Mechanical arm control system for spline injection molding

Abstract

The invention relates to a mechanical arm control system for spline injection molding, which is characterized in that mechanical arm working data, including position information of each joint and an end effector of a mechanical arm, acting force of the mechanical arm and appearance images of injection molding spline products, in historical periods are acquired in batches; training a random forest model by taking the position information of the mechanical arm and the acting force of the mechanical arm as input variables and taking the corresponding working link as an output label; the trained random forest model is used for judging the working link of the current mechanical arm; according to the working link of the current mechanical arm, the next working link is determined in combination with the detection result of the machine vision model on the appearance image of the spline product; and finally, a control instruction is generated according to the next working link information, and the motion state of the mechanical arm is controlled. The problem that in the prior art, it is difficult to determine the next working link according to the working data of the mechanical arm, and consequently spline injection molding control is inaccurate is solved.

Description

Technical Field

[0001] The invention belongs to the technical field of injection molding control and relates to a mechanical arm control system for spline injection molding. Background Art

[0002] In the field of injection molding production, traditional manual operation methods are gradually being replaced by robotic arm technology due to their low efficiency, prone to fatigue errors and insufficient precision control, making it difficult to meet the increasing product precision requirements and expanding production scale.

[0003] However, the existing technology has the problem of difficulty in automatically identifying the current working link based on the working data of the robot arm, and judging the next working link in combination with the detection information of the spline product, which leads to the problem of insufficient automation and inaccuracy in the spline injection molding control. Summary of the invention

[0004] The present invention provides a robotic arm control system for spline injection molding, which aims to improve the relevant performance of the robotic arm body unit, drive unit, and sensor unit in the hardware module, optimize the functions of the motion control unit, human-computer interaction unit, data management unit, and other parts in the software module, and add a reliable safety control unit, so that the robotic arm can accurately complete various action tasks such as taking and placing splines, handling, and product testing during the spline injection molding production process, reduce the defective rate, avoid equipment damage and safety accidents, and better adapt to different injection molding tasks and production scenarios.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] The present application provides a robot control system for spline injection molding, including a hardware module and a software module; the hardware module includes a robot body unit, a drive unit and a sensor unit,

[0007] The software module includes a motion control unit, a human-computer interaction unit and a data management unit connected in communication; the data management unit includes a data acquisition subunit, a data analysis subunit and a data storage subunit, the data acquisition subunit is used to collect working data of the sensor unit robot arm during operation, the working data includes position information of each joint and end effector of the robot arm, the robot arm force, and the appearance image of the injection molded spline product; the robot arm force includes the gripping force of the end effector to grab the injection molded spline and the contact force between the robot arm and the surrounding objects;

[0008] The data analysis subunit is used to analyze the working state of the robot arm according to the collected working data, and generate a control instruction according to the working state and send it to the motion control unit to control the motion state of the robot arm.

[0009] Furthermore, the sensor unit includes a position sensor, a force sensor and a vision sensor. The position sensor is used to provide real-time feedback of position information of each joint and the end effector of the robotic arm. The force sensor is used to collect the gripping force of the end effector in grasping the spline and the contact force in the process of interacting with the surrounding environment. The vision sensor is used to obtain the appearance image of the spline product and the position of the injection mold.

[0010] Furthermore, the motion control unit includes a trajectory planning subunit, a speed control subunit and a motion coordination subunit. The trajectory planning subunit is used to plan the motion trajectory of the robot arm joint space or Cartesian space according to the injection molding process requirements; the speed control subunit is used to control the speed and acceleration of the robot arm during the movement; the motion coordination subunit is used to coordinate the movement sequence and time rhythm between each robot arm or the robot arm and the injection molding machine when multiple robot arms work together or the robot arm and the injection molding machine cooperate.

[0011] Furthermore, the robot body unit is used to achieve the transformation of spatial position and posture to complete the actions of picking up, placing and carrying splines during the injection molding process, and the drive unit is used to provide power for the movement of the robot joints; the human-computer interaction unit includes an operation interface subunit and a programming interface subunit, and the operation interface subunit provides a visual operation interface for setting parameters, sending action instructions and displaying the movement status of the robot; the programming interface subunit is used by technicians to program and modify the action flow of the robot.

[0012] Furthermore, analyzing the working state of the robot arm according to the collected working data specifically includes the following steps:

[0013] S1. Batch obtain the robot arm working data of the historical period from the data storage subunit;

[0014] S2, taking the position information of each joint and the end effector of the robotic arm and the force of the robotic arm as input variables and the corresponding working links as output labels, training a random forest model; the working links include a pick-and-place link, a transport link and a collaboration link;

[0015] S3, using the trained random forest model to determine the working link of the current robotic arm;

[0016] S4. According to the current working link of the robot arm and the detection result of the appearance image of the spline product by the machine vision model, the next working link is determined.

[0017] Furthermore, the training of the random forest model comprises the following steps:

[0018] Convert the position information of each joint and end effector of the robot arm into three-dimensional coordinate data;

[0019] Convert the robot arm force data into two-dimensional data of magnitude and direction;

[0020] Randomly extract N transformed input variables and construct a decision tree with the output labels;

[0021] Repeat the process of building decision trees until all input variables are traversed to generate a random forest;

[0022] According to the prediction ability of each decision tree for the output label, determine the key input variables with significant judgment ability for the work link;

[0023] Optimize the goodness of fit of the random forest based on key input variables.

