A visual guidance-based method and system for middle taping of battery cell

Abstract

A kind of visual guidance-based middle gluing method and system of battery cell, it is related to battery gluing field, in the method, first, the battery cell column image information is obtained using visual acquisition device, adaptive path fitting is carried out based on the space coordinates of representative point, and vertical gluing reference path is generated;Then, the position data of the area that has completed gluing is collected in real time during the gluing process, the deviation trend characteristics are calculated to carry out dynamic compensation, and the corrected gluing track is obtained;Finally, the corrected gluing track is used to plan the horizontal gluing path according to adaptive spacing.This application not only improves the planning accuracy of vertical gluing reference path, but also suppresses the trajectory deviation caused by process parameter fluctuation during gluing process, ensures the uniformity and consistency of gluing line, and makes the horizontal gluing and vertical gluing better match, improves the gluing quality of intersection area.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of battery rubberizing, and particularly relates to a method and system for middle rubberizing of battery cells based on visual guidance. BACKGROUND

[0002] In the production process of lithium batteries, middle rubberizing of battery cells is an important process step for improving the structural stability of battery modules. The quality of middle rubberizing of battery cells directly affects the heat dissipation performance, structural strength and service life of the battery modules. The traditional method of middle rubberizing of battery cells mainly adopts preset trajectories for fixed-path rubberizing, and determines the rubberizing position through robot teaching or visual positioning to form a grid-shaped rubberizing path. In actual production, although this fixed-trajectory rubberizing method is simple to operate, it is difficult to adapt to fluctuations in process parameters during the rubberizing process.

[0003] In related technologies, the rubberizing process usually determines the starting position of rubberizing by mechanical positioning or visual positioning. In actual rubberizing, the position of the battery cell is fixed by mechanical limiting or jig, and then equal-distance grid-shaped rubberizing is performed based on a preset fixed spacing. Some improved schemes use a vision system to position the battery cell position to determine the starting and ending points of rubberizing, but the rubberizing process still uses a fixed straight-line interpolation motion method.

[0004] However, in actual rubberizing, fluctuations in process parameters such as glue output and rubberizing speed will cause deviations between the actual rubberizing trajectory and the planned trajectory, and thus cause quality problems such as uneven glue amount in local areas and inconsistent rubberizing line width. As new energy vehicles continue to improve the performance requirements of power batteries, the integration and energy density of battery modules continue to improve, which puts higher requirements on the consistency and reliability of middle rubberizing of battery cells. SUMMARY

[0005] The application provides a method and system for middle rubberizing of battery cells based on visual guidance, which is used to improve the accuracy and consistency of middle rubberizing of battery cells. The method uses a vision system to collect position information in real time during the rubberizing process, and corrects the rubberizing trajectory through dynamic compensation calculation, effectively solving the rubberizing deviation problem caused by fluctuations in process parameters, and achieving high-quality grid-shaped rubberizing.

[0006] In a first aspect, the application provides a method for middle rubberizing of battery cells based on visual guidance, comprising:

[0007] acquiring image information of the battery cell column using a visual acquisition device to obtain visual data of the battery cell;

[0008] extracting feature point information of the battery cell column according to the visual data to obtain spatial coordinates of representative points, and performing adaptive path fitting calculation based on the spatial coordinates of the representative points to obtain a first-direction rubberizing reference path equation;

[0009] control the movement of the taping head according to the first direction taping reference path equation to obtain an initial taping track;

[0010] collect position data of the completed taping area in real time, calculate a deviation trend feature of the taping position, perform dynamic compensation calculation according to the deviation trend feature to obtain a corrected taping head movement track;

[0011] based on the corrected taping track, plan a plurality of second direction taping paths at adaptive intervals, the second direction being perpendicular to the first direction.

[0012] By adopting the technical solution, the application first acquires cell column image information by using a visual acquisition device, performs adaptive path fitting based on the spatial coordinates of representative points to generate a vertical taping reference path; then collects position data of the completed taping area in real time during the taping process, performs dynamic compensation by calculating a deviation trend feature to obtain a corrected taping track; finally, plans a horizontal taping path based on the corrected taping track at adaptive intervals; not only improves the planning accuracy of the vertical taping reference path, but also suppresses the track deviation caused by process parameter fluctuation during the taping process, ensures the uniformity and consistency of the taping line, and makes the horizontal taping and vertical taping better match to improve the taping quality of the intersection area.

[0013] In combination with some embodiments of the first aspect, in some embodiments, the obtaining of the spatial coordinates of the representative points based on the feature point information of the cell column extracted from the visual data and the adaptive path fitting calculation based on the spatial coordinates of the representative points to obtain the first direction taping reference path equation specifically comprises:

[0014] extracting cell column edge contour features from the visual data to obtain a set of cell column edge points;

[0015] selecting feature points of the head, middle and tail regions of the cell column from the set of edge points to obtain spatial coordinates of three representative points;

[0016] performing fitting on the three representative points by using a least square method to obtain the first direction taping reference path equation.

[0017] By adopting the technical solution, the application extracts cell column edge contour features from visual data to obtain a set of edge points, selects feature points of the head, middle and tail regions of the cell column as representative points to obtain spatial coordinates, and performs fitting on the three representative points by using a least square method to obtain a vertical taping reference path equation; by selecting representative points at key positions for path fitting, the calculation process is simplified, and the fitting accuracy of the reference path is ensured, which can provide an accurate path reference for subsequent taping.

