Butt joint guiding method and device of socket assembling machine

Abstract

The invention provides a butt joint guiding method and device for a socket assembling machine, is applied to the technical field of automation, and obtains the residual position error of a to-be-assembled socket through building an error attenuation model. And outputting the position error of the pre-positioned socket through an error attenuation model, and providing error data for subsequent steps. And then, according to the position error of the socket after pre-positioning, the time required by the visual servo system for fine adjustment is determined. Thirdly, combining the position error of the pre-positioned socket with the visual fine adjustment time, and constructing a beat time model of single socket assembly; the model can reflect the time composition required by single assembly. And finally, according to the established beat time model, the pre-positioning time of the conical positioning mechanism is adjusted. By adjusting the pre-positioning time and optimizing the pre-positioning effect, the subsequent visual fine adjustment time is further influenced, and finally the conical positioning mechanism is controlled to pre-position the socket to be assembled according to the adjusted pre-positioning time, so that the effects of guiding the assembly and butt joint of the socket and improving the assembly and butt joint speed are achieved.

Description

Technical Field

[0001] This application relates to the field of automation technology, and particularly to a docking guidance method and device for a socket assembly machine. Background Art

[0002] In a modern high-speed production workshop for electronic products, the socket assembly link is a key bottleneck in the product manufacturing process, directly affecting the overall production efficiency. To meet the growing production capacity requirements and market delivery pressure, the production line continuously pursues a shorter production cycle. Traditionally, visual guidance docking technology is widely used in socket assembly, which uses a vision system to accurately identify the position information of the socket and the connector, and guides the robotic arm to complete the precise docking operation. However, in the actual application scenario of high-speed production, the processing speed of the traditional visual guidance method has gradually become the key bottleneck restricting the further improvement of production efficiency.

[0003] Specifically, the visual guidance docking process usually includes multiple links such as rapid image acquisition, efficient image information processing, accurate calculation of position deviation, and timely response of the servo system. Each link takes a certain amount of time. Especially on a high-speed production line that pursues an extreme cycle, the time accumulated by these links becomes lengthy and it is difficult to fully meet the production requirements of an extremely short cycle. Although the visual guidance speed can be increased by continuously improving the hardware performance of the vision system and continuously optimizing the image processing algorithm, there are always technical bottlenecks and objective cost limitations. On the other hand, if a high-precision mechanical device is completely used to achieve the rapid docking of the socket, it often requires the investment of very expensive customized equipment, and may require a large-scale transformation and upgrade of the existing production line, and its economic cost and actual implementation difficulty are very high.

[0004] For a high-speed production workshop of electronic products that has widely deployed a visual guidance system, how to economically and efficiently break through the speed bottleneck of the existing visual guidance docking method and achieve an effective speed increase in the socket assembly link without significantly increasing additional costs and without substantially replacing existing equipment has become a key technical problem to be solved urgently. Especially in a cost-sensitive high-speed production environment, it is crucial to seek a solution with low cost, easy to quickly integrate, and capable of significantly improving the docking speed.

[0005] In view of the above problems, the existing technology urgently needs to be improved. Summary of the Invention

[0006] In view of the deficiencies of the above-mentioned prior art, this application provides a docking guidance method and device for a socket assembly machine, which has the beneficial effect of improving the docking speed of socket assembly.

[0007] First aspect, a docking guidance method for a socket assembly machine, which is used for the socket assembly link with vision guidance in a high-speed production workshop of electronic products to improve the socket assembly docking speed. The method includes the following steps:

[0008] S1: Establish an error attenuation model, and output the position error of the socket after preliminary positioning according to the error attenuation model;

[0009] S2: Obtain the vision fine adjustment time according to the position error;

[0010] S3: Construct a beat time model for a single socket assembly based on the position error and the vision fine adjustment time;

[0011] S4: According to the beat time model, adjust the preliminary positioning time, and control the conical positioning mechanism to perform preliminary positioning on the socket to be assembled according to the adjusted preliminary positioning time, so as to guide the socket assembly docking.

[0012] Further, step S1 includes:

[0013] S11: Obtain the geometric parameters of the conical positioning mechanism, the initial position error of the socket to be assembled, and the range of preliminary positioning time;

[0014] S12: Establish an error attenuation model according to the geometric parameters of the conical positioning mechanism;

[0015] S13: Select multiple preliminary positioning time points at a preset time interval within the range of preliminary positioning time, and calculate the position error of the first socket after preliminary positioning corresponding to each preliminary positioning time point according to the error attenuation model;

[0016] S14: Establish a mapping relationship between the preliminary positioning time and the position error attenuation effect according to the initial position error of the socket to be assembled and the position error of the first socket after preliminary positioning, and output the position error of the socket after preliminary positioning according to the mapping relationship.