[0024] Furthermore, the machine vision model is configured as a convolutional neural network model, including the following construction steps:

[0025] Collect samples of spline product appearance image data from historical periods;

[0026] The images are marked separately according to whether the splines are qualified;

[0027] Construct a convolutional network structure to extract image features from spline product appearance images;

[0028] The convolutional neural network model is trained with image features as independent variables and the corresponding annotation information as dependent variables.

[0029] Furthermore, the detection result of the appearance image of the spline product in combination with the machine vision model is used to determine the next working step, which is specifically:

[0030] When the test result of the spline product is judged to be qualified, the next work link is carried out according to the normal working order;

[0031] When the test result of the sample product is determined to be unqualified, the next step is to recycle the sample product.

[0032] Furthermore, the step of generating a control instruction according to the working state and sending it to a motion control unit to control the motion state of the robot arm specifically includes the following steps:

[0033] Get the next work link information;

[0034] Generate control instructions based on the information, the control instructions including the motion trajectory, speed, acceleration of the robot arm, and the action sequence and time rhythm of each robot arm or between the robot arm and the injection molding machine;

[0035] Send control instructions to the motion control unit;

[0036] The motion control unit controls the motion state of the robot arm according to the command instructions.

[0037] Furthermore, the software module also includes a safety control unit for monitoring in real time whether the robot arm body collides with the injection molding machine and the surrounding environment, and stopping or adjusting the action in time once a collision is detected.

[0038] Beneficial effects of the present invention:

[0039] (1) By batch acquiring the working data of the robot arm in the historical period, including the position information of each joint and end effector of the robot arm, the force of the robot arm, and the appearance image of the injection molded spline product; using the position information and force of the robot arm as input variables and the corresponding working link as output label, the random forest model is trained; the trained random forest model is used to judge the working link of the current robot arm; according to the working link of the current robot arm, combined with the detection result of the appearance image of the spline product by the machine vision model, the next working link is determined; finally, the control instruction is generated according to the information of the next working link to control the motion state of the robot arm. The present invention solves the problem that it is difficult to determine the next working link according to the working data of the robot arm in the prior art, resulting in inaccurate control of spline injection molding.

[0040] (2) By constructing a random forest model, the relationship between work data and work links is quantified, the accuracy of judgment on the current work link is improved, and a theoretical basis is provided for determining the next work link.

[0041] (3) The machine vision model is used to detect whether the current spline product is qualified, and based on the test results, it is determined whether the next work link should be carried out in the normal working order or the spline product should be recycled. The combination of the two models improves the ability to identify the working status of the robot arm, thereby improving the efficiency and accuracy of the robot arm injection molding. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0043] Figure 1 The structure diagram of a robot arm control system for spline injection molding in the present invention.

[0044] Figure 2 This is a flow chart of analyzing the working status of a robot arm based on collected working data in one embodiment of the present invention.

[0045] Figure 3 A flowchart of a random forest model constructed in one embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation mode, structure, characteristics and effects of the present invention are described in detail below in combination with the accompanying drawings and preferred embodiments.

[0047] See also Figure 1-Figure 3 The present application provides a robot control system for spline injection molding, including a hardware module and a software module; the hardware module includes a robot body unit, a drive unit and a sensor unit,

[0048] The software module includes a motion control unit, a human-computer interaction unit and a data management unit connected in communication; the data management unit includes a data acquisition subunit, a data analysis subunit and a data storage subunit, the data acquisition subunit is used to collect working data of the sensor unit robot arm during operation, the working data includes position information of each joint and end effector of the robot arm, the robot arm force, and the appearance image of the injection molded spline product; the robot arm force includes the gripping force of the end effector to grab the injection molded spline and the contact force between the robot arm and the surrounding objects;

[0049] The data analysis subunit is used to analyze the working status of the robot arm based on the collected working data.

[0050] A control instruction is generated according to the working state and sent to the motion control unit to control the motion state of the robot arm.

[0051] Furthermore, analyzing the working state of the robot arm according to the collected working data specifically includes the following steps:

[0052] Step S1: batch-acquire the robot arm working data of the historical period from the data storage subunit;

[0053] As a module for long-term storage of various data of the robot arm operation, the data storage subunit accumulates detailed records of the robot arm's working process in different time periods and different injection molding production tasks. These data cover the position information of each joint and end effector of the robot arm, the force of the robot arm, and the appearance image of the injection molding spline product, forming the basic data pool for subsequent analysis.

[0054] By acquiring in batches, we can collect enough representative data samples at one time. This helps to more comprehensively reflect the performance of the robot under various actual working conditions, and avoid inaccurate subsequent analysis results due to too little or one-sided data. For example, in different batches of injection molding production, the specifications and materials of the splines may be different, and the operation of the robot arm will also be different. Batch acquisition of data can include all these diverse situations and provide sufficient materials for accurate analysis.

[0055] Data screening and sorting: After acquiring the data, certain data screening and sorting work is performed according to specific needs. For example, some obviously abnormal data points (which may be caused by sensor failure, etc.) are removed, or the data is classified according to time sequence, work task type, etc., so that the data can be used more efficiently for model training later.

[0056] Step S2: taking the position information of each joint and the end effector of the robotic arm and the force of the robotic arm as input variables and the corresponding working links as output labels, training a random forest model;

[0057] The position information of each joint and end effector of the robot arm directly reflects the posture and specific position of the robot arm in space, and is one of the key factors in judging its working process. For example, in the pick-and-place process, the end effector will move to a specific position to grab or place the spline; in the handling process, its position trajectory will show the characteristics of transfer from one fixed point to another, so the position information can help distinguish different working stages.