[0018] In some embodiments of the first aspect, in some embodiments, the feature point information of the cell column is extracted from the visual data to obtain spatial coordinates of representative points, and adaptive path fitting is performed based on the spatial coordinates of the representative points to obtain a first-direction adhesive application reference path equation, specifically including:

[0019] An edge contour feature of the cell column is extracted from the visual data to obtain an edge point set of the cell column;

[0020] The edge point set is subjected to gray gradient analysis to identify position information of cell gaps, and a center connecting line of adjacent cells is calculated based on the position information of the cell gaps to obtain spatial coordinates of a plurality of groups of local reference points;

[0021] The plurality of groups of local reference points are fitted to obtain a first-direction adhesive application reference path equation.

[0022] By using the above technical solution, after the edge contour feature of the cell column is extracted from the visual data to obtain an edge point set, the position information of the cell gaps is identified through gray gradient analysis, the center connecting line of adjacent cells is calculated based on the cell gaps to obtain spatial coordinates of a plurality of groups of local reference points, and the local reference points are fitted to obtain a vertical adhesive application reference path equation; more local reference points are obtained by identifying the cell gaps, the sampling density of path fitting is improved, the vertical adhesive application reference path better fits the actual distribution characteristics of the cell column, and thus the accuracy of the adhesive application path is improved.

[0023] In some embodiments of the first aspect, in some embodiments, the position data of the completed adhesive application area is collected in real time, an adhesive application position deviation trend feature is calculated, dynamic compensation calculation is performed according to the deviation trend feature to obtain a corrected adhesive head motion trajectory, specifically including:

[0024] The completed adhesive application area is scanned by the visual collection device to collect image data of the completed adhesive application area, and position information of an actual adhesive application line is extracted;

[0025] The position information of the actual adhesive application line is compared with the planned path to obtain a position deviation, and the cell contour of the forward area to be applied with adhesive is analyzed by the visual collection device to obtain position features of a target area;

[0026] Compensation calculation is performed according to the historical change data of the position deviation and the position features of the target area to obtain a corrected adhesive head motion trajectory, and the historical change data includes a cumulative deviation amount of the backward area, a deviation change rate, and speed and adhesive amount parameters in the adhesive application process.

[0027] By adopting the technical scheme, the application realizes real-time scanning of the backward area to which the adhesive has been applied by the visual acquisition device, extracts actual adhesive application line position information, and obtains position deviation by comparing the position information with the planned path. Meanwhile, the application analyzes the forward area to which the adhesive is to be applied to obtain target area position characteristics. The application performs compensation calculation based on historical change data of the position deviation and the target area position characteristics, and obtains a corrected adhesive application head motion trajectory. By analyzing the deviation characteristics of the backward area to which the adhesive has been applied and the position characteristics of the forward area to which the adhesive is to be applied in real time, the application establishes a dynamic compensation mechanism, which can not only discover and correct trajectory deviation in the adhesive application process in a timely manner, but also takes into account the influence of process parameters on the adhesive application quality, and realizes accurate control of the adhesive application trajectory.

[0028] In combination with some embodiments of the first aspect, in some embodiments, the planning of the plurality of second direction adhesive application paths according to the adaptive spacing based on the corrected adhesive application trajectory specifically comprises:

[0029] extracting edge contour characteristics of each first direction adhesive application line based on the corrected adhesive application trajectory;

[0030] determining intersection positions of each second direction adhesive application path and all first direction adhesive application lines according to adaptive spacing between adjacent two first direction adhesive application lines based on the edge contour characteristics, wherein the adaptive spacing is determined according to actual spacing between adjacent first direction adhesive application lines and a width parameter of the first direction adhesive application line, or the adaptive spacing is determined according to actual spacing of the cell column and adhesive application process parameters;

[0031] obtaining target point positions of each second direction adhesive application based on the intersection positions.

[0032] By adopting the technical scheme, the application extracts edge contour characteristics of each vertical adhesive application line based on the corrected adhesive application trajectory, and determines intersection positions of the horizontal adhesive application path and the vertical adhesive application line according to the adaptive spacing based on the characteristics, and finally obtains target point positions of the horizontal adhesive application. By analyzing actual characteristics of the vertical adhesive application line to determine the adaptive spacing, the horizontal adhesive application can better adapt to actual position and width change of the vertical adhesive application line, avoiding uneven distribution of adhesive in the intersection area, and improving overall quality of the grid-shaped adhesive application structure.

[0033] In combination with some embodiments of the first aspect, in some embodiments, the planning of the plurality of second direction adhesive application paths according to the adaptive spacing based on the corrected adhesive application trajectory specifically comprises:

[0034] performing gray scale gradient analysis on the first direction adhesive application line based on the corrected adhesive application trajectory to obtain center line positions and second direction width distribution characteristics of each adhesive application line;

[0035] According to the center line position and the width distribution feature, a position of an intersection of each second direction taping path and all first direction taping lines is determined according to an adaptive interval between two adjacent first direction taping lines, and the adaptive interval is determined according to an actual interval between the adjacent first direction taping lines, a width parameter of the first direction taping line, and a glue amount distribution feature of the intersection region;

[0036] A second direction taping target trajectory is obtained based on the intersection position and the intersection region feature.

[0037] By adopting the technical solution, the center line position and the width distribution feature of the vertical taping line are obtained by performing gray gradient analysis on the vertical taping line based on the corrected taping trajectory, the adaptive interval is determined according to the features and the glue amount distribution feature of the intersection region, and then the horizontal taping target trajectory is obtained. The more complete feature information of the vertical taping line is obtained by the gray gradient analysis, and the glue amount distribution feature of the intersection region is considered, so that the horizontal taping better adapts to the actual state of the vertical taping line, the problem of excessive or insufficient glue amount in the intersection region can be effectively avoided, and the uniformity and reliability of the grid-shaped taping structure are improved.