[0017] Further, step S12 includes:

[0018] S121: Establish an initial error attenuation model according to the geometric parameters of the conical positioning mechanism;

[0019] S121: Collect the historical wear data of the conical positioning mechanism, the vibration data of the robotic arm, the workshop ambient temperature data, and the manufacturing error data of the socket to be assembled;

[0020] S122: Modify the initial error attenuation model according to the historical wear data of the conical positioning mechanism, the vibration data of the robotic arm, the workshop ambient temperature data, and the manufacturing error data of the socket to be assembled to obtain the error attenuation model.

[0021] Further, step S122 includes:

[0022] S1221: respectively establish a wear model of the conical positioning mechanism, a vibration model of the robotic arm, a temperature compensation model, and a manufacturing error model of the socket to be assembled according to the historical wear data of the conical positioning mechanism, the vibration data of the robotic arm, the workshop ambient temperature data, and the manufacturing error data of the socket to be assembled;

[0023] S1222: based on the wear model of the conical positioning mechanism, the vibration model of the robotic arm, the temperature compensation model, and the manufacturing error model of the socket, use a weighted fusion algorithm to correct the initial error attenuation model to obtain the corrected error attenuation model.

[0024] Further, step S1222 includes:

[0025] S12221: determine the influence factors of each model on the pre-positioning accuracy;

[0026] S12222: based on the influence factors, use the analytic hierarchy process to determine the weights of each model;

[0027] S12223: according to the weights of each model, use the weighted average method to correct the initial error attenuation model to obtain the corrected error attenuation model.

[0028] Further, step S2 includes:

[0029] S21: obtain the performance parameters of the visual servo system, and establish a mapping function between the position error and the visual adjustment time according to the position error of the socket after pre-positioning and the performance parameters of the visual servo system;

[0030] S22: calculate the initial visual fine adjustment time corresponding to the position error according to the mapping function;

[0031] S23: obtain the workshop ambient light intensity, and establish a correction model of the light intensity on the visual adjustment time according to the relationship between the light intensity and the visual recognition accuracy, and correct the initial visual fine adjustment time according to the correction model to obtain the visual fine adjustment time.

[0032] Further, step S3 includes:

[0033] S31: obtain the pre-positioning time;

[0034] S32: according to the pre-positioning time, query the pre-established corresponding relationship between the pre-positioning time and the pre-positioning error attenuation amount to obtain the pre-positioning error attenuation amount corresponding to the pre-positioning time;

[0035] S33: Modify the visual fine adjustment time according to the pre-positioning error attenuation amount to obtain the modified visual fine adjustment time;

[0036] S34: Construct a tact time model for single socket assembly according to the pre-positioning time and the modified visual fine adjustment time.

[0037] Further, step S34 includes:

[0038] S341: According to the pre-positioning time and the modified visual fine adjustment time, use the time series analysis method to predict the tact time of the next N single socket assemblies, where N is an integer greater than 1;

[0039] S342: Calculate the variance of the tact time according to the predicted tact time of the next N single socket assemblies, and determine whether the variance exceeds a preset threshold;

[0040] S343: If the variance exceeds the preset threshold, adjust the pre-positioning time and re-execute steps S31 to S34 to reduce the volatility of the tact time; if the variance does not exceed the preset threshold, sum up the pre-positioning time and the modified visual fine adjustment time to construct a tact time model for single socket assembly.

[0041] Further, step S4 includes:

[0042] S41: Determine the pre-positioning time adjustment range according to the tact time model;

[0043] S42: Select multiple adjusted pre-positioning times within the pre-positioning time adjustment range, and calculate the predicted tact time corresponding to each adjusted pre-positioning time according to the tact time model;

[0044] S43: Select the optimal pre-positioning time by using an optimization algorithm, where the optimization algorithm aims to minimize the tact time;

[0045] S44: Control the conical positioning mechanism to pre-position the socket to be assembled according to the optimal pre-positioning time to guide the socket assembly docking.