[0058] Relevance of robot arm force: The robot arm force includes the gripping force of the end effector to grasp the injection molding spline and the contact force with the surrounding objects. In the pick-and-place process, the gripping force will have a specific change pattern. When grasping, the gripping force gradually increases until the spline is stably held, and the gripping force decreases to zero when placing. In the handling process, the force is relatively stable. In the collaborative process, if it cooperates with other robot arms or equipment, the contact force situation will be different. Therefore, the force data can help to define the work process more accurately.

[0059] Introduction to Random Forest Model:

[0060] Principle overview: Random forest is an ensemble learning algorithm based on multiple decision trees. It extracts multiple sub-datasets from the original training data set with replacement through bootstrap sampling, and then builds a decision tree based on each sub-dataset. In the process of building a decision tree, for each node, it randomly selects a part of the features from all the features (here, the position information and force-related features) to find the best split point, so as to reduce the variance of the model and improve the generalization ability.

[0061] Training process: The selected position information and force data are used as input variables, and the corresponding work links such as pick-and-place, handling and collaboration are used as output labels. A large number of historical data samples are input into the random forest algorithm for training. Each decision tree will learn a mapping relationship between input variables and output labels. As the number of decision trees increases, the entire random forest model can integrate the judgment results of each tree and give a more accurate and stable work link judgment for the newly input data.

[0062] Model evaluation and optimization:

[0063] During the training process, some evaluation indicators (such as accuracy, precision, recall, F1 value, etc.) are usually used to evaluate the model on the reserved validation data set to check whether the model is overfitting or underfitting. If the model performance is found to be unsatisfactory, the model may be optimized by adjusting the model parameters (such as the number of decision trees, the maximum depth of each tree, etc.), increasing the amount of data, or performing feature engineering (further processing, screening or combination of input variables, etc.) to ensure that it can accurately determine the corresponding working link based on the input real-time position and force information.

[0064] Furthermore, the evaluation index is calculated as follows:

[0065]

[0066] In the formula, ACC represents accuracy; PPV represents precision; TRP represents recall; F1 represents F1 value; TP represents true positive examples; TN represents true negative examples; FP represents false positive examples; and FN represents false negative examples.

[0067] Step S3: Using the trained random forest model to determine the working link of the current robotic arm;

[0068] During the actual operation of the robot arm, the sensor unit will collect the current position information of each joint and end effector of the robot arm and the force information of the robot arm in real time. These real-time data will be continuously transmitted to the data analysis subunit as the input data of the model. The real-time collected position information and force data are sorted in the format required by the model and input into the trained random forest model. Each decision tree in the model will perform calculations and judgments based on these input data, and then combine the output results of all decision trees (using voting) to finally determine whether the current working link of the robot arm is the pick-and-place link, the handling link, or the collaborative link. For example, if most decision trees judge that the current situation meets the characteristics of the pick-and-place link, then the model outputs the result that the current robot arm is in the pick-and-place link.

[0069] The results of the current working link determined by the model will be fed back to the subsequent processes of the data analysis subunit and other related system modules (such as the motion control unit, etc.) in a timely manner. This result can help the entire system understand the current working status of the robot arm so as to make corresponding decisions and adjustments. For example, the motion control unit can control the movement of the robot arm according to the pre-set motion parameters of the pick-and-place link according to the current pick-and-place link.

[0070] Step S4: Determine the next working link according to the current working link of the robot arm and the detection result of the appearance image of the spline product by the machine vision model.

[0071] Machine vision model detection of appearance images:

[0072] Image acquisition: The visual sensor will collect the appearance images of the injection molded spline products in real time. These images contain visual information such as the shape, size, surface quality, etc. of the spline. The collected images will be transmitted to the machine vision model for processing. The machine vision model uses image processing algorithms (such as image recognition, feature extraction, target detection and other technologies) to analyze the appearance images. For example, it detects whether there are appearance defects (such as scratches, holes, etc.) on the spline, and determines whether the placement of the spline in the current position meets the requirements (whether the angle is correct, whether it is aligned with other parts, etc.). Through these test results, the actual situation of the current spline product can be understood, providing a visual basis for determining the next work link.

[0073] Furthermore, the sensor unit includes a position sensor, a force sensor and a vision sensor. The position sensor is used to provide real-time feedback of position information of each joint and the end effector of the robotic arm. The force sensor is used to collect the gripping force of the end effector in grasping the spline and the contact force in the process of interacting with the surrounding environment. The vision sensor is used to obtain the appearance image of the spline product and the position of the injection mold.

[0074] Furthermore, the motion control unit includes a trajectory planning subunit, a speed control subunit and a motion coordination subunit. The trajectory planning subunit is used to plan the motion trajectory of the robot arm joint space or Cartesian space according to the injection molding process requirements; the speed control subunit is used to control the speed and acceleration of the robot arm during the movement; the motion coordination subunit is used to coordinate the movement sequence and time rhythm between each robot arm or the robot arm and the injection molding machine when multiple robot arms work together or the robot arm and the injection molding machine cooperate.