[0038] In combination with some embodiments of the first aspect, in some embodiments, the method further comprises:

[0039] According to the size of the cell column, the taping region is divided into two regions, and corresponding working regions of two taping robots are obtained;

[0040] Based on the working regions, motion parameters of the two taping robots are configured, and synchronous taping operation control instructions of the two taping robots are obtained;

[0041] According to the control instructions, a real-time state monitoring mechanism of the two taping robots is established, and position information and taping state data of the two taping robots are obtained;

[0042] Based on the taping state data, a motion speed compensation amount of the two taping robots is calculated, and dynamically adjusted synchronous motion parameters are determined;

[0043] According to the synchronous motion parameters, an abnormal state of the taping robots is monitored, and a collaborative control strategy of the taping robots is obtained.

[0044] By adopting the technical solution, the present application divides the adhesive area into two working areas according to the size of the battery cell column, configures the motion parameters of the two adhesive mechanical hands to realize synchronous adhesive, and establishes a real-time state monitoring mechanism to obtain position information and adhesive state data, calculates the speed compensation amount based on the data to determine the dynamically adjusted synchronous motion parameters, and then obtains the collaborative control strategy of the mechanical hands; the synchronous adhesive efficiency is improved through the collaborative work of the two mechanical hands, the synchronism of the two mechanical hands is ensured through real-time monitoring and dynamic adjustment, the adhesive quality problem caused by the asynchronization of the mechanical hands is effectively avoided, and efficient and stable large-area adhesive work is realized.

[0045] In a second aspect, the embodiments of the present application provide an electric cell middle adhesive system based on visual guidance, comprising: one or more processors and a memory; the memory is coupled with the one or more processors, and the memory is configured to store computer program codes, the computer program codes comprising computer instructions, and the one or more processors invoke the computer instructions to enable the system to perform the method described in the first aspect and any possible implementation manner of the first aspect.

[0046] In a third aspect, the embodiments of the present application provide a computer readable storage medium comprising instructions, which, when executed on a system, cause the system to perform the method described in the first aspect and any possible implementation manner of the first aspect.

[0047] In a fourth aspect, the embodiments of the present application provide a computer program product, which, when executed on a system, causes the system to perform the method described in any possible implementation manner of the first aspect.

[0048] The one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0049] 1. The present application provides an electric cell middle adhesive method based on visual guidance, first, an image information of a battery cell column is acquired by using a visual acquisition device, an adaptive path fitting is performed based on the spatial coordinates of representative points to generate a vertical adhesive reference path; then, position data of an adhesive area completed in the adhesive process is acquired in real time, a dynamic compensation is performed by calculating deviation trend characteristics to obtain a corrected adhesive track; finally, a horizontal adhesive path is planned according to the adaptive spacing based on the corrected adhesive track; not only the planning accuracy of the vertical adhesive reference path is improved, but also the track deviation caused by the process parameter fluctuation in the adhesive process is inhibited, the uniformity and consistency of the adhesive line are ensured, the horizontal adhesive and the vertical adhesive are better matched, and the adhesive quality of the cross region is improved.

[0050] 2. The application provides a visual guidance-based middle cell gluing method, which extracts cell column edge profile features from visual data to obtain an edge point set, selects feature points at the head, middle and tail regions of the cell column as representative points to obtain spatial coordinates, and adopts a least square method to fit the three representative points to obtain a vertical gluing reference path equation; the representative points at key positions are selected for path fitting, which simplifies the calculation process and ensures the fitting accuracy of the reference path, and can provide accurate path reference for subsequent gluing.

[0051] 3. The application provides a visual guidance-based middle cell gluing method, which extracts edge profile features of each vertical gluing line based on the corrected gluing trajectory, and determines the intersection positions of the horizontal gluing path and the vertical gluing line according to the features at an adaptive interval to finally obtain target points of the horizontal gluing; the adaptive interval is determined by analyzing the actual features of the vertical gluing line, so that the horizontal gluing can better adapt to the actual position and width change of the vertical gluing line, avoiding uneven distribution of glue in the intersection area and improving the overall quality of the grid-shaped gluing structure. BRIEF DESCRIPTION OF DRAWINGS

[0052] Figure 1 is a flowchart of a visual guidance-based middle cell gluing method in an embodiment of the application.

[0053] Figure 2 is a gluing schematic diagram of a visual guidance-based middle cell gluing method in an embodiment of the application.

[0054] Figure 3 is a cell arrangement schematic diagram in a visual guidance-based middle cell gluing method in an embodiment of the application.

[0055] Figure 4 is a schematic diagram of an entity device structure of a visual guidance-based middle cell gluing system provided in an embodiment of the application. DETAILED DESCRIPTION

[0056] The terms used in the following embodiments of the application are only for the purpose of describing the specific embodiments and are not intended to be limiting to the application. As used in the specification and the appended claims of the application, the singular forms "a," "an," and "the" are intended to include both singular and plural forms, unless the context clearly indicates otherwise. It will be further understood that the terms "and / or" used in the application means any or all possible combinations of one or more of the listed items.

[0057] The terms "first", "second", "third", etc. are used only for the purpose of description and do not imply or suggest relative importance or imply a specific number of the technical features indicated. Therefore, the features defined as "first", "second", etc. can explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, the meaning of "a plurality of" is two or more, unless otherwise specified.

