Accumulated box conveying device and method for full-automatic box pasting machine

Abstract

The invention provides an accumulated box conveying device and method for a full-automatic box pasting machine. The method comprises the steps that edge features of all paper boxes are extracted from depth images of the paper boxes on an accumulation type conveying chain at different visual angles; determining the displacement deviation gradient of each group of adjacent paper boxes in the conveying process according to the edge characteristics of each group of adjacent paper boxes, and predicting the accumulated position deviation of each paper box when reaching the storage area based on all the displacement deviation gradients; the adjustable range of the optimal compensation angle is obtained through all the edge features when the guide roller dynamically presets all the paper boxes; and based on all the accumulated position deviations and the adjustable range of the optimal compensation angle, the queue interval between the paper boxes in the conveying process is cooperatively adjusted, a cooperative queue of the paper boxes in the conveying process is obtained, and the paper boxes on the production line are conveyed according to the cooperative queue. By the adoption of the scheme, group posture cooperative control can be achieved so as to reduce linkage deviation caused by single-point adjustment of the paper boxes on the power and free conveying chain.

Description

Technical Field

[0001] The present application relates to the technical field of carton stacking and conveying, and more specifically, to a carton stacking and conveying device and method for a full-automatic carton gluing machine. Background Art

[0002] Carton stacking and conveying is a key technology in a full-automatic carton gluing machine for precise stacking and orderly transmission of cartons. Its core goal is to ensure that cartons accurately reach the storage area in a preset posture and position through dynamic perception, prediction, and collaborative control during long-distance and high-speed conveying.

[0003] With the increasing demand for automation in the packaging industry, the full-automatic carton gluing machine faces higher precision requirements in the carton forming, stacking, and conveying links. The traditional conveying system relies on mechanical guiding devices with fixed parameters. During long-distance conveying, due to speed fluctuations of the accumulation conveyor chain, differences in vibration characteristics, and interactions between multiple cartons, cartons are prone to cumulative displacement errors and attitude deflections. In the prior art, the position of cartons is usually detected from a single perspective, and single-point adjustment is performed on the cartons, but it is difficult to dynamically predict the misalignment trend of cartons under complex working conditions; the correction angle of the guiding roller is mostly based on a preset threshold, lacking real-time analysis of the spatial correlation between cartons. This results in a high stacking misalignment rate of cartons in the storage area and frequent secondary adjustments, seriously affecting the production rhythm and packaging quality. Therefore, how to achieve group attitude collaborative control to reduce the chain deviation caused by single-point adjustment of cartons on the accumulation conveyor chain has become a problem faced by the industry. Summary of the Invention

[0004] The present application provides a carton conveying device and method for a full-automatic carton gluing machine, which can achieve group attitude collaborative control to reduce the chain deviation caused by single-point adjustment of cartons on the accumulation conveyor chain.

[0005] In a first aspect, the present application provides a carton conveying method for a production line, the production line including a carton stacking and conveying device for a full-automatic carton gluing machine, a long-distance accumulation conveyor chain, and guiding rollers. The method includes the following steps: Collect depth images of cartons from different perspectives on the accumulation conveyor chain; Extract the edge features of each carton from all the depth images; Determine the displacement deviation gradient of each group of adjacent cartons during conveying according to the edge features of each group of adjacent cartons and the conveying speed of the accumulation conveyor chain, and predict the cumulative position deviation of each carton when reaching the storage area based on all the displacement deviation gradients and the vibration characteristics of the accumulation conveyor chain; Perform multi-angle correlation constraints on the attitude angles of each carton on the conveying line through all the edge features to obtain the adjustable range of the optimal compensation angle for the guiding rollers to perform dynamic pre-adjustment on each carton; Based on all the cumulative position deviations and the adjustable range of the optimal compensation angle, the queue interval between each carton during transportation is adjusted collaboratively to obtain a collaborative queue of cartons during transportation, and then the cartons on the production line are transported according to the collaborative queue.

[0006] In some embodiments, extracting the edge features of each carton from all the depth images specifically includes: Performing grayscale processing on all the depth images to obtain a plurality of grayscale images; Performing edge detection on each grayscale image to obtain the edge features of each carton.

[0007] In some embodiments, determining the displacement deviation gradient of each group of adjacent cartons during transportation according to the edge features of each group of adjacent cartons and the transportation speed of the accumulation conveyor chain specifically includes: Obtaining the transportation speed of the accumulation conveyor chain; Determining the central representative points of each carton according to all the edge features; Selecting a group of adjacent cartons as the selected adjacent cartons, and determining the distance between the selected adjacent cartons according to the central representative points corresponding to each carton in the selected adjacent cartons; Determining the displacement deviation gradient of the selected adjacent cartons during transportation through the distance and the transportation speed; Continuing to determine the displacement deviation gradient of the remaining adjacent cartons during transportation.

[0008] In some embodiments, predicting the cumulative position deviation of each carton when reaching the storage area based on all the displacement deviation gradients and the vibration characteristics of the accumulation conveyor chain specifically includes: Determining the vibration characteristics of the accumulation conveyor chain; Selecting a carton as the selected carton and determining the estimated time for the selected carton to reach the storage area; Determining the position influence characteristics of the selected carton according to the estimated time and the vibration characteristics; Determining the cumulative position deviation of the selected carton when reaching the storage area through each displacement deviation gradient corresponding to the selected carton and the position influence characteristics; Continuing to determine the cumulative position deviation of the remaining cartons when reaching the storage area.