[0075] Furthermore, the robot body unit is used to achieve the transformation of spatial position and posture to complete the actions of picking up, placing and carrying splines during the injection molding process, and the drive unit is used to provide power for the movement of the robot joints; the human-computer interaction unit includes an operation interface subunit and a programming interface subunit, and the operation interface subunit provides a visual operation interface for setting parameters, sending action instructions and displaying the movement status of the robot; the programming interface subunit is used by technicians to program and modify the action flow of the robot.

[0076] Furthermore, the training of the random forest model comprises the following steps:

[0077] (1) Convert the position information of each joint and end effector of the robot arm into three-dimensional coordinate data;

[0078] In the actual operation scenario of the robot arm, its joints and end effectors have exact positions in space, and using three-dimensional coordinates to represent these positions is the most intuitive and convenient way for subsequent analysis and processing. The original position information is often based on various data forms of sensor feedback, which may be complicated and do not directly reflect the spatial coordinate situation. In order to make the data more in line with the requirements of model training, conversion operations are required.

[0079] Specifically, the conversion work is carried out based on the structural parameters of the robot itself (such as the rotation angle range of each joint, the length of the connecting rod, etc.) and the pre-built robot kinematic model. Through a series of geometric calculations and coordinate transformations, those relative and scattered sensor readings are converted into unified coordinate values ​​in a specific three-dimensional coordinate system (such as the common Cartesian coordinate system), that is, x, y, and z coordinates. For example, for a robot with multiple rotating joints, key information such as the current rotation angle of each joint and the connection length between the connecting rods are known. Using the trigonometric function relationship and the rules of coordinate translation and rotation, the specific three-dimensional coordinate position of the end effector in the entire space is gradually calculated. In this way, the posture and location of the robot in space can be clearly and accurately presented, laying the foundation for the subsequent model to use these position data to judge the working links.

[0080] (2) Convert the robot arm force data into two-dimensional data of magnitude and direction;

[0081] As a vector, the magnitude and direction of the robot force carry key working status information. In order to make the force data more effectively used by the random forest model, it needs to be refined and converted into two-dimensional data in the form of magnitude and direction.

[0082] In actual operation, the original force data collected by the force sensor (which may be force components in multiple directions, or in the form of complex vector representation in a specific coordinate system) will be processed using corresponding mathematical methods. For example, the Pythagorean theorem is used to calculate the modulus of the force vector to determine its size; then, through trigonometric functions (such as the inverse tangent function, etc.), combined with the direction setting of the coordinate system, the angle between the force direction and the positive direction of a reference coordinate axis is calculated, and then its direction angle is clarified, and finally the original force data is converted into a concise two-dimensional data form containing key information. This form can not only fully retain the information that is valuable for judging the work link, but also meet the requirements of the model input variables, which helps the subsequent decision tree to more accurately learn and judge the characteristic laws of the forces in different work links.

[0083] (3) Randomly extract N transformed input variables and construct a decision tree with the output labels;

[0084] The unique construction method of the random forest model aims to improve the generalization ability and stability of the overall model by increasing the diversity between decision trees and avoiding overfitting problems. Among them, the random extraction of input variables plays a key role.

[0085] After completing the transformation of position information and force data, we have a series of input variables that can be used by the model. At this time, N variables are randomly extracted from these transformed input variable sets each time, and combined with the corresponding work links that have been clearly marked (such as pick-up and placement links, handling links, and collaboration links as output labels), a decision tree is constructed. In the process of building a decision tree, based on the N extracted input variables, according to the intrinsic relationship between them and the output labels, by constantly looking for the best split nodes and division rules (for example, different branches are divided according to the value range of a certain variable, so that the purity of different categories of output labels under the branches is maximized), the decision tree gradually learns how to accurately judge the corresponding work links according to the different values ​​of the input variables. Each decision tree is gradually grown and constructed based on the randomly extracted variables and the corresponding output labels.

[0086] (4) Repeat the process of building decision trees until all input variables are traversed to generate a random forest;

[0087] The construction of random forest does not rely on a single decision tree to complete the judgment task, but requires the cooperation of many decision trees. Therefore, the above process of building decision trees should be repeated many times until all transformed input variables have the opportunity to be extracted and participate in the construction of decision trees.

[0088] In other words, by continuously randomly extracting different combinations of input variables, we build one decision tree after another with different feature learning capabilities. As the number of decisions gradually increases, many decision trees are combined to form a random forest. When judging the current working link of the robot arm in the future, the random forest can integrate the judgment results of each decision tree (for example, determine the final output category through voting, etc.), give full play to the advantages of each decision tree, and thus more accurately and stably judge the current working link of the robot arm. Compared with a single decision tree, this integrated method can better cope with complex and changeable actual working conditions and improve the overall judgment accuracy.

[0089] (5) Based on the predictive power of each decision tree for the output label, determine the key input variables that have significant judgment power for the work link;

[0090] After the random forest is constructed, although each decision tree participates in the learning process of judging the work link, in fact, different input variables play different roles in accurately judging the work link in different decision trees.

[0091] In order to further explore and utilize the most critical and influential information for judging work links, it is necessary to analyze and evaluate each decision tree. Specifically, the accuracy of each decision tree's prediction of the output label (i.e., work link) based on the input variables will be examined, which can be measured by some evaluation indicators (such as accuracy, recall, etc.). In this process, the contribution of different input variables to improving prediction accuracy in multiple decision trees is statistically analyzed. Those input variables that can frequently and significantly help decision trees to more accurately judge work links will be identified as key input variables. These key input variables often contain deep-level characteristic laws that are closely related to work links, and are an important basis for subsequent optimization of the random forest model and improving its overall judgment ability.