[0058] In the field of power battery manufacturing, the middle adhesive of the battery cell is a key process to ensure the stability and heat dissipation performance of the battery module structure. With the increasing demand for energy density of power batteries for electric vehicles, the structure of the battery module is becoming more compact, which puts higher requirements on the precision and consistency of the middle adhesive of the battery cell. The traditional adhesive method cannot monitor and adjust the adhesive quality in real time, and is difficult to adapt to the fluctuations of process parameters such as adhesive output and adhesive speed, which may cause problems such as inconsistent adhesive line width and uneven adhesive distribution in the cross region, affecting the heat dissipation performance and structural strength of the battery module.

[0059] In related technologies, the adhesive process usually uses mechanical positioning or single visual positioning to determine the adhesive position, and performs grid-shaped adhesive through a preset trajectory and fixed spacing. Specifically, the battery cell is first positioned based on mechanical limiting or jigs, then vertical adhesive is completed by simple straight-line interpolation motion, and horizontal adhesive is performed in an equal-interval manner. Although this method is simple, it has obvious shortcomings in actual production: first, the fixed trajectory method cannot adapt to the adhesive deviation caused by process parameter fluctuations; second, the preset spacing method cannot guarantee the uniformity of adhesive in the cross region; in addition, the lack of real-time monitoring and compensation mechanism leads to significant fluctuations in adhesive quality.

[0060] The present application is mainly applied to the middle adhesive of the battery cell of the power battery module, and is especially suitable for grid-shaped adhesive process of large-size battery cells. In these application scenarios, it is necessary to ensure the precision of vertical adhesive and horizontal adhesive, and to ensure the uniform distribution of adhesive in the cross region. In order to solve the above technical problems, the present application provides a middle adhesive method and system for battery cell based on visual guidance. In the following, an embodiment is used to describe the present application in combination with Figure 1 A middle adhesive method and system for battery cell based on visual guidance in an embodiment of the present application are described as follows:

[0061] Please refer to Figure 1 A flowchart of a middle adhesive method for battery cell based on visual guidance in an embodiment of the present application is shown in the following figure.

[0062] S101, acquiring image information of the battery cell column by using a visual acquisition device to obtain visual data of the battery cell.

[0063] Specifically, as Figure 2As shown, the present application adopts a double-robot symmetrical arrangement scheme, and two industrial robots, i.e., the adhesive applying robots, are respectively installed on the two sides of the battery cell conveying line. Each robot is configured with two adhesive applying heads, one for production operation and the other as a backup, and a total of four adhesive applying heads, Figure 2 The two adhesive applying heads corresponding to the lower adhesive applying robot are shown (one backup and one in use), and when the adhesive applying head is not working, it is placed on the adhesive applying head base beside it. When in use, the adhesive applying robot is connected to the quick-change disc above the adhesive applying head and performs adhesive application. When the battery cell columns are synchronously conveyed along the conveying line, the two adhesive applying robots can simultaneously perform adhesive application at the middle position of different battery cell columns.

[0064] Specifically, the vision acquisition device includes an industrial camera and an image transmission line. The industrial camera adopts a high-resolution and high-frame-rate model to ensure clear and fast acquisition of image information of the battery cell columns. The industrial camera is installed above the adhesive applying equipment and maintains a fixed distance from the battery cell columns, and the field of view covers the entire adhesive applying work area. In addition, a high-speed camera can also be used instead of the industrial camera to obtain better dynamic image acquisition effect. In the present embodiment, a multi-sensor vision acquisition scheme is adopted to acquire image information of the battery cell columns. Each adhesive applying robot is equipped with three vision acquisition devices: the first vision acquisition device is installed vertically to obtain the overall distribution characteristics of the battery cell columns; the second vision acquisition device is used to monitor the adhesive amount distribution in real time during the adhesive applying process; and the third vision acquisition device is responsible for detecting the position deviation of the adhesive applying track. These vision acquisition devices can be industrial cameras. The multi-sensor systems on the two adhesive applying robots work in a synchronous triggering mode to ensure the time sequence consistency of data acquisition. This multi-sensor cooperative vision acquisition scheme realizes all-around monitoring of the adhesive applying process, can timely discover and correct adhesive applying quality abnormalities, and ensures adhesive applying precision and stability.

[0065] Further, to realize the cooperative work of the two robots, first, the rubber sticking area is divided into two corresponding work areas according to the size (such as length, width, etc.) of the battery cell column. Then, based on the work area, the motion parameters of the two rubber sticking robots are configured, and the synchronous rubber sticking operation control instructions of the two robots are obtained. At the same time, the real-time state monitoring mechanism of the two rubber sticking robots is established through the sensor to obtain the position information and the rubber sticking state data. Based on the rubber sticking state data, the motion speed compensation amount of the two rubber sticking robots is calculated, and the dynamically adjusted synchronous motion parameters are determined. Finally, the abnormal state of the robot is monitored according to the synchronous motion parameters, and the cooperative control strategy of the robot is obtained. Through this cooperative work mode, efficient and stable large-area rubber sticking operation is realized. Specifically, the system first obtains the overall length L of the battery cell group in the transverse direction, the width W of a single battery cell, and the spacing D between adjacent battery cells. According to the formula N = (L + D) / (W + D), the number of single-row battery cells N is obtained, wherein the overall length L is the length of the battery cell group in the transverse direction, including the width of N battery cells plus the width of N-1 spacing. Then, according to the principle of N / 2, the left and right two work areas are divided. It can be understood that, in order to ensure the integrity of the grid structure, if the two work areas need to be rubbered at the junction, the rubbering task is completed by one of the rubbering robots. For abnormal situations, such as when a robot needs to pause work, the system automatically recalculates the work area allocation scheme to ensure production continuity.

[0066] S102, extracting feature point information of the battery cell column according to the visual data, obtaining spatial coordinates of representative points, and performing adaptive path fitting calculation based on the spatial coordinates of the representative points to obtain a first direction rubber sticking reference path equation.