[0009] In some embodiments, performing multi-angle correlation constraints on the attitude angles of each carton on the conveyor line through all the edge features to obtain the adjustable range of the optimal compensation angle when the guiding roller dynamically pre-adjusts each carton specifically includes: Selecting a carton as the selected carton and generating point cloud data of the selected carton according to the edge features of the selected carton; Determine the attitude offset angle of the selected carton based on the point cloud data; Continue to determine the attitude offset angles of the remaining cartons; Obtain the attitude angles of each carton on the conveyor line; Perform multi-angle correlation on the attitude angles of each carton through all the attitude offset angles to obtain the multi-angle correlation information of the cartons on the accumulation conveyor chain; Determine the angle adjustment information of the guide roller; Determine the adjustable range of the optimal compensation angle when the guide roller performs dynamic pre-adjustment on each carton according to the multi-angle correlation information and the angle adjustment information.

[0010] In some embodiments, based on all the cumulative position deviations and the adjustable range of the optimal compensation angle, the queue intervals between each carton during the conveying process are coordinately adjusted to obtain the coordinated queue of the cartons during the conveying process, which specifically includes: Determine the coordinated analysis quantity when each carton arranges its orientation during the conveying process according to all the cumulative position deviations; Determine the coupling degree between the orientations of adjacent cartons during the arrangement process of the cartons through all the coordinated analysis quantities and the adjustable range of the optimal compensation angle; Determine the coupling degree threshold between adjacent cartons; Extract each coupling degree that is lower than the coupling degree threshold among the coupling degrees between the orientations of each adjacent carton; Adjust the orientations of the corresponding adjacent cartons according to the extracted coupling degrees to obtain the arrangement results of the corresponding adjacent cartons for the extracted coupling degrees; Keep the orientations of the corresponding adjacent cartons for the coupling degrees that are not extracted unchanged to obtain the arrangement results of the corresponding adjacent cartons for the coupling degrees that are not extracted; Use the above two arrangement results as the coordinated queue of the cartons during the conveying process.

[0011] In some embodiments, the accumulation conveyor chain includes a loading area, a conveying area, an unloading area, a tensioning area, a monitoring area, a turning area, and a buffer area.

[0012] In a second aspect, the present application provides an accumulation carton conveying device for a full-automatic carton gluing machine. The accumulation carton conveying device includes a carton conveying unit, and the carton conveying unit includes: An acquisition module for acquiring depth images of the cartons on the accumulation conveyor chain from different perspectives; A processing module for extracting the edge features of each carton from all the depth images; The processing module is further configured to determine the displacement deviation gradient of each group of adjacent cartons during the conveying process according to the edge features of each group of adjacent cartons and the conveying speed of the accumulation conveyor chain, and predict the cumulative position deviation of each carton when it reaches the storage area based on all the displacement deviation gradients and the vibration characteristics of the accumulation conveyor chain; The processing module is further configured to perform multi-angle correlation constraints on the attitude angles of each carton on the conveying line through all the edge features, so as to obtain the adjustable range of the optimal compensation angle when the guiding roller dynamically pre-adjusts each carton; The execution module is configured to perform collaborative adjustment on the queue intervals between each carton during the conveying process based on all the cumulative position deviations and the adjustable range of the optimal compensation angle, so as to obtain a collaborative queue of the cartons during the conveying process, and then convey the cartons on the production line according to the collaborative queue.

[0013] In a third aspect, the present application provides a computer device, which includes a memory and a processor. The memory stores code, and the processor is configured to obtain the code and execute the above-mentioned carton conveying method for a production line.

[0014] In a fourth aspect, the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the above-mentioned carton conveying method for a production line.

[0015] The technical solution provided by the disclosed embodiments of the present application has the following beneficial effects: In the carton conveying device and method for a full-automatic carton gluing machine provided by the present application, first, depth images of cartons on an accumulation conveyor chain from different perspectives are collected; edge features of each carton are extracted from all the depth images; the displacement deviation gradient of each group of adjacent cartons during the conveying process is determined according to the edge features of each group of adjacent cartons and the conveying speed of the accumulation conveyor chain, and the cumulative position deviation of each carton when it reaches the storage area is predicted based on all the displacement deviation gradients and the vibration characteristics of the accumulation conveyor chain; multi-angle correlation constraints are performed on the attitude angles of each carton on the conveying line through all the edge features, so as to obtain the adjustable range of the optimal compensation angle when the guiding roller dynamically pre-adjusts each carton; collaborative adjustment is performed on the queue intervals between each carton during the conveying process based on all the cumulative position deviations and the adjustable range of the optimal compensation angle, so as to obtain a collaborative queue of the cartons during the conveying process, and then convey the cartons on the production line according to the collaborative queue.