[0092] (6) Optimize the goodness of fit of random forest based on key input variables.

[0093] After determining the key input variables, the random forest model can be optimized and adjusted based on these variables to improve its goodness of fit, that is, to enable the model to better judge the working links based on the input variables.

[0094] There are many specific optimization methods. For example, when constructing a decision tree or adjusting an existing decision tree, the probability of extracting key input variables can be increased, so that the decision tree can learn and divide more based on these important variables, and enhance the accuracy of its judgment on the work links. Or readjust some parameter settings of the decision tree (such as node splitting conditions, tree depth, etc.) according to the key input variables, so that the decision tree structure is more adapted to the characteristic laws reflected by these key variables. Through such a series of optimization operations, the entire random forest model can be more accurate and stable when using input variables to judge the working links of the robot arm, and better adapt to different actual working scenarios, thereby providing a reliable basis for the robot arm control system to more reasonably plan subsequent work links.

[0095] Furthermore, the machine vision model is configured as a convolutional neural network model, including the following construction steps:

[0096] (1) Collecting spline product appearance image data samples from historical periods;

[0097] In the injection molding process, a large number of spline product appearance images produced in different batches and under different working conditions will be accumulated over time. These image data samples from historical periods constitute the basic material for building machine vision models. Collecting these images helps to cover various possible spline appearance situations, such as different shapes, colors, surface textures, and possible defect types, so that the model can learn a rich variety of spline product appearance features to cope with complex and changing situations in actual production.

[0098] For example, the collected images may include complete and defect-free splines with a smooth appearance and standard size; there may also be images of splines with various defects such as scratches, holes, and deformations; there are also images acquired under different lighting conditions and shooting angles, which fully reflect the appearance of spline products in actual production scenarios and provide sufficient data support for the accurate training of subsequent models.

[0099] (2) Label the images according to whether the splines are qualified;

[0100] After collecting the appearance images of the spline products, each image needs to be accurately annotated. The basis for annotation is the key standard of whether the spline is qualified. A qualified spline means that its appearance meets the production process requirements and has no defects that affect product quality and subsequent use; while an unqualified spline has problems such as surface defects and shape deviations.

[0101] Through manual or semi-automatic annotation tools, carefully observe the appearance of the splines on each image, and clearly mark the image as "qualified" or "unqualified". For example, if the spline surface in the image is flat, without obvious scratches and holes, and the size meets the design requirements, it is marked as "qualified"; on the contrary, if the spline surface has visible scratches, uneven color, or distorted shape, etc., which do not meet the requirements, it is marked as "unqualified". This clear annotation provides a clear target output for subsequent model training, allowing the model to learn what kind of image features correspond to qualified splines and what kind correspond to unqualified splines, thereby establishing a relationship between image features and whether the spline is qualified or not.

[0102] (3) Construct a convolutional network structure to extract image features from the spline product appearance image;

[0103] Convolutional Neural Network (CNN) has powerful image feature extraction capabilities, and its core lies in the unique convolutional network structure. When building a convolutional network structure, it usually includes multiple convolutional layers, pooling layers, and fully connected layers.

[0104] Convolutional layer: The convolutional layer performs convolution operations by sliding convolution kernels (also called filters) on the image, which can automatically extract local features in the image, such as edges, textures, lines, etc. Different convolution kernels can capture different types of features, and multiple convolution kernels can be operated in parallel to extract a variety of rich local feature information. For example, some convolution kernels can detect the straight line texture of the spline surface, while others can highlight features such as edge contours.

[0105] Pooling layer: A pooling layer is often set after the convolution layer. Its main function is to reduce the dimension of the feature map obtained by convolution, reduce the amount of data while retaining key feature information to prevent overfitting. Common pooling methods include maximum pooling and average pooling. For example, through maximum pooling, the maximum value of a local area in the feature map is selected as the representative value of the area, which can compress the data scale without losing important features and improve computing efficiency.

[0106] Fully connected layer: After feature extraction and dimensionality reduction through multiple convolutional layers and pooling layers, all the previously extracted features are finally integrated and mapped through the fully connected layer, and converted into a feature representation suitable for subsequent classification tasks (determining whether the spline is qualified) so as to establish a connection with the annotation information for model training.

[0107] Through such a convolutional network composed of a multi-layer structure, the collected spline product appearance images are processed in sequence, and key information that can effectively characterize the spline appearance characteristics is gradually extracted, laying the foundation for accurately judging whether the spline is qualified.

[0108] (4) Using image features as independent variables and corresponding annotation information as dependent variables, the convolutional neural network model is trained.

[0109] After using the convolutional network structure to extract the features of the spline product appearance image, these image features are used as independent variables, and the information previously labeled based on whether the spline is qualified is used as the dependent variable to train the convolutional neural network model.

[0110] During the training process, the model will try to predict the corresponding annotation information (i.e., whether the spline is qualified) based on the input image features, and then calculate the loss value (such as the commonly used cross entropy loss, etc.) by comparing it with the actual annotation. The loss value reflects the degree of difference between the model's prediction result and the actual situation. Then, based on this loss value, the optimization algorithm (such as stochastic gradient descent, etc.) is used to back-propagate and adjust the model's parameters (such as the weight and bias of the convolution kernel), so that the model can reduce the loss value in the next prediction, that is, make the model's prediction result closer to the actual annotation situation.