[0067] As shown in Figure 3 In the battery assembly process, intermediate rubber sticking is a process of integrating multiple dispersed battery cells into a stable module. Through precise coating of the glue layer, mechanical fixation, thermal management cooperation, and electrical safety protection are simultaneously achieved, preventing damage to the internal structure of the battery cell due to vibration displacement, strengthening the heat conduction efficiency between battery cells with heat-conducting glue, and blocking the risk of short circuit with insulating glue. When executed, the glue line is often arranged in a cross pattern along the transverse and longitudinal directions of the battery cell array, forming a grid or frame type fixing structure.

[0068] Specifically, the path fitting includes feature extraction, representative point selection, and path fitting. First, the edge contour features of the battery cell column are extracted from the visual data obtained by the visual acquisition module to obtain a set of edge points of the battery cell column. Image processing algorithms, such as edge detection algorithms, can be used to complete this task, which can accurately identify the edge contour of the battery cell column. Machine learning algorithms can also be used to automatically extract edge contour features by training models. Then, feature points in the head, middle, and tail regions of the battery cell column are selected from the edge point set to obtain the spatial coordinates of the three representative points. The representative point selection unit can determine the head, middle, and tail regions according to the preset rules, and then select the most representative feature points in these regions. The path fitting uses the least squares method to fit the three representative points to obtain the vertical gluing reference path equation. The least squares method is a commonly used mathematical method that can find an optimal fitting curve that minimizes the sum of squared errors of the representative points to the curve. In addition to the least squares method, other fitting methods such as polynomial fitting can also be used. These three units work together to first obtain the edge point set by the feature extraction unit, then determine the representative points by the representative point selection unit, and finally perform fitting by the path fitting unit to obtain an accurate vertical gluing reference path equation, providing a reliable path reference for subsequent gluing operations.

[0069] Another implementation of path fitting includes the following steps: first, the edge contour features of the battery cell column are extracted from the visual data to obtain a set of edge points of the battery cell column. This step can use Canny and other edge detection algorithms, or use deep learning models to extract edge contour features. Then, the edge point set is analyzed for gray gradient to identify the location information of the battery cell gap. Gray gradient analysis can accurately detect the location of the gap between battery cells by calculating the gray value change rate of adjacent pixel points. Based on the location information of the battery cell gap, the center line of adjacent battery cells is calculated to obtain the spatial coordinates of multiple sets of local reference points. Finally, the multiple sets of local reference points are fitted to obtain the vertical gluing reference path equation. This method obtains more local reference points by identifying the battery cell gap, improving the sampling density of path fitting. Compared with the three-point fitting method, this method can better fit the actual distribution characteristics of the battery cell column, thereby improving the accuracy of the gluing path.

[0070] In one implementation, the system first divides the entire image into multiple feature blocks, each block having a size matching the size of the battery cell. Within each block, the system establishes a local coordinate system and determines the main direction of the battery cell edge by analyzing the directional features of the gray scale distribution. A directional weight matrix is introduced to dynamically adjust the response characteristics of the detection operator according to the edge direction. Higher weight is given to the edge parallel to the main direction, and lower weight is given to the edge perpendicular to the main direction. This directional enhancement strategy significantly improves the accuracy of edge detection. In the battery cell gap identification process, a lower threshold is first used to obtain the complete gap profile, and then a higher threshold is used to extract key feature points. Finally, the two levels of features are connected through a region growing algorithm to avoid the gap breakage or adhesion problems that are prone to occur in traditional single-threshold methods. In another implementation, a multi-scale feature pyramid is constructed to extract edge information at different resolutions. The low-resolution layer is used to obtain the overall trend of the battery cell arrangement, and the high-resolution layer is used to accurately locate local details. Then, in the gap identification stage, a template library is established using the geometric characteristics of the battery cell, and the template matching results are weighted and fused with the gray scale gradient analysis results. Through the multi-source information fusion strategy, the reliability of feature extraction is greatly improved.

[0071] In one embodiment, a laser scanning device is used to obtain the three-dimensional information of the battery cell column. The laser scanning device can quickly and accurately obtain the three-dimensional shape and position information of the battery cell column. It emits a laser beam to irradiate the battery cell column, and calculates the three-dimensional coordinates of the object surface by measuring the time and angle of the reflected light. The laser scanning device includes a laser emitter, a receiver, and a data processing circuit. The laser emitter emits a laser beam, the receiver receives the reflected laser signal, and the data processing circuit processes the signal to obtain the three-dimensional information of the battery cell column. Then, the path fitting is performed using these three-dimensional information, and the vertical taping reference path equation is also obtained. This method can provide more accurate battery cell column information and further improve the accuracy of path fitting.

[0072] S103、According to the vertical taping reference path equation, control signals are generated to control the movement of the taping head.

[0073] Specifically, according to the vertical taping reference path equation obtained by path fitting, control signals are generated to control the movement of the taping head. The control signals are generated by a programmable logic controller (PLC), which converts the reference path equation into the motion parameters of the taping head according to a preset program. The taping robot is driven by the control signals to control the taping head to move according to the vertical taping reference path equation, thereby obtaining the initial taping trajectory.

[0074] S104、Real-time acquisition of position data of the completed taping area, calculation of taping position deviation trend characteristics, dynamic compensation calculation according to the deviation trend characteristics, and obtaining of the corrected taping head motion trajectory.