[0016] It can be seen that during the accumulation box transportation process of this application, first, through the feature fusion processing of multi-source depth images, a three-dimensional edge feature model of the paper box contour is established, effectively eliminating the feature extraction error caused by occlusion or stacking. Then, according to the edge features and transportation speed of each group of adjacent paper boxes, the displacement deviation gradient is determined, and the cumulative position deviation is predicted based on the vibration feature, realizing the distributed perception and collaborative prediction of the displacement deviation of the paper box group during the transportation process, breaking through the limitations of traditional single-point adjustment, and effectively avoiding the chain position interference caused by local adjustment. Then, through the edge features, multi-angle correlation constraints are imposed on the posture angle of the paper box, obtaining the adjustable range of the best compensation angle of the guide roller, and dynamically optimizing the compensation angle of the guide roller by using multi-angle correlation constraints, so that the compensation action can satisfy both the current paper box correction requirements and the position constraints of adjacent paper boxes. Then, based on the cumulative position deviation and the best compensation angle, the queue interval is collaboratively adjusted to form a collaborative queue, realizing the distributed collaborative optimization of the paper box queue, thereby improving the overall transportation stability and effectively avoiding the chain reaction deviation caused by single-point adjustment. Finally, the paper boxes on the production line are transported according to the collaborative queue to complete the transportation of the paper boxes on the accumulation conveyor chain. By adopting the above scheme, the collaborative control of the group posture can be realized to reduce the chain deviation caused by single-point adjustment of the paper boxes on the accumulation conveyor chain. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is an exemplary flowchart of a paper box transportation method for a production line shown according to some embodiments of this application; Figure 2 is a structural diagram of different perspectives on an accumulation conveyor chain shown according to some embodiments of this application; Figure 3 is an exemplary flowchart of determining the displacement deviation gradient shown according to some embodiments of this application; Figure 4 is a schematic structural diagram of a paper box transportation unit shown according to some embodiments of this application; Figure 5 is a schematic structural diagram of a computer device for implementing the paper box transportation method for a production line shown according to some embodiments of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] To better understand the technical solutions of this application, the technical solutions of this application will be described in detail below in combination with the accompanying drawings of the specification and specific embodiments.

[0019] Refer to Figure 1, This figure is an exemplary flowchart of a carton conveying method for a production line shown in some embodiments of the present application. In this embodiment, the production line includes a carton stacking and conveying device for a fully automatic carton gluing machine, a long-distance accumulation conveyor chain, and guide rollers. The carton conveying method 100 for the production line mainly includes the following steps: In step 101, depth images of the cartons from different perspectives on the accumulation conveyor chain are collected.

[0020] Among them, the depth image represents the depth image of the cartons on the accumulation conveyor chain in the detection area, and can be used to determine the geometric shape of the cartons on the accumulation conveyor chain in the detection area.

[0021] Specifically, acquisition nodes are arranged at different perspectives in the detection area of the accumulation conveyor chain, and depth images corresponding to the perspectives of each acquisition node are collected at each acquisition node. Specific acquisition devices can be, for example, charge-coupled device cameras, complementary metal-oxide semiconductor cameras, 3D cameras, etc., which are not specifically limited here. In other embodiments, other methods can also be used for acquisition, which are not limited here.

[0022] It should be noted that in the present application, the detection area refers to the area on the accumulation conveyor chain for detecting the state of the cartons. In some embodiments, refer to Figure 2 As shown in the figure, this figure is a structural diagram of different perspectives on the accumulation conveyor chain in some embodiments of the present application. As Figure 2 described, the different perspectives include a top perspective (such as installed vertically above the detection area), a side perspective (such as symmetrically installed on both sides of the detection area, horizontal or slightly inclined to the plane of the detection area), a front and rear perspective (such as installed at the starting end and the ending end of the detection area, facing the moving direction of the cartons), and an oblique perspective (with an angle of 45° - 60° to the detection area, installed on a bracket or gantry).

[0023] In step 102, the edge features of each carton are extracted from all the depth images.

[0024] In some embodiments, the extraction of the edge features of each carton from all the depth images can be implemented by the following steps: All the depth images are grayscaled to obtain a plurality of grayscale images; Edge detection is performed on each grayscale image to obtain the edge features of each carton.

[0025] In specific implementation, grayscale processing is performed on all depth images to obtain multiple grayscale images, which can be achieved in the following manner: First, preprocessing operations of denoising and smoothing are performed on each depth image through filtering methods in the prior art (such as bilateral filtering and guided filtering) to improve the image quality and reduce noise interference. Then, each preprocessed depth image is converted into a grayscale image through linear normalization, thereby obtaining multiple grayscale images. The grayscale images can be used for edge detection operations on the cartons on the accumulation conveyor chain; edge detection is performed on each grayscale image to obtain the edge features of each carton, which can be achieved in the following manner: An edge detection algorithm is used to perform edge detection on the cartons in each grayscale image, and regions with obvious grayscale changes in the grayscale image are detected, thereby determining the edge positions of each carton in each grayscale image. Edge points in the overlapping regions of the same carton at different perspective edge positions are merged through multi-view edge point clustering methods in the prior art (such as DBSCAN and Euclidean clustering) to obtain the three-dimensional edges of each carton. The edge contours of the cartons are extracted from the three-dimensional edges, and the edge contours of each carton are used as the edge features of the corresponding carton; in other embodiments, other methods can also be used to determine, which is not limited here.

[0026] It should be noted that the edge features in this application represent the features of the edge contours of the cartons on the accumulation conveyor chain and can be used to distinguish the cartons on the accumulation conveyor chain.

[0027] In step 103, according to the edge features of each group of adjacent cartons and the conveying speed of the accumulation conveyor chain, the displacement deviation gradient of each group of adjacent cartons during the conveying process is determined, and based on all the displacement deviation gradients and the vibration characteristics of the accumulation conveyor chain, the cumulative position deviation of each carton when reaching the storage area is predicted.

[0028] In some embodiments, referring to Figure 3 As shown, this figure is an exemplary flowchart for determining the displacement deviation gradient in some embodiments of this application. In this embodiment, the displacement deviation gradient of each group of adjacent cartons during the conveying process can be determined according to the edge features of each group of adjacent cartons and the conveying speed of the accumulation conveyor chain, which can be achieved through the following steps: First, in step 1031, the conveying speed of the accumulation conveyor chain is obtained; Secondly, in step 1032, the central representative points of each carton are determined according to all the edge features; Furthermore, in step 1033, a group of adjacent cartons is selected as the selected adjacent cartons, and the distance between the selected adjacent cartons is determined according to the central representative points corresponding to each carton in the selected adjacent cartons; Thus, in step 1034, the displacement deviation gradient of the selected adjacent cartons during the conveying process is determined through the distance and the conveying speed; Finally, in step 1035, continue to determine the displacement deviation gradient of the remaining adjacent cartons during the conveying process.