[0111] Through multiple iterations of training and continuous adjustment of model parameters, the model gradually learns the inherent rules between image features and whether the spline is qualified or not, and finally reaches the goal of being able to accurately judge whether the input spline product appearance image is qualified or not, thereby providing a strong visual inspection basis for the robot control system to determine the next working link in combination with the current robot working link.

[0112] Furthermore, the detection result of the appearance image of the spline product in combination with the machine vision model is used to determine the next working step, which is specifically:

[0113] (1) When the test result of the sample product is determined to be qualified, the next work link shall be carried out in the normal working order;

[0114] In the spline injection molding production process involving robotic arms, each work link is closely linked and has a set sequence. For example, the first step is the pick-and-place link, followed by the transportation link, and then it may be coordinated with other equipment for subsequent processing. These links are carried out in an orderly manner to ensure that the entire injection molding production process is completed efficiently and with high quality.

[0115] When the machine vision model detects the appearance image of the spline product, if the test result is determined to be qualified, it means that the appearance of the current spline product meets the production process requirements and there are no defects that affect subsequent use or processing. At this time, the next work link of the robot arm can be carried out in sequence according to the pre-set normal working order. For example, if the current robot arm is in the handling link and the spline product has passed the inspection, then it will enter the pick-and-place link of placing the spline at the specified position according to the process, or cooperate with other robots and equipment according to the production arrangement, etc., to ensure that the entire injection molding production task continues in an orderly manner, and to ensure production efficiency and the final quality of the product.

[0116] (2) When the test result of the sample product is determined to be unqualified, the next step is to recycle the sample product.

[0117] However, in the actual production process, it is inevitable that the appearance of the spline products does not meet the requirements due to various reasons. At this time, the machine vision model will determine that the test result is unqualified. Unqualified spline products may have appearance defects such as surface scratches, holes, irregular shapes, color deviations, etc. If they continue to participate in subsequent production links, it may not only affect the quality of the final product, but also cause damage to other equipment or cause the entire production process to be disrupted.

[0118] Therefore, when the test results show that the spline product is unqualified, in order to avoid these adverse effects, the next work link of the robot arm is set to recycle the spline product. The robot arm will move the unqualified spline products to the designated recycling area according to the corresponding recycling path planning and operation instructions, so as to carry out centralized processing, cause analysis or attempt to repair these defective products. Such an arrangement helps to isolate unqualified products in time and maintain the normal operation of other links on the production line. At the same time, it is also convenient for enterprises to count and trace quality problems that occur in the production process, so as to take corresponding improvement measures to optimize the injection molding production process, reduce the defective rate, and improve the overall production efficiency.

[0119] Furthermore, the step of generating a control instruction according to the working state and sending it to a motion control unit to control the motion state of the robot arm specifically includes the following steps:

[0120] (1) Obtain information about the next work step;

[0121] In the entire robotic arm control system for spline injection molding, after a series of data collection, analysis, and related model judgment operations, the relevant information of the next working link of the robotic arm has been determined. This information clarifies the specific task that the robotic arm will perform next, such as taking and placing splines, moving splines to a specific location, or coordinating with other robotic arms or injection molding machines to complete a certain processing action. It is the key basis for the subsequent generation of precise control instructions. Different working links correspond to different robotic arm action requirements and operation details. Only by clearly knowing what the next working link is, can the subsequent movement state of the robotic arm be reasonably planned.

[0122] (2) generating control instructions based on the information, wherein the control instructions include the motion trajectory, speed, acceleration of the robot arm, and the action sequence and time rhythm between each robot arm or between the robot arm and the injection molding machine;

[0123] Based on the information of the next work link, the corresponding control instructions are generated. This control instruction covers multiple key elements and aims to accurately control the motion state of the robot arm in all directions so that it can successfully complete the tasks of the corresponding work link.

[0124] Motion trajectory planning: Depending on the work process, the end effector of the robot arm needs to move along a specific path in three-dimensional space. For example, in the pick-and-place process, it is necessary to plan the optimal motion trajectory from the current position to the position of the spline (when grasping) or from the current position of the spline to the target placement position (when placing). This may involve using mathematical methods such as spline curves to make the trajectory smoother, avoid abrupt changes in the movement of the robot arm, improve the stability and efficiency of the movement, and ensure that the target point can be accurately reached for operation.

[0125] Speed ​​and acceleration control: It is also extremely important to properly set the speed and acceleration parameters of the robot during movement. Different work links have different requirements for speed and acceleration. For example, in the handling link, if the spline is relatively fragile or the placement accuracy is extremely high, a lower speed and a relatively stable acceleration may be required to ensure that the spline will not shake, fall, or deviate from the placement position due to inertia and other factors during the handling process; and in some scenarios where time efficiency is high and the spline is relatively stable, appropriately increasing the speed and reasonably adjusting the acceleration can speed up the production rhythm.

[0126] Coordination of action sequence and time rhythm (for multiple robotic arms or cooperation with injection molding machines): When there are multiple robotic arms working together or the robotic arms need to cooperate with the actions of the injection molding machine, the control instructions must also clarify the action sequence and time rhythm between the robotic arms or between the robotic arms and the injection molding machine. For example, after the injection molding is completed, a robotic arm needs to wait for the injection molding machine to open the mold before entering the mold area to pick up and place the sample. Here, it is necessary to accurately specify the time node for the robotic arm to start the action and the order of each related action to ensure that there will be no interference and conflict between each other, and to ensure that the entire injection molding production process is seamlessly connected and runs efficiently.