[0075] Specifically, the system first scans the completed rearward area of the rubberizing based on the visual acquisition device, collects image data of the completed rubberizing area, and extracts the position information of the actual rubberizing line. This step can reuse the aforementioned visual acquisition device, and by scanning the rearward area multiple times, accurate actual rubberizing line position information is obtained. Then the position information of the actual rubberizing line is compared with the planned path to obtain the position deviation. At the same time, the cell profile of the forward area to be rubberized is analyzed based on the visual acquisition device to obtain the position characteristics of the target area. The position deviation and cell profile analysis can be completed using image processing algorithms. Finally, compensation calculation is performed according to the historical change data of the position deviation and the position characteristics of the target area to obtain the corrected motion trajectory of the rubberizing head. The historical change data includes the cumulative deviation amount of the rearward area, the deviation change rate, and the speed and amount of rubber during the rubberizing process. By establishing a compensation mathematical model, the motion trajectory of the rubberizing head is adjusted in real time, thereby ensuring the accuracy of the rubberizing.

[0076] The rearward area refers to the area of the last N rubberizing lines that have been completed, and the value of N ranges from 3 to 5, which is used to ensure data continuity and real-time performance; the cumulative deviation amount is defined as the integral value of the vertical distance between the center of the actual rubberizing line and the center of the planned path, which reflects the overall deviation trend; the deviation change rate represents the change value of the deviation amount per unit time, which is used to predict the deviation development trend; the position characteristics of the target area include the curvature, inclination angle, and relative position relationship of the cell edge profile, which are used to predict possible trajectory deviations. These characteristic parameters constitute the basic data set of the dynamic compensation system.

[0077] In an implementation, the application adopts a hierarchical calculation strategy to process the compensation data. At the bottom layer, the actual trajectory curve y = f(x) of the completed taping line is fitted by the least square method, and the deviation function d(x) = |f(x) - g(x)| from the planned path r = g(x) is calculated. When max{d(x)} exceeds the preset threshold δ1, the compensation calculation is triggered. In the middle layer, the sliding window method is used to calculate the deviation change rate v(t) = Δd(x) / Δt, and when |v(t)| exceeds the threshold δ2, the compensation weight is increased. At the top layer, a prediction model P(x) = h[d(x), v(t), F] is established in combination with the target region features, where F is the target region feature vector, which is used to generate the final compensation amount. In specific implementation, the actual trajectory is fitted by using a cubic spline function: f(x) = ax3+bx2+cx+d, where the coefficients a, b, c, and d are solved by minimizing the sum of squared errors. The deviation change rate is calculated by using the five-point difference method: v(t) = [d(t+2) - 8d(t+1) + 8d(t-1) - d(t-2)] / 12Δt. In another implementation, the application adopts a frequency domain analysis method based on Fourier transform. First, the fast Fourier transform is performed on the deviation function d(x) to obtain the frequency spectrum component D(ω). By analyzing the amplitude distribution of D(ω), periodic deviation and random deviation can be identified. When the amplitude of a certain frequency component exceeds the threshold δ3, it indicates that there is a significant periodic deviation, and harmonic compensation is needed. At the same time, the wavelet transform is used to perform multi-scale decomposition on the deviation signal to obtain the deviation features of different frequency bands. On this basis, an adaptive filter H(ω) is established, which dynamically adjusts according to the spectral features: H(ω) = α·D(ω) + β·W(ω), where W(ω) is the wavelet coefficient, and α and β are adaptive weight coefficients. The time domain compensation amount is obtained by inverse Fourier transform.

[0078] S105, based on the corrected taping trajectory, a plurality of second direction taping paths are planned according to an adaptive spacing.

[0079] Specifically, the system first extracts the edge contour features of each vertical taping line based on the corrected taping trajectory using an image processing algorithm. Then, according to the edge contour features, the intersection positions of each horizontal taping path with all vertical taping lines are determined between adjacent two vertical taping lines according to an adaptive spacing. The adaptive spacing can be determined according to the actual spacing between adjacent vertical taping lines and the width parameter of the vertical taping line, or according to the actual spacing of the cell column and the taping process parameters. Finally, the target point positions of each horizontal taping are obtained based on the intersection positions.

[0080] There is another implementation of the transverse path planning. First, based on the corrected adhesive trajectory, the gray scale gradient of the vertical adhesive line is analyzed to obtain the center line position and the transverse width distribution characteristics of each adhesive line. Then, according to the center line position and the width distribution characteristics, the intersection positions of each transverse adhesive path and all vertical adhesive lines are determined at an adaptive interval between adjacent two vertical adhesive lines. The adaptive interval is determined according to the actual interval between adjacent vertical adhesive lines, the width parameters of the vertical adhesive lines, and the cross region adhesive amount distribution characteristics. Finally, the transverse adhesive target trajectory is obtained based on the intersection positions and the cross region characteristics. The two methods can plan appropriate transverse adhesive paths according to different situations and improve the overall quality of the grid-shaped adhesive structure.

[0081] The center line position refers to the symmetry axis of the vertical adhesive line in the transverse direction, which is obtained by analyzing the center position of the gray scale distribution of the adhesive line. The transverse width distribution characteristics include the average value and the degree of change of the adhesive line width, which are used to represent the transverse uniformity of the adhesive line. The cross region adhesive amount distribution characteristics include the adhesive layer thickness at the intersection point and the adhesive diffusion range, which are used to evaluate the bonding strength of the intersection point. The adaptive interval refers to the transverse adhesive line interval dynamically adjusted according to the process requirements, which needs to consider the battery size and heat dissipation requirements in the value range.