[0029] When specifically implemented, the conveying speed of the accumulation conveyor chain can be obtained in the following manner: collect the conveying speed of the accumulation conveyor chain from the database of the full-automatic carton gluing machine, where the conveying speed represents the speed of the accumulation conveyor chain when conveying cartons, and the conveying speed includes the speeds of various regions on the accumulation conveyor chain, and each region includes a loading area, a conveying area, an unloading area, a tensioning area, a monitoring area, a turning area, and a buffer area. Since the functions of each region are different, the transmission speeds of each region are different; in other embodiments, other methods can also be used to determine it, and it is not limited here.

[0030] When specifically implemented, the center representative point of each carton can be determined according to all edge features in the following manner: select a carton as the selected carton, and calculate the center point of the selected carton based on the average of polygon vertices. The steps are as follows: First, perform polygon approximation on all edge points based on the edge features of the selected carton to obtain the coordinates of each vertex, and then take the average value of all vertex coordinates as the center point, and use this center point as the center representative point of the selected carton, and continue to determine the center representative points of the remaining cartons, where the center representative point represents the representative point of the center position of the carton; in other embodiments, other determinations can also be used, and it is not limited here.

[0031] When specifically implemented, the distance between the selected adjacent cartons can be determined according to the center representative points corresponding to each carton in the selected adjacent cartons in the following manner: calculate the distance between the selected adjacent cartons by combining the Euclidean distance in the prior art with the coordinates of the center representative points corresponding to each carton in the selected adjacent cartons; the displacement deviation gradient of the displacement deviation of the selected adjacent cartons during the conveying process can be determined by the distance and the conveying speed in the following manner: the displacement deviation represents the offset degree of the distance between adjacent cartons during the movement on the accumulation conveyor chain. Based on the speed change, construct a displacement deviation gradient model of the displacement deviation, consider the influence of the conveying speed of the accumulation conveyor chain on the carton spacing, and establish a relationship model between the phase difference and the speed change. When the speed of the accumulation conveyor chain changes, the spacing between the cartons will also change accordingly, resulting in a phase difference. Therefore, use the phase difference change data of the historical cartons on the accumulation conveyor chain during the conveying process for this relationship model, and optimize this model by the method of cross-validation, and use the optimized model as the displacement deviation gradient model. Input the current distance between the selected adjacent cartons and the conveying speed of the accumulation conveyor chain into this displacement deviation gradient model, and calculate the displacement deviation gradient of the selected adjacent cartons during the conveying process through this displacement deviation gradient model; in some embodiments, other methods can also be used to determine it, and it is not limited here.

[0032] It should be noted that the displacement deviation gradient in this application represents the change gradient of the displacement deviation between adjacent cartons during the conveying process, which can be used to adjust the positions of the cartons on the accumulation conveyor chain, thereby reducing the accumulation of cartons on the accumulation conveyor chain.

[0033] In some embodiments, predicting the cumulative position deviation of each carton when reaching the storage area based on all the displacement deviation gradients and the vibration characteristics of the accumulation conveyor chain can be achieved by the following steps: Determine the vibration characteristics of the accumulation conveyor chain; Select a carton as the selected carton and determine the estimated time for the selected carton to reach the storage area; Determine the position influence characteristics of the selected carton according to the estimated time and the vibration characteristics; Determine the cumulative position deviation of the selected carton when reaching the storage area through each displacement deviation gradient corresponding to the selected carton and the position influence characteristics; Continue to determine the cumulative position deviation of the remaining cartons when reaching the storage area.

[0034] Specifically, determining the vibration characteristics of the accumulation conveyor chain can be achieved in the following way, that is: collect the vibration frequencies of each area on the accumulation conveyor chain through vibration sensors, and use the set of all vibration frequencies as the vibration characteristics, where the vibration characteristics represent the characteristics of the vibration frequencies on the accumulation conveyor chain; determining the estimated time for the selected carton to reach the storage area can be achieved in the following way, that is: extract the time when each historical carton reaches the storage area from the time recorder of the fully automatic carton gluing machine, and use the average time of all the times as the estimated time for the selected carton to reach the storage area, where the estimated time represents the estimated time for the carton to reach the storage area from the starting end of the accumulation conveyor chain; determining the position influence characteristics of the selected carton according to the estimated time and the vibration characteristics can be achieved in the following way, that is: collect the weight of the selected carton through a weight sensor, calculate the variance of all the vibration frequencies in the vibration characteristics to obtain the degree of vibration fluctuation on the accumulation conveyor chain, divide this variance by this weight, which means normalizing the vibration characteristics to the frequency fluctuation per unit weight, multiply the obtained value by the estimated time, perform a natural exponential operation on the multiplied value to simulate the non-linear growth or decay of the vibration of the carton during the conveying process, and use the reciprocal of the value obtained from the natural exponential operation as the characterization value of the position influence characteristics of the selected carton, where the position influence characteristics represent the characteristics of the influence degree of time and the friction situation of the accumulation conveyor chain on the position of the carton; in other embodiments, other methods can also be used for determination, which are not limited here.

