions and production efficiency of the packaging machine need to be improved.

[0003] Film pulling bag moving is an important process of a packaging machine. With the development of technology, people have developed double film pulling bag moving equipment, which can complete bag moving alternately on both sides to improve efficiency. However, if errors occur during bag moving, the final packaging effect will be seriously reduced, affecting the working precision. However, there is no specific detection and optimization method for the errors in the bag moving process in the prior art.

[0004] Therefore, the skilled in the art provides a film pulling double bag moving method to solve the problems in the background art. SUMMARY

[0005] The application aims to provide a film pulling double bag moving method, which can accurately detect various errors occurring in the bag moving process and optimize them, thereby avoiding the influence on the final packaging effect, to solve the problems in the background art.

[0006] To achieve the above-mentioned purpose, the application provides the following technical scheme:

[0007] A film pulling double bag moving method, comprising the following steps:

[0008] Determine which side of the mechanical arm reaches the position first, and the other side of the mechanical arm waits at the waiting position;

[0009] The mechanical arm that reaches the position first performs bag moving work, and after the bag moving work is completed, the other side of the mechanical arm performs bag moving work;

[0010] Detect and optimize the errors occurring in the bag moving work process, including camera positioning errors, servo motor errors, mechanical arm deformation errors and coordinate system conversion errors.

[0011] As a further scheme of the application, the specific performance process of the bag moving work is:

[0012] The position information of the target bag is acquired through the camera, and the corresponding global coordinates are obtained through coordinate system conversion.

[0013] The rotation angle of the servo motor is obtained through the motion control of the mechanical arm.

[0014] The mechanical arm is controlled to complete the grabbing.

[0015] As a further scheme of the application, the specific process of detecting and optimizing the error generated in the bag moving process is:

[0016] Step one: randomly generate a group of servo motor rotation angles in the mechanical arm motion space, and control the mechanical arm to move;

[0017] Step two: after the mechanical arm moves, delay for 2s, and wait for the camera to position the center point of the mechanical arm clamp;

[0018] Step three: repeat steps one and two N times, and finally save N groups of servo motor rotation angles and corresponding bag position coordinates as the input of the parameter identification system;

[0019] Step four: establish the mechanical arm servo motor frequency response control, mechanical arm motion model, and coordinate system conversion model as the solution target of the identification system, and configure the optimization algorithm parameters;

[0020] Step five: use the adaptive parameter adjustment differential algorithm to iteratively update the model parameters, and when the accuracy convergence condition is met or the given number of iterations is reached, terminate the algorithm, and complete the solution of the control system model.

[0021] As a further scheme of the application, the center point position of the mechanical arm clamp refers to the position of the bag.

[0022] As a further scheme of the application, in the repeated execution of steps one and two N times, the size of N is set according to the accuracy requirement.

[0023] As a further scheme of the application, the specific process of using the adaptive parameter adjustment differential algorithm to iteratively update the model parameters is:

[0024] Step (1): initialize the first dynamic solution set, the second dynamic solution set, the data set containing the variance of all dimensions of the control system error, and the data set with large variance, let the dominated set P be an empty set, and set the adaptive optimization state switching rule;

[0025] Step (2): when the fitness function meets the convergence condition or reaches the given number of iterations, output the global optimal solution, and terminate the algorithm; otherwise, immediately execute the dynamic solution set update selection strategy of the adaptive parameter adjustment differential algorithm;

[0026] Step (3): Determine the three optimization states according to the adaptive optimization state switching rules, and jump to steps (4), (5) and (6) respectively;

[0027] Step (4): Use principal component analysis to calculate the data variance of each dimension parameter in the entire dataset, update the state and start the transition mechanism, and jump to step (2);

[0028] Step (5): Fix the values ​​of all parameters in the dataset with smaller variance, enter the transfer mechanism, optimize the components in the dataset with larger variance, and jump to step (2);

[0029] Step (6): Release all parameters in the datasets with smaller variance and the datasets with larger variance, enter the transfer mechanism, and jump to step (2).

[0030] As a further aspect of the present invention: the frequency response control uses the SFRA method to test the frequency characteristics of the servo system.