[0082] The present application determines the transverse adhesive path by using a multi-objective optimization method. The optimization process considers the interval uniformity, adhesive layer thickness, and bonding strength, which are given corresponding weights according to different importance. The transverse adhesive line interval must be kept within the allowed range, the adhesive layer thickness must meet the process requirements, and the adhesive diffusion range needs to reach the minimum coverage area. When a transverse adhesive line conflicts with any constraint condition, the system automatically adjusts the position parameters of the adjacent path until the best layout scheme is found. In specific implementation, an iterative optimization method is used to find the optimal solution that meets all requirements through multiple rounds of calculation. In this embodiment, the range of transverse adhesive interval is determined by the heat dissipation requirement of the battery; the adhesive layer thickness range is based on the comprehensive consideration of bonding strength and material cost.

[0083] In the above embodiment, the application based on the visual guidance of the middle cell gluing method, first from the acquisition of cell column image information, gradually path fitting, trajectory generation, dynamic compensation and lateral path planning, while using double robot collaborative work scheme. This method makes full use of the advantages of visual guidance and dynamic compensation, effectively solves the problem of gluing deviation caused by process parameter fluctuation in traditional gluing method. The whole gluing process is executed according to the clear logical sequence, which can adjust the gluing path in real time according to the actual situation, and ensure the uniformity and consistency of the gluing line. Through multi-angle visual acquisition and real-time compensation, the gluing quality of the intersection area is improved; using double robot collaborative work improves the gluing efficiency, realizes efficient and stable large-area gluing operation. Compared with the traditional method, the application has significant improvement in gluing precision and reliability, and can better meet the high quality requirements of modern lithium battery production for middle cell gluing.

[0084] The system in the embodiment of the application will be described from the perspective of hardware processing. Please refer to Figure 4 , which is a schematic diagram of the entity device structure of a middle cell gluing system based on visual guidance provided by the embodiment of the application.

[0085] It should be noted that, Figure 4 The structure of the system shown is only an example, and should not bring any limitation to the function and use range of the embodiment of the application.

[0086] As Figure 4 shown, the system includes a CPU 401, which can perform various appropriate actions and processes according to the program stored in the read-only memory ROM 402 or the program loaded into the random access memory RAM 403 from the storage part 408, such as performing the method in the above embodiment. In the RAM 403, various programs and data required for system operation are also stored. The CPU 401, the ROM 402 and the RAM 403 are connected to each other through a bus 404. The I / O interface 405 is also connected to the bus 404.

[0087] The following components are connected to the I / O interface 405: an input section 406 including a camera, a microphone, and the like; an output section 407 including a liquid crystal display (LCD), a speaker, and the like; a storage section 408 including a hard disk and the like; and a communication section 409 including a network interface card such as a LAN (Local Area Network) card, a modem, and the like. The communication section 409 performs a communication process via a network such as the Internet. A drive 410 is also connected to the I / O interface 405 as necessary. A removable medium 411 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, and the like is attached to the drive 410 as necessary, so that a computer program read out therefrom is installed in the storage section 408 as necessary.

[0088] In particular, the processes described above with reference to the flowcharts can be implemented as a computer software program according to embodiments of the present application. For example, embodiments of the present application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing a computer program for executing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network by the communication section 409, and / or from the removable medium 411. When the computer program is executed by the CPU 401, various functions defined in the present application are executed.

[0089] It should be noted that the computer-readable medium in the embodiments of the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. The computer-readable storage medium may, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination of the above. More specific examples of the computer-readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (Compact Disc Read-Only Memory, CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or apparatus. In the present application, the computer-readable signal medium can include a data signal carrying computer-readable computer programs in a baseband or as a part of a carrier wave. Such a propagated data signal can take various forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above.

[0090] The flowcharts and block diagrams in the drawings show the possible implementation architectures, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In the flowcharts or block diagrams, each block can represent a module, a program segment or a part of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur in different orders than that shown in the drawings. For example, two blocks represented in succession can actually be executed substantially in parallel, and sometimes in reverse order, depending on the functions involved. It should also be noted that each block in the block diagram or flowchart, and the combination of blocks in the block diagram or flowchart, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0091] As another aspect, the present application also provides a computer readable storage medium, which can be included in the system described in the above embodiments, or can exist independently without being assembled into the system. The above storage medium carries one or more computer programs, which, when executed by a processor of a system, enable the system to implement the method provided in the above embodiments.

[0092] The above embodiments are only used to illustrate the technical solutions of the present application, but not limit the present application; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

[0093] In the above embodiments, according to the context, the term "when" can be interpreted as "if" or "after" or "in response to determining" or "in response to detecting". Similarly, according to the context, the phrase "upon determining" or "if detecting (the stated condition or event)" can be interpreted as "if determining" or "in response to determining" or "upon detecting (the stated condition or event)" or "in response to detecting (the stated condition or event)".

[0094] In the above embodiments, all or part of the embodiments can be realized by software, hardware, firmware or any combination thereof. When realized by software, all or part of the embodiments can be realized in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through wired (such as coaxial cable, optical fiber, digital subscriber line) or wireless (such as infrared, wireless, microwave, etc.) mode. The computer readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. integrated with one or more available media. The available media can be magnetic media (for example, floppy disk, hard disk, magnetic tape), optical media (for example, DVD), or semiconductor media (for example, solid state disk) and the like.

[0095] Those skilled in the art can understand that all or part of the processes in the above-mentioned method embodiments can be implemented by a computer program instructing relevant hardware to complete, the program can be stored in a computer readable storage medium, and the program can include the processes of the above-mentioned method embodiments when executed. The aforementioned storage medium includes ROM or random storage memory RAM, magnetic disc or optical disc and various storage code medium.