[0035] In specific implementation, the cumulative position deviation of the selected carton when reaching the storage area can be determined by the respective displacement deviation gradients corresponding to the carton and the position influence features in the following manner: Based on Newtonian mechanics, the relationship between the phase difference and displacement is derived, a cumulative position deviation model is initialized, and the cumulative position deviation model is constructed through machine learning algorithms (such as linear regression, neural network). The relationship between the phase difference and displacement is used to train the cumulative position deviation model, and the weight coefficients A and B are determined. The cumulative position deviation model is optimized through the cross-validation method. The framework of this model is: cumulative position deviation = phase difference * A + position influence feature * B. The respective displacement deviation gradients and position influence features corresponding to the selected carton are input into this cumulative position deviation model, and the cumulative position deviation of the selected carton when reaching the storage area is output through this cumulative position deviation model; in other embodiments, it can also be determined by other methods, which are not limited here.

[0036] It should be noted that the cumulative position deviation in this application represents a parameter value indicating the degree of deviation of the carton's position on the accumulation conveyor chain, which can be used to globally adjust the cartons on the accumulation conveyor chain, thereby reducing the interference between the cartons on the accumulation conveyor chain.

[0037] In step 104, the attitude angles of each carton on the conveyor line are subjected to multi-angle correlation constraints through all the edge features, and the adjustable range of the best compensation angle when the guiding roller dynamically pre-adjusts each carton is obtained.

[0038] In some embodiments, the adjustable range of the best compensation angle when the guiding roller dynamically pre-adjusts each carton by subjecting the attitude angles of each carton on the conveyor line to multi-angle correlation constraints through all the edge features can be implemented by the following steps: Select a carton as the selected carton, and generate point cloud data of the selected carton according to the edge features of the selected carton; Determine the attitude offset angle of the selected carton through the point cloud data; Continue to determine the attitude offset angles of the remaining cartons; Obtain the attitude angles of each carton on the conveyor line; Perform multi-angle correlation on the attitude angles of each carton through all the attitude offset angles to obtain the multi-angle correlation information of the cartons on the accumulation conveyor chain; Determine the angle adjustment information of the guiding roller; Determine the adjustable range of the best compensation angle when the guiding roller dynamically pre-adjusts each carton according to the multi-angle correlation information and the angle adjustment information.

[0039] In specific implementation, generating the point cloud data of the selected carton according to the edge features of the selected carton can be achieved by the following method, that is: calculating the 3D edge point coordinates of the carton corresponding to the edge features of the selected carton in the three-dimensional space of the depth image through triangulation in the prior art, and generating the 3D edge points of the carton from the 3D edge point coordinates of the carton through SFM (Structure from Motion). When using multi-view stereo to generate all 3D point clouds based on the 3D edge points, all the 3D point clouds are used as the point cloud data of the selected carton. Among them, the point cloud data represents the data of the three-dimensional points on the edge of the carton on the accumulation conveyor chain and can be used to represent the spatial set structure of the carton; determining the attitude offset angle of the selected carton from the point cloud data can be achieved by the following method, that is: registering the point cloud data with the pre-established three-dimensional model of the carton, the iterative closest point algorithm can be used to align the point cloud, and the current attitude angle of the carton after point cloud registration is calculated through Euler angles. The target attitude is obtained from the database of the fully automatic carton gluing machine, and the difference between the current attitude angle and the attitude angle of the target attitude is used as the attitude offset angle of the selected carton. Among them, the attitude offset angle represents the angle of the offset degree between the current carton and the target carton attitude angle and can be used to analyze the correction situation of the carton attitude; in other embodiments, other methods can also be used to determine, which is not limited here.

[0040] In specific implementation, obtaining the attitude angles of each carton on the conveyor line can be achieved by the following method, that is: obtaining the edge features of each carton, locating the 4 corner points of each carton by combining the edge features through a corner point detection method (such as Harris corner detection, Shi-Tomasi algorithm), taking the accumulation conveyor chain as the reference plane, and calculating the attitude angle of the corresponding carton based on the attitude calculation method (such as Perspective-n-Point) combined with the coordinates of the 4 corner points of each carton. Among them, the attitude angle represents the angle of the carton attitude relative to the accumulation conveyor chain; in other embodiments, other methods can also be used to determine, which is not limited here.

[0041] In specific implementation, the attitude angles of each paper box are associated from multiple perspectives through all the attitude offset angles. The multi-perspective association information of the paper boxes on the accumulation conveyor chain can be implemented in the following manner: Select a group of adjacent paper boxes as the selected adjacent paper boxes. Take the difference between the attitude angle of the first paper box and the attitude angle of the second paper box in the selected adjacent paper boxes as the first angle, and take the difference between the attitude offset angle of the first paper box and the attitude offset angle of the second paper box in the selected adjacent paper boxes as the second angle. Take the vector composed of the first angle and the second angle as the multi-perspective association vector between adjacent paper boxes. Continue to determine the multi-perspective association vectors between the remaining adjacent paper boxes. Take the set of all multi-perspective association vectors as the multi-perspective association information of the paper boxes on the accumulation conveyor chain. Among them, the multi-perspective association information represents the information on the association between multiple angles of the paper boxes on the accumulation conveyor chain; in other embodiments, it can also be determined in other ways, which is not limited here.