[0031] Compared with the prior art, the beneficial effects of the present invention are:

[0032] This application can accurately detect and optimize various errors that occur during bag handling. It can not only automatically calibrate normal robotic arms but also effectively model robotic arms with defective components, limiting the disturbances of these defective components to a reasonable range, thereby preventing any impact on the final packaging effect. Specifically, this application establishes a robotic arm servo motor frequency response control model, a robotic arm motion model, and a coordinate system transformation model as the solution objectives for the identification system. It configures and optimizes algorithm parameters, and then uses an adaptive parameter adjustment differential algorithm to iteratively update the model parameters. The algorithm terminates when the accuracy convergence condition is met or a given number of iterations is reached, thus completing the solution for the control system model and effectively improving the accuracy of bag handling. Attached Figure Description

[0033] Figure 1 A flowchart of a film-stretching double-bag transfer method;

[0034] Figure 2 This is a flowchart illustrating the bag-handling process in a double-bag-handling method for film stretching;

[0035] Figure 3 This is a schematic diagram illustrating the detection and optimization of errors generated during the bag-handling process in a double-bag-handling method for film stretching;

[0036] Figure 4 This is a schematic diagram illustrating the use of an adaptive parameter adjustment difference algorithm to iteratively update model parameters in a film stretching double-bag method.

[0037] Figure 5 This is a schematic diagram of frequency response control in a film stretching double-bag method. Detailed Implementation

[0038] 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, and 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.

[0039] As mentioned in the background section of this application, the inventors have discovered through research that although people have developed double-sided film-pulling bag-handling equipment that can alternately handle bags from both sides to improve efficiency, if errors occur during the bag-handling process, the final packaging effect can easily be severely reduced, affecting the accuracy of the work. However, there is no specific method for detecting and optimizing errors in the bag-handling process in the existing technology, which has certain shortcomings.

[0040] To address the aforementioned shortcomings, this application discloses a film-pulling double-bag-handling method that can accurately detect and optimize various errors that occur during the bag-handling process. It can not only automatically calibrate normal robotic arms, but also effectively model robotic arms with defective parts and limit the disturbance of defective parts to a reasonable range, thereby avoiding affecting the final packaging effect.

[0041] The following will describe in detail, with reference to the accompanying drawings, how the solution of this application solves the above-mentioned technical problems.

[0042] Please see Figure 1 This invention provides a double-arm bag-moving method for film stretching, comprising the following steps: determining which robotic arm arrives at its position first, while the other robotic arm waits in a waiting position; the robotic arm that arrives first performs the bag-moving operation, and after completion, the other robotic arm takes over; and detecting and optimizing errors generated during the bag-moving process, including camera positioning errors, servo motor errors, robotic arm deformation errors, and coordinate system transformation errors. This application can accurately detect and optimize various errors occurring during the bag-moving process, thereby preventing the final packaging effect from being affected.

[0043] In this embodiment: as Figure 2As shown, the specific implementation process of the bag moving work is: obtaining target bag position information through a camera, and then obtaining corresponding global coordinates through coordinate system conversion; obtaining servo motor rotation angle through mechanical arm motion control; and controlling the mechanical arm to complete grabbing. The overall optimization purpose of the vision-mechanical arm system is to improve the grabbing accuracy, therefore, the application analyzes the error sources from the bag moving work process, and the coordinate system conversion refers to the process of converting from one coordinate system to another coordinate system. The coordinate system conversion is commonly used in fields such as geophysics, geography, and astronomy, for example: conversion between local rectangular coordinate system and national rectangular coordinate system; conversion between astronomical coordinate system and geographical coordinate system; conversion between spherical coordinate system and plane coordinate system, etc.

[0044] In the embodiment, as Figure 3As shown, the specific process of detecting and optimizing the error generated in the bag carrying work process is: step one: randomly generate a group of servo motor angles in the mechanical arm motion space, control the mechanical arm motion; step two: after the mechanical arm motion is completed, delay for 2s, wait for the camera to position the center point position of the mechanical arm clamp; step three: repeat steps one and two N times, finally save N groups of servo motor angles and corresponding bag position coordinates as the input of the parameter identification system. It should be noted that the parameter identification system refers to the process of determining system parameters through experiment or data analysis. The parameter identification system usually includes data collection, parameter estimation and model verification steps. In parameter identification, the input and output data of the system need to be analyzed and processed to estimate the parameter values of the system. These parameters usually refer to certain physical or statistical parameters, such as damping coefficient, elastic coefficient, heat transfer coefficient, etc.; step four: establish the mechanical arm servo motor frequency response control, mechanical arm motion model, coordinate system conversion model as the solution target of the identification system, and configure the optimization algorithm parameters. It should be noted that the mechanical arm motion model is a mathematical model used to describe the position, velocity, acceleration and other motion parameters of the mechanical arm during motion. A typical mechanical arm is composed of a series of connected joints and links, each joint has one degree of freedom, which can be translation or rotation. For a mechanical arm with n joints, the joint number is from 1 to n, and there are n+1 links, numbered from 0 to n. The research of mechanical arm motion model includes forward kinematics and inverse kinematics, where forward kinematics studies how to drive the mechanical arm to reach the target position, and inverse kinematics studies how to calculate the state and length of each joint muscle according to the pose and position of the end effector. The coordinate system conversion model refers to a mathematical method for transforming coordinates from one system to another system according to different coordinate conversion requirements. The optimization algorithm parameters are the hyperparameters used in the optimization algorithm. Common hyperparameters in optimization algorithms include: the number of classes in clustering algorithms, the step size in gradient descent methods, the coefficient of regularization distribution items, the number of layers in neural networks, kernel functions in support vector machines, etc. The selection of optimization algorithm parameters is generally a combinatorial optimization problem, which is difficult to learn automatically through optimization algorithms; step five: use the adaptive parameter adjustment differential algorithm to iteratively update the model parameters, and when the precision convergence condition is met or the given number of iterations is reached, terminate the algorithm, and complete the solution of the control system model. In addition, the detection and optimization of the error generated in the bag carrying work process mainly includes data collection and model optimization, which is completed by using a camera to assist positioning for data collection.