Claims

1. A method for bonding adhesive between battery cells based on visual guidance, characterized in that, include: Visual data of the battery cells is obtained by acquiring image information of the battery cell array using a visual acquisition device. Based on the visual data, feature point information of the battery cell column is extracted to obtain the spatial coordinates of the representative points. Based on the spatial coordinates of the representative points, adaptive path fitting calculation is performed to obtain the first direction adhesive bonding reference path equation. The adhesive application head movement is controlled according to the first direction adhesive application reference path equation to obtain the initial adhesive application trajectory; Real-time acquisition of position data of the completed adhesive application area, calculation of adhesive application position deviation trend characteristics, dynamic compensation calculation based on the deviation trend characteristics, and obtaining the corrected adhesive application head movement trajectory; Based on the corrected adhesive application trajectory, multiple second-direction adhesive application paths are planned according to adaptive spacing, wherein the second direction is perpendicular to the first direction; Specifically, the real-time acquisition of position data of the completed adhesive application area, calculation of adhesive application position deviation trend characteristics, and dynamic compensation calculation based on the deviation trend characteristics to obtain the corrected adhesive application head movement trajectory include: The visual acquisition device scans the rear area where the adhesive has been applied, acquires image data of the area where the adhesive has been applied, and extracts the position information of the actual adhesive line. The actual adhesive application line position information is compared with the planned path to obtain the position deviation, and the cell outline of the forward adhesive application area is analyzed based on the vision acquisition device to obtain the position characteristics of the target area. Compensation calculations are performed based on the historical change data of the positional deviation and the positional characteristics of the target area to obtain the corrected trajectory of the adhesive applicator. The historical change data includes the cumulative deviation in the backward region, the rate of deviation change, and the speed and amount of adhesive during the application process.

2. The method according to claim 1, characterized in that, The step of extracting feature point information of the cell array based on the visual data to obtain the spatial coordinates of representative points, and performing adaptive path fitting calculation based on the spatial coordinates of the representative points to obtain the first direction adhesive bonding reference path equation, specifically includes: Extract the edge contour features of the battery cell column from the visual data to obtain the set of edge points of the battery cell column; Feature points from the head, middle and tail regions of the cell array are selected from the edge point set to obtain the spatial coordinates of three representative points; The least squares method was used to fit the three representative points to obtain the first direction adhesive reference path equation.

3. The method according to claim 1, characterized in that, The step of extracting feature point information of the cell array based on the visual data to obtain the spatial coordinates of representative points, and performing adaptive path fitting calculation based on the spatial coordinates of the representative points to obtain the first direction adhesive bonding reference path equation, specifically includes: Extract the edge contour features of the battery cell column from the visual data to obtain the set of edge points of the battery cell column; Gray-scale gradient analysis is performed on the edge point set to identify the position information of the cell gaps. Based on the position information of the cell gaps, the center line connecting adjacent cells is calculated to obtain the spatial coordinates of multiple sets of local reference points. By fitting the multiple sets of local reference points, the first direction adhesive application reference path equation is obtained.

4. The method according to claim 2, characterized in that, The step of planning multiple second-direction adhesive application paths based on the corrected adhesive application trajectory and according to adaptive spacing specifically includes: Based on the corrected adhesive application trajectory, the edge contour features of each adhesive application line in the first direction are extracted; Based on the edge contour features, the intersection position of each second direction adhesive path with all first direction adhesive lines is determined according to an adaptive spacing between two adjacent first direction adhesive lines. The adaptive spacing is determined based on the actual spacing between adjacent first direction adhesive lines and the width parameter of the first direction adhesive line, or the adaptive spacing is determined based on the actual spacing of the cell array and the adhesive application process parameters. The target points for applying adhesive in each second direction are obtained based on the intersection points.

5. The method according to claim 3, characterized in that, The step of planning multiple second-direction adhesive application paths based on the corrected adhesive application trajectory and according to adaptive spacing specifically includes: Based on the corrected adhesive trajectory, grayscale gradient analysis is performed on the adhesive lines in the first direction to obtain the centerline position of each adhesive line and the width distribution characteristics in the second direction. Based on the centerline position and width distribution characteristics, the intersection position of each second-direction adhesive path with all first-direction adhesive lines is determined according to an adaptive spacing between two adjacent first-direction adhesive lines. The adaptive spacing is determined based on the actual spacing between adjacent first-direction adhesive lines, the width parameter of the first-direction adhesive lines, and the adhesive amount distribution characteristics of the intersection area. The second-direction adhesive target trajectory is obtained based on the intersection point location and the characteristics of the intersection area.

6. The method according to claim 1, characterized in that, The method further includes: The adhesive application area is divided into two regions based on the cell array size, resulting in the corresponding working areas for the two adhesive application robots. Based on the motion parameters of the two adhesive application robots configured in the work area, synchronous adhesive application operation control commands for the two robots are obtained. A real-time status monitoring mechanism for the two adhesive application robots is established based on the control instructions to obtain the position information and adhesive application status data of the two adhesive application robots. Based on the adhesive application status data, calculate the motion speed compensation amount of the two adhesive application robots and determine the dynamically adjusted synchronous motion parameters. Based on the synchronous motion parameters, abnormal states of the robotic arm are monitored to obtain a collaborative control strategy for the robotic arm.

7. A vision-guided interlayer adhesive bonding system for battery cells, characterized in that, The system includes: One or more processors and a memory; the memory is coupled to the one or more processors, the memory being used to store computer program code, the computer program code including computer instructions, the one or more processors invoking the computer instructions to cause the system to perform the method as described in any one of claims 1-6.

8. A computer-readable storage medium comprising instructions, characterized in that, When the instructions are executed on the system, the system performs the method as described in any one of claims 1-6.

9. A computer program product, characterized in that, When the computer program product is run on the system, the system performs the method as described in any one of claims 1-6.

Images