[0042] In specific implementation, the angle adjustment information of the guiding roller can be determined in the following manner: The guiding roller is a roller used to adjust the direction and angle of the paper boxes on the accumulation conveyor chain. The position and angle of the guiding roller on the accumulation conveyor chain are collected through a position sensor and an angle sensor. The geometric relationship between the guiding roller and related components is analyzed through geometric analysis methods combined with the position and angle, and the angle adjustment amount that the guiding roller needs to be adjusted is determined accordingly. Take the set of the angle adjustment amount and the current angle as the angle adjustment information of the guiding roller. Among them, the angle adjustment information represents the information on the direction adjustment of the paper boxes on the accumulation conveyor chain by the guiding roller; in other embodiments, it can also be determined in other ways, which is not limited here.

[0043] In specific implementation, the adjustable range of the optimal compensation angle for the guiding roller to perform dynamic pre-adjustment on each paper box can be determined according to the multi-perspective association information and the angle adjustment information in the following manner: Construct an angle constraint model based on a machine learning library (such as random forest, neural network). Train this angle constraint model through the angle adjustment data of the guiding roller and each historical paper box, and optimize this angle constraint model through the cross-validation method. Take the association information and the adjustment information as the input features of this association model, and take the adjustable range of the optimal compensation angle as the output feature of this association model. Input the multi-perspective association information and the angle adjustment information into this association model to update the association information and the adjustment information, and output the adjustable range of the optimal compensation angle for the guiding roller to perform dynamic pre-adjustment on each paper box through this association model; in other embodiments, it can also be determined in other ways, which is not limited here.

[0044] It should be noted that the adjustable range in this application represents the adjustment range within which the guide roller can efficiently correct the angles of various paper boxes during angle adjustment, and can be used to control the angle correction of all paper boxes on the accumulation conveyor chain by the guide roller, thereby reducing the angle rotation rate of the guide roller to achieve low-power control of the guide roller for angle control.

[0045] In step 105, based on all the cumulative position deviations and the adjustable range of the optimal compensation angle, the queue intervals between various paper boxes during transportation are coordinately adjusted to obtain a coordinated queue of the paper boxes during transportation, and then the paper boxes on the production line are transported according to the coordinated queue.

[0046] In some embodiments, the coordinated adjustment of the queue intervals between various paper boxes during transportation based on all the cumulative position deviations and the adjustable range of the optimal compensation angle to obtain a coordinated queue of the paper boxes during transportation can be achieved by the following steps: Determine the coordinated analysis quantity of each paper box during the orientation arrangement during transportation according to all the cumulative position deviations; Determine the coupling degree between the orientations of adjacent paper boxes during the arrangement of the paper boxes through all the coordinated analysis quantities and the adjustable range of the optimal compensation angle; Determine the coupling degree threshold between adjacent paper boxes; Extract the coupling degrees that are lower than the coupling degree threshold among the coupling degrees between the orientations of the adjacent paper boxes; Adjust the orientations of the corresponding adjacent paper boxes according to the extracted coupling degrees to obtain the arrangement results of the corresponding adjacent paper boxes for the extracted coupling degrees; Keep the orientations of the adjacent paper boxes corresponding to the coupling degrees that are not extracted unchanged to obtain the arrangement results of the corresponding adjacent paper boxes for the coupling degrees that are not extracted; Use the above two arrangement results as the coordinated queue of the paper boxes during transportation.

[0047] In specific implementation, the collaborative analysis quantity for the orientation arrangement of each carton during the conveying process can be determined based on all the cumulative position deviations by the following method: According to the kinematic and dynamic characteristics of the cartons during the conveying process, a mathematical relationship model between the carton orientation arrangement and the cumulative position deviations is established. By combining all the cumulative position deviations with this mathematical relationship model, the movement trajectories and force conditions of each carton on the conveying equipment are analyzed, and thus the collaborative analysis quantity corresponding to each carton is derived. Among them, the collaborative analysis quantity represents the parameter value of the adjustment degree during the collaborative analysis of each carton on the accumulation conveyor chain; The coupling degree between the orientations of adjacent cartons during the arrangement of cartons can be determined by all the collaborative analysis quantities and the adjustable range of the optimal compensation angle in the following way: Based on the coupling coordination degree model, the coupling degree of the cartons is defined as the matching degree between the collaborative analysis quantity and the ideal adjustment quantity under the adjustable range of the optimal compensation angle, and a specific mathematical method (such as weighted summation, multiple regression) is used as the calculation formula for this matching degree. A group of adjacent cartons is selected as the selected adjacent cartons, and the mean value of the two collaborative analysis quantities corresponding to the selected adjacent cartons and the adjustable range of the optimal compensation angle are input into this calculation formula, and the value calculated by this calculation formula is used as the coupling degree between the orientations of the adjacent cartons, and then the coupling degrees between the orientations of the remaining adjacent cartons are continuously determined; In other embodiments, other methods can also be adopted, which are not limited here.

[0048] It should be noted that the coupling degree in this application represents the parameter value of the coupling degree between the orientations of two cartons in the orientation arrangement during the conveying process of the accumulation conveyor chain, and can be used to arrange the cartons on the accumulation conveyor chain to facilitate obtaining a better coordinated orientation arrangement.