[0045] In this embodiment: the center point position of the mechanical arm clamp refers to the position of the bag.

[0046] In this embodiment: in the repeated execution of steps one and two N times, the size of N is set by the accuracy requirement. This setting allows workers to collect the corresponding number of angle data sets and coordinate data sets as needed.

[0047] In the embodiment, as shown in Figure 4 The specific process of using the adaptive parameter adjustment differential algorithm to iteratively update the model parameters is as follows: step (1): initialize the first dynamic solution set, the second dynamic solution set, the variance smaller data set containing all dimensions of the control system error, and the variance larger data set, let the dominating set P be an empty set, and set the adaptive optimization state switching rule; step (2): when the fitness function meets the convergence condition or reaches the given number of iterations, output the global optimal solution, and the algorithm terminates; otherwise, immediately execute the dynamic solution set update selection strategy of the adaptive parameter adjustment differential algorithm; step (3): judge the three optimization states according to the adaptive optimization state switching rule, and jump to step (4), step (5), and step (6); step (4): use the principal component analysis method to calculate the data variance of each dimension parameter in the entire data set, update the state and start the transfer mechanism at the same time, and jump to step (2); step (5): fix the values of all parameters in the variance smaller data set, enter the transfer mechanism, and optimize the components in the variance larger data set, and jump to step (2); step (6): release all parameters in the variance smaller data set and the variance larger data set, enter the transfer mechanism, and jump to step (2). The adaptive parameter adjustment differential algorithm has the advantages of being able to adjust the algorithm parameters according to the characteristics of the problem and the performance of the algorithm, improving the robustness and global optimization ability of the algorithm, and the adaptive characteristics of the algorithm can also partially alleviate the impact of parameter adjustment. In the implementation of the adaptive parameter adjustment differential algorithm, multiple tests are needed to obtain the basic characteristics of the problem, and the algorithm parameters (such as the differential weight factor and the crossover probability) are adjusted according to the characteristics of the problem and the performance of the algorithm, so that the algorithm can automatically adjust the parameters when facing complex or changing optimization problems, and improve the search efficiency.

[0048] In the embodiment, as shown in Figure 5 The frequency response control uses the SFRA method to test the frequency characteristics of the servo system. Traditional control system design requires the use of a dynamic signal analyzer for testing. The dynamic signal analyzer generates a test sine sweep signal Id* according to the instructions of the upper computer, the DSP converts the signal collected by A / D into a test reference signal, and simultaneously sends the feedback signal Id calculated by A / D to the dynamic signal analyzer. The dynamic signal analyzer collects the output of the current sensor as the feedback signal of the system, and the dynamic signal analyzer obtains the object characteristics of the PMSM according to the test signal and the feedback signal. After using the SFRA software, the DSP itself integrates the functions of the dynamic signal analyzer, so the dynamic signal analyzer and the A / D and D / A on the DSP processing board can be omitted. The principle of testing the characteristics of the AC servo system based on SFRA is as shown in Figure 5As shown, due to the lack of 4 times D / A and A / D conversion, the noise in the signal Id* and Id is reduced, which is beneficial to the analysis of system characteristics, especially in the high frequency band with low signal-to-noise ratio, improves the identification accuracy of the system, compared with the SFRA method in the power supply, the SFRA in the servo system increases the ABC / dq coordinate transformation, dq / ABC coordinate transformation and SVPWM and several parts. Through the analysis of the SFRA method and the alternating current servo system, the method is innovatively introduced from the power supply test of TI company to the alternating current servo control system. On the one hand, the cost of the dynamic signal analyzer is saved, and on the other hand, the noise introduced by the dynamic signal analyzer is eliminated, which is worth popularizing in the servo system. The SFRA method is carried out in the current loop of the alternating current servo system, and the method is also suitable for the application of speed loop, position loop and other control systems, which has important significance for the object characteristic test in high-precision alternating current servo control system such as mechanical arm.