[0127] (3) Sending control instructions to the motion control unit;

[0128] The generated control instructions need to be accurately transmitted to the motion control unit. As the part of the software module that is specifically responsible for controlling the actual movement of the robot arm, the motion control unit drives the drive unit in the hardware module based on the received instructions, and then drives the robot body unit to make corresponding actions. This issuance process needs to ensure the integrity and accuracy of the instruction data. Through a stable communication link within the system (such as wired or wireless communication, etc., depending on the system design), the control instructions containing detailed information such as motion trajectory, speed, acceleration, action sequence and time rhythm are sent to the motion control unit to prepare for the actual motion control of the robot arm.

[0129] (4) The motion control unit controls the motion state of the robot arm according to the command instructions.

[0130] After receiving the control command, the motion control unit begins to play its core role and accurately controls the motion state of the robot. It will decompose the trajectory into specific parameters such as the angle change or displacement change of each joint according to the motion trajectory information specified in the command through complex algorithms, and then command the drive unit to provide corresponding power to each joint of the robot arm, so that the joints move according to the predetermined trajectory, speed, acceleration and other requirements, thereby driving the entire robot body to achieve the desired spatial position and posture change.

[0131] In the case of multi-robot coordination or cooperation with injection molding machines, the motion control unit will strictly follow the action sequence and time rhythm set in the instructions to coordinate the actions between different robots and between the robots and the injection molding machine to ensure that they can work together in an orderly manner. For example, the actions of different robots are started in sequence at the specified time interval, or the robot action is paused at a specific time point to wait for the injection molding machine to complete a certain process before continuing. Through such precise control, the entire spline injection molding production process can be carried out efficiently, stably and accurately, improving production efficiency and product quality.

[0132] Furthermore, the software module also includes a safety control unit for monitoring in real time whether the robot arm body collides with the injection molding machine and the surrounding environment, and stopping or adjusting the action in time once a collision is detected.

[0133] In the production scenario of spline injection molding, the robot arm usually needs to work in a relatively complex environment. There are not only large equipment such as injection molding machines around it, but also various tooling fixtures, material storage areas, and other auxiliary production facilities. When the robot arm is taking and placing splines, carrying, and cooperating with other equipment, its range of motion is large and its movements are frequent. It is easy to have the risk of collision with the injection molding machine or the surrounding environment. Once a collision occurs, it may not only cause structural damage to the robot arm itself, affecting its subsequent normal use, but also damage key production equipment such as the injection molding machine, thereby causing a series of problems such as production interruption and increased maintenance costs, and may even endanger the safety of the operator. Therefore, the safety control unit plays a vital role in the entire robot arm control system. It is committed to real-time monitoring and preventing possible collisions to ensure that the entire production process continues safely and stably.

[0134] The safety control unit monitors in real time whether the robot body collides with the injection molding machine and the surrounding environment. It mainly relies on the coordinated work of multiple parts in the system, specifically:

[0135] Sensor data utilization: The sensor unit in the hardware module plays a key role in data collection. For example, the position sensor can provide real-time feedback on the precise position information of each joint and end effector of the robot arm. Through these position data, the safety control unit can grasp the specific posture and position of the robot arm in space, so as to determine whether it is approaching the injection molding machine or other objects in the surrounding environment, and whether there is a potential risk of collision. Force sensors are equally important. When the robot arm approaches the surrounding objects to a certain extent, changes in contact force may occur. Force sensors can capture abnormal changes in force in a timely manner, providing a strong auxiliary judgment basis for collision monitoring.

[0136] Software algorithm coordination: The safety control unit itself is equipped with a special software algorithm for real-time analysis and processing of the data collected by the sensor. These algorithms will build a corresponding collision detection model based on the kinematic model of the robot arm, the range of the workspace, and the pre-set safety area of ​​the injection molding machine and the surrounding environment. By constantly comparing the real-time position, force and other data of the robot arm with the safety threshold and spatial range set in the collision detection model, it can accurately determine whether the robot arm is about to collide with the injection molding machine or surrounding objects. For example, based on the current motion trajectory and speed of the robot arm, the algorithm can predict whether it will enter a spatial area defined as dangerous (close to the key parts of the injection molding machine or the collision restricted area of ​​surrounding objects, etc.) in the next very short time, so as to achieve the functions of early warning and real-time monitoring.

[0137] Once the safety control unit detects that there is a risk of collision or a collision has occurred between the robot body and the injection molding machine or the surrounding environment, it will quickly take appropriate measures to avoid more serious consequences:

[0138] Stop the action in time: The most direct and effective way is to immediately stop all the current actions of the robot arm, cut off the power supply of the drive unit to the robot arm joints, keep the robot arm in the current position and stop moving, prevent the collision from further aggravating, and reduce the damage to the robot arm and peripheral equipment. For example, if it is detected that the end effector of the robot arm is about to hit the mold structure of the injection molding machine, the safety control unit will immediately issue a stop command to stop the robot arm from moving at the moment before the collision occurs, avoiding damage to the mold and the robot arm.