[0049] In specific implementation, the coupling degree threshold between adjacent cartons can be determined in the following way: If the coupling degree between adjacent cartons is low, a large stacking misalignment amount will occur when transporting the adjacent cartons to the storage area. The coupling degree data between adjacent cartons with a low historical misalignment amount are extracted from the fully automatic carton gluing machine, and the mean value of all the coupling degrees in the coupling degree data is used as the coupling degree threshold between adjacent cartons. When the coupling degree between the orientations of adjacent cartons during the conveying process reaches the coupling degree threshold, the misalignment amount of the adjacent cartons transported to the storage area is low, and thus no secondary stacking is required; The orientations of the corresponding adjacent cartons can be adjusted according to the extracted coupling degrees to obtain the arrangement results of the corresponding adjacent cartons for the extracted coupling degrees in the following way: The orientations of the corresponding adjacent cartons for the extracted coupling degrees are adjusted by the guide rollers until the coupling degree between each adjacent carton is greater than the coupling degree threshold, then the orientations of the adjacent cartons do not need to be adjusted, and the orientation arrangement of each carton obtained after adjustment is used as the arrangement result of the corresponding adjacent cartons for the coupling degrees that are not extracted.

[0050] In specific implementation, conveying the cartons on the production line according to the collaborative queue can be achieved in the following manner, namely: conveying the collaborative queue to the control system of the production line, adjusting the control parameters of the production line through the adjustment model of the control system, and thereby controlling the carton conveying process of the production line through the adjusted control parameters; in other embodiments, other control methods can also be used, which are not limited here.

[0051] In addition, in another aspect of the present application, in some embodiments, the present application provides a box stacking and conveying device for a fully automatic box gluer, the box stacking and conveying device comprising a carton conveying unit, referring to Figure 4 , which is a structural schematic diagram of a carton conveying unit according to some embodiments of the present application, the carton conveying unit 400 includes: a collection module 401, a processing module 402 and an execution module 403, which are respectively described as follows: Acquisition module 401, in this application, acquisition module 401 is mainly used to acquire depth images of cartons on the accumulation conveyor chain at different viewing angles; Processing module 402, in this application, the processing module 402 is used to extract edge features of each paper box from all depth images; It should be noted that the processing module 402 in the present application is also used to determine the displacement deviation gradient of each group of adjacent cartons during the conveying process according to the edge characteristics of each group of adjacent cartons and the conveying speed of the accumulation type conveyor chain, and predict the cumulative position deviation of each carton when it arrives at the storage area based on all the displacement deviation gradients and the vibration characteristics of the accumulation type conveyor chain; In addition, it should be noted that the processing module 402 in the present application is also used to perform multi-angle association constraints on the posture angles of each paper box on the conveyor line through all edge features, so as to obtain the adjustable range of the optimal compensation angle when the guide roller dynamically pre-adjusts each paper box; Execution module 403. In the present application, execution module 403 is mainly used to coordinately adjust the queue intervals between each carton during the conveying process based on all accumulated position deviations and the adjustable range of the optimal compensation angle, so as to obtain a coordinated queue of the carton during the conveying process, and then convey the carton on the production line according to the coordinated queue.

[0052] In addition, the present application also provides a computer device, which includes a memory and a processor, wherein the memory stores codes, and the processor is configured to obtain the codes and execute the above-mentioned carton conveying method for a production line.

[0053] In some embodiments, reference Figure 5, which is a schematic structural diagram of a computer device for implementing a carton conveying method for a production line according to some embodiments of the present application. The carton conveying method for a production line in the above embodiments can be implemented by Figure 5 the computer device shown. The computer device 500 includes at least one processor 501, a communication bus 502, a memory 503, and at least one communication interface 504.

[0054] The processor 501 can be a general-purpose central processing unit (CPU) or an application specific integrated circuit (ASIC).

[0055] The communication bus 502 can be used to transmit information between the above components.

[0056] The memory 503 can be a read only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read only memory (EEPROM), a compact disc read only memory (CD ROM) or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disks or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 503 can exist independently and be connected to the processor 501 through the communication bus 502. The memory 503 can also be integrated with the processor 501.

[0057] Among them, the memory 503 is used to store the program code for executing the solution of the present application and is controlled by the processor 501 to execute. The processor 501 is used to execute the program code stored in the memory 503. The program code can include one or more software modules. The method used in the above embodiments can be implemented by one or more software modules in the program code in the processor 501 and the memory 503.

[0058] A communication interface 504, using any transceiver-like device, is used to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0059] In a specific implementation, as an example, a computer device may include multiple processors, and each of these processors may be a single-core (single CPU) processor or a multi-core (multi CPU) processor. Here, the processor may refer to one or more devices, circuits, and / or processing cores for processing data (such as computer program instructions).

[0060] The above computer device may be a general-purpose computer device or a special-purpose computer device. In a specific implementation, the computer device may be a desktop computer, a laptop computer, a network server, a personal digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. The embodiments of the present application do not limit the type of the computer device.

[0061] In addition, the present application also provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, it implements the above-mentioned carton conveying method for a production line.

[0062] Although the preferred embodiments of the present application have been described, those skilled in the art can make additional changes and modifications once they know the basic creative concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications falling within the scope of the present application.

[0063] Obviously, those skilled in the art can make various changes 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.