[0049] The application can accurately detect various errors occurring in the bag moving process and optimize, not only can complete the automatic calibration of normal mechanical arm, but also can effectively model the mechanical arm with defective parts, and limit the disturbance of the defective parts within a reasonable range, and avoid affecting the final packaging effect. Among them, the application establishes the mechanical arm servo motor frequency response control, mechanical arm motion model and coordinate system conversion model as the solving target of the identification system, configures the optimization algorithm parameter, and then uses the adaptive parameter adjustment differential algorithm to iteratively update the model parameter, meets the precision convergence condition or reaches the given iteration number, terminates the algorithm, completes the solution of the control system model, and effectively improves the accuracy of bag moving.

[0050] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these modifications and variations.

[0051] The above is only the preferred embodiment of the present application, but the protection scope of the present application is not limited to this, any skilled person in the art can make equivalent replacement or change according to the technical scheme and the inventive concept of the present application within the technical range disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. A method of draw film double bagging, characterized by, The method comprises the following steps: Determine which side of the robot reaches the position first, and the other side of the robot waits at the waiting position; The robot that reaches the position first performs the bag moving work, and after the bag moving work is completed, the other side of the robot performs the bag moving work; Detect and optimize errors generated in the bag moving work, including camera positioning errors, servo motor errors, mechanical arm deformation errors and coordinate system conversion errors; The specific process of the bag moving work is as follows: Obtain the target bag position information through the camera, and then obtain the corresponding global coordinates through coordinate system conversion; Obtain the servo motor rotation angle through the mechanical arm motion control; Control the mechanical arm to complete the grabbing; The specific process of detecting and optimizing the errors generated in the bag moving work is as follows: Step 1: Randomly generate a group of servo motor rotation angles in the mechanical arm motion space, and control the mechanical arm to move; Step 2: After the mechanical arm motion is completed, delay for 2s, and wait for the camera to position the center point position of the mechanical arm clamp; Step 3: Repeat steps 1 and 2 N times, and finally save N groups of servo motor rotation angles and corresponding bag position coordinates as the input of the parameter identification system; Step 4: Establish the mechanical arm servo motor frequency response control, mechanical arm motion model and coordinate system conversion model as the solving target of the identification system, and configure the optimization algorithm parameters; Step 5: Use the adaptive parameter adjustment differential algorithm to iteratively update the model parameters, and when the precision convergence condition is met or the given number of iterations is reached, terminate the algorithm, and complete the solution of the control system model; The specific process of using the adaptive parameter adjustment differential algorithm to iteratively update the model parameters is as follows: Step (1): Initialize the first dynamic solution set, the second dynamic solution set, the small-variance data set containing all dimensions of the control system error and the large-variance data set, let the dominated set P be an empty set, and set the adaptive optimization state switching rule; Step (2): When the fitness function meets the convergence condition or reaches the given number of iterations, output the global optimal solution, and the algorithm terminates; otherwise, immediately execute the dynamic solution set update selection strategy of the adaptive parameter adjustment differential algorithm; Step (3): Determine the three optimization states according to the adaptive optimization state switching rule, and jump to step (4), step (5) or step (6); Step (4): Calculate the data variance of each dimension parameter in the entire data set using the principal component analysis method, and simultaneously update the state and start the transfer mechanism, and jump to step (2); Step (5): Fix the values of all parameters in the small-variance data set, enter the transfer mechanism, and optimize the components in the large-variance data set, and jump to step (2); Step (6): Release all parameters in the small-variance data set and the large-variance data set, enter the transfer mechanism, and jump to step (2).

2. A method of pulling a film and carrying a bag according to claim 1, wherein The center point position of the mechanical arm clamp refers to the position of the bag.

3. The method of claim 1, wherein, In the repeated execution of steps 1 and 2 N times, the size of N is set by the accuracy requirement.

4. The method of claim 1, wherein, The frequency response control uses the SFRA method to test the frequency characteristics of the servo system.

Images