[0139] Adjusting the action: In some cases, depending on the specific circumstances of the collision and the system's preset strategy, the safety control unit can also make appropriate adjustments to the robot's action so that it can avoid the collision object and return to a safe state of motion. This may involve replanning the robot's motion trajectory, changing its direction of motion, or adjusting its speed. For example, when it is detected that the robot is close to the surrounding material storage area and has a tendency to collide during the process of carrying the spline, the safety control unit can quickly calculate a new motion trajectory that avoids the area, and instruct the motion control unit to control the robot to continue moving according to the new trajectory, ensuring that the production process can continue safely with minimal interruption.

[0140] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A robotic arm control system for spline injection molding, characterized in that: It includes a hardware module and a software module; the software module includes a motion control unit, a human-computer interaction unit and a data management unit that are communicatively connected; the data management unit includes a data acquisition subunit, a data analysis subunit and a data storage subunit, and the data analysis subunit is used to analyze the working state of the robot arm according to the collected working data, and generate a control instruction according to the working state and send it to the motion control unit to control the motion state of the robot arm; Analyzing the working state of the robot arm according to the collected working data specifically includes the following steps: S1. Batch obtain the robot arm working data of the historical period from the data storage subunit; S2, taking the position information of each joint and the end effector of the robotic arm and the force of the robotic arm as input variables and the corresponding working links as output labels, training a random forest model; the working links include a pick-and-place link, a transport link and a collaboration link; S3, using the trained random forest model to determine the working link of the current robotic arm; S4. According to the current working link of the robot arm and the detection result of the appearance image of the spline product by the machine vision model, the next working link is determined.

2. A robotic arm control system for spline injection molding according to claim 1, characterized in that: The training of the random forest model comprises the following steps: Convert the position information of each joint and end effector of the robot arm into three-dimensional coordinate data; Convert the robot arm force data into two-dimensional data of magnitude and direction; Randomly extract N transformed input variables and construct a decision tree with the output labels; Repeat the process of building decision trees until all input variables are traversed to generate a random forest; According to the prediction ability of each decision tree for the output label, determine the key input variables with significant judgment ability for the work link; Optimize the goodness of fit of the random forest based on key input variables.

3. A robot arm control system for spline injection molding according to claim 1, characterized in that: The machine vision model is configured as a convolutional neural network model, including the following construction steps: Collect samples of spline product appearance image data from historical periods; The images are marked separately according to whether the splines are qualified; Construct a convolutional network structure to extract image features from spline product appearance images; The convolutional neural network model is trained with image features as independent variables and the corresponding annotation information as dependent variables.

4. A robot arm control system for spline injection molding according to claim 1, characterized in that: The detection result of the appearance image of the spline product combined with the machine vision model is used to determine the next work link, which is specifically: When the test result of the spline product is judged to be qualified, the next work link is carried out according to the normal working order; When the test result of the sample product is determined to be unqualified, the next step is to recycle the sample product.

5. A robot arm control system for spline injection molding according to claim 1, characterized in that: The step of generating a control instruction according to the working state and sending it to a motion control unit to control the motion state of the robot arm specifically includes the following steps: Get the next work link information; Generate control instructions based on the information, the control instructions including the motion trajectory, speed, acceleration of the robot arm, and the action sequence and time rhythm of each robot arm or between the robot arm and the injection molding machine; Send control instructions to the motion control unit; The motion control unit controls the motion state of the robot arm according to the command instructions.

6. A robot arm control system for spline injection molding according to claim 1, characterized in that: The software module also includes a safety control unit for real-time monitoring of whether the robot arm body collides with the injection molding machine and the surrounding environment, and stopping or adjusting the action in time once a collision is detected.

7. A robot arm control system for spline injection molding according to claim 1, characterized in that: The sensor unit includes a position sensor, a force sensor and a vision sensor. The position sensor is used to provide real-time feedback on the position information of each joint and the end effector of the robotic arm. The force sensor is used to collect the gripping force of the end effector when grasping the spline and the contact force during the interaction with the surrounding environment. The vision sensor is used to obtain the appearance image of the spline product and the position of the injection mold.

8. A robot arm control system for spline injection molding according to claim 1, characterized in that: The motion control unit includes a trajectory planning subunit, a speed control subunit and a motion coordination subunit. The trajectory planning subunit is used to plan the motion trajectory of the robot arm joint space or Cartesian space according to the injection molding process requirements; the speed control subunit is used to control the speed and acceleration of the robot arm during the movement process; the motion coordination subunit is used to coordinate the movement sequence and time rhythm between each robot arm or the robot arm and the injection molding machine when multiple robot arms work together or the robot arm and the injection molding machine cooperate with each other.

9. A robot arm control system for spline injection molding according to claim 1, characterized in that: The robot body unit is used to achieve the transformation of spatial position and posture to complete the action of picking up and carrying the spline during the injection molding process, and the drive unit is used to provide power for the movement of the robot joint; the human-computer interaction unit includes an operation interface subunit and a programming interface subunit, and the operation interface subunit provides a visual operation interface for setting parameters, sending action instructions and displaying the movement status of the robot; The programming interface subunit is used by technicians to program and modify the motion flow of the robotic arm.

10. A robot arm control system for spline injection molding according to claim 1, characterized in that: The hardware module includes a robot body unit, a drive unit and a sensor unit. The robot force includes the gripping force of the end effector for grabbing the injection molded spline and the contact force between the robot and surrounding objects. The data acquisition subunit is used to collect working data of the sensor unit robot during operation. The working data includes position information of each joint of the robot and the end effector, the robot force, and the appearance image of the injection molded spline product.

Images