Claims

1. A carton conveying method for a production line, the production line comprising a carton accumulating and conveying device for a full-automatic carton gluing machine, a long-distance accumulation conveyor chain, and guide rollers, characterized in that, The method comprises the following steps: Collect depth images of the cartons at different perspectives on the accumulation conveyor chain; Extract the edge features of each carton from all the depth images; Determine the displacement deviation gradient of each group of adjacent cartons during the conveying process according to the edge features of each group of adjacent cartons and the conveying speed of the accumulation conveyor chain, and predict the cumulative position deviation of each carton when it reaches the storage area based on all the displacement deviation gradients and the vibration characteristics of the accumulation conveyor chain; Perform multi-angle correlation constraints on the attitude angles of each carton on the conveying line through all the edge features to obtain the adjustable range of the optimal compensation angle when the guiding roller dynamically pre-adjusts each carton; Based on all the cumulative position deviations and the adjustable range of the optimal compensation angle, synergistically adjust the queue intervals between each carton during the conveying process to obtain the collaborative queue of the cartons during the conveying process, and then convey the cartons on the production line according to the collaborative queue; 2. The method according to claim 1, wherein Specifically, extracting the edge features of each carton from all the depth images includes: Perform grayscale processing on all the depth images to obtain a plurality of grayscale images; Perform edge detection on each grayscale image to obtain the edge features of each carton; 3. The method according to claim 1, wherein Specifically, determining the displacement deviation gradient of each group of adjacent cartons during the conveying process according to the edge features of each group of adjacent cartons and the conveying speed of the accumulation conveyor chain includes: Obtain the conveying speed of the accumulation conveyor chain; Determine the central representative point of each carton according to all the edge features; Select a group of adjacent cartons as the selected adjacent cartons, and determine the distance between the selected adjacent cartons according to the central representative points corresponding to each carton in the selected adjacent cartons; Determine the displacement deviation gradient of the selected adjacent cartons during the conveying process through the distance and the conveying speed; Continue to determine the displacement deviation gradients of the remaining adjacent cartons during the conveying process; 4. The method according to claim 1, characterized in that, Specifically, predicting the cumulative position deviation of each carton when it reaches the storage area based on all the displacement deviation gradients and the vibration characteristics of the accumulation conveyor chain includes: Determine the vibration characteristics of the accumulation conveyor chain; Select a carton as the selected carton and determine the estimated time for the selected carton to reach the storage area; Determine the position influence characteristics of the selected carton according to the estimated time and the vibration characteristics; Determine the cumulative position deviation of the selected carton when it reaches the storage area through each displacement deviation gradient corresponding to the selected carton and the position influence characteristics; Continue to determine the cumulative position deviations of the remaining cartons when they reach the storage area; 5. The method according to claim 1, wherein Specifically, performing multi-angle correlation constraints on the attitude angles of each carton on the conveying line through all the edge features to obtain the adjustable range of the optimal compensation angle when the guiding roller dynamically pre-adjusts each carton includes: Select a carton as the selected carton and generate the point cloud data of the selected carton according to the edge features of the selected carton; Determine the attitude offset angle of the selected carton through the point cloud data; Continue to determine the attitude offset angles of the remaining cartons; Obtain the attitude angles of each carton on the conveying line; Perform multi-angle correlation on the attitude angles of each carton through all the attitude offset angles to obtain the multi-angle correlation information of the cartons on the accumulation conveyor chain; Determine the angle adjustment information of the guiding roller; Determine the adjustable range of the optimal compensation angle when the guiding roller dynamically pre-adjusts each paper box according to the multi-angle correlation information and the angle adjustment information.

6. The method according to claim 1, wherein Based on all the cumulative position deviations and the adjustable range of the optimal compensation angle, coordinately adjust the queue intervals between each paper box during the conveying process, and the specific steps for obtaining the coordinated queue of the paper boxes during the conveying process include: Determine the coordinated analysis quantity when each paper box arranges its orientation during the conveying process according to all the cumulative position deviations; Determine the coupling degree between the orientations of adjacent paper boxes during the arrangement process of the paper boxes through all the coordinated analysis quantities and the adjustable range of the optimal compensation angle; Determine the coupling degree threshold between adjacent paper boxes; Extract the coupling degrees that are lower than the coupling degree threshold among the coupling degrees between the orientations of adjacent paper boxes; Adjust the orientations of the corresponding adjacent paper boxes according to the extracted coupling degrees to obtain the arrangement results of the corresponding adjacent paper boxes for the extracted coupling degrees; Keep the orientations of the adjacent paper boxes corresponding to the coupling degrees that are not extracted unchanged to obtain the arrangement results of the corresponding adjacent paper boxes for the coupling degrees that are not extracted; Use the above two arrangement results as the coordinated queue of the paper boxes during the conveying process.

7. The method according to claim 1, wherein The accumulation conveyor chain includes a loading area, a conveying area, an unloading area, a tensioning area, a monitoring area, a turning area, and a buffer area.

8. A box stacking and conveying device for a full-automatic box gluing machine, the box stacking and conveying device comprising a paper box conveying unit, characterized in that, The paper box conveying unit includes: An acquisition module, configured to acquire depth images of the paper boxes on the accumulation conveyor chain from different perspectives; A processing module, configured to extract the edge features of each paper box from all the depth images; The processing module is further configured to determine the displacement deviation gradient of each group of adjacent paper boxes during the conveying process according to the edge features of each group of adjacent paper boxes and the conveying speed of the accumulation conveyor chain, and predict the cumulative position deviation of each paper box when it reaches the storage area based on all the displacement deviation gradients and the vibration characteristics of the accumulation conveyor chain; The processing module is further configured to perform multi-angle correlation constraints on the attitude angles of each paper box on the conveying line through all the edge features to obtain the adjustable range of the optimal compensation angle when the guiding roller dynamically pre-adjusts each paper box; An execution module, configured to coordinately adjust the queue intervals between each paper box during the conveying process based on all the cumulative position deviations and the adjustable range of the optimal compensation angle to obtain the coordinated queue of the paper boxes during the conveying process, and then convey the paper boxes on the production line according to the coordinated queue.

9. A computer device, characterized in that, The computer device includes a memory and a processor, the memory stores code, and the processor is configured to obtain the code and execute the paper box conveying method for a production line according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the paper box conveying method for a production line according to any one of claims 1 to 7.

Images