[0046] In a second aspect, a docking guiding device for a socket assembly machine is applied in the steps of the docking guiding method for a socket assembly machine described in any one of the above. The device includes:

[0047] Error attenuation model construction module: used to establish an error attenuation model and output the position error of the socket after pre-positioning according to the error attenuation model;

[0048] Visual adjustment time acquisition module: used to obtain the visual fine adjustment time according to the position error;

[0049] Beat time model construction module: used to construct a beat time model for single socket assembly based on the position error and the visual fine adjustment time;

[0050] Guided assembly docking module: used to adjust the pre-positioning time according to the beat time model, and control the conical positioning mechanism to pre-position the socket to be assembled according to the adjusted pre-positioning time, so as to guide the socket assembly docking.

[0051] Beneficial effects: A docking guidance method and device for a socket assembly machine proposed in this application. The method first establishes an error attenuation model, which is used to predict the remaining position error of the socket to be assembled after the pre-positioning of the conical positioning mechanism. The position error of the socket after pre-positioning is output through the error attenuation model, providing error data for subsequent steps. Then, the method determines the time required for the visual servo system to perform fine adjustment according to the position error of the socket after pre-positioning. The smaller the position error, the shorter the visual fine adjustment time. Next, the method combines the position error of the socket after pre-positioning and the visual fine adjustment time to construct a beat time model for single socket assembly. This model can reflect the time composition required for single assembly. Finally, the method adjusts the pre-positioning time of the conical positioning mechanism according to the established beat time model. By adjusting the pre-positioning time, the pre-positioning effect is optimized, which in turn affects the subsequent visual fine adjustment time. Finally, the conical positioning mechanism is controlled to pre-position the socket to be assembled according to the adjusted pre-positioning time, realizing the guidance of socket assembly docking and achieving the purpose of improving the assembly docking speed. Description of the Drawings

[0052] Figure 1 It is a flowchart of a docking guidance method for a socket assembly machine proposed in this application.

[0053] Figure 2 It is a structural diagram of a docking guidance device for a socket assembly machine proposed in this application.

[0054] Label description: 201, error attenuation model construction module; 202, visual adjustment time acquisition module; 203, beat time model construction module; 204, guided assembly docking module. Detailed Embodiments

[0055] Next, in combination with the accompanying drawings in the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Usually, the components of the embodiments of the present application described and marked in the accompanying drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed present application, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present application.

[0056] It should be noted that similar reference numerals and letters denote similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of the present application, terms such as "first" and "second" are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0057] Please refer to Figure 1 , in a first aspect, a docking guidance method for a socket assembly machine, which is used for the socket assembly link with visual guidance in a high-speed production workshop of electronic products to improve the docking speed of socket assembly. The method includes the steps:

[0058] S1: Establish an error attenuation model and output the position error of the socket after pre-positioning according to the error attenuation model;

[0059] S2: Obtain the visual fine adjustment time according to the position error;

[0060] S3: Construct a beat time model for a single socket assembly based on the position error and the visual fine adjustment time;

[0061] S4: According to the beat time model, adjust the pre-positioning time, and control the conical positioning mechanism to pre-position the socket to be assembled according to the adjusted pre-positioning time to guide the socket assembly docking.

[0062] Among them, in step S1, for the docking guidance method of the socket assembly machine, an error attenuation model is established to predict the remaining position error of the socket to be assembled after the conical positioning mechanism has been positioned. The error attenuation model can be constructed based on information such as the geometric parameters of the conical positioning mechanism, the initial position error of the socket to be assembled, and the pre-positioning time range. For example, the geometric parameters can include the cone angle and cone surface length of the conical positioning mechanism; the initial position error can be obtained by taking pictures and identifying the socket to be assembled by the vision system before pre-positioning; the pre-positioning time range can be preset according to the production beat requirements and the motion performance of the mechanism. After the model is established, the position error of the socket after pre-positioning can be output through the error attenuation model according to the set pre-positioning time.

[0063] In step S2, the visual fine-tuning time is obtained according to the position error of the socket after pre-positioning. Performance parameters of the visual servo system, such as visual recognition accuracy, image processing speed, servo motor response time, etc., can be obtained in advance. A mapping function between the position error and the visual adjustment time can be established based on the performance parameters of the visual servo system. For example, the larger the position error, the longer the required visual adjustment time usually is. The mapping function can take forms such as linear function, non-linear function or look-up table. Through the mapping function, the initial visual fine-tuning time corresponding to the position error can be calculated. Further, the illumination intensity of the workshop environment can be obtained, and the influence of the illumination intensity on the visual recognition accuracy can be considered. A correction model for the influence of the illumination intensity on the visual adjustment time can be established according to the relationship between the illumination intensity and the visual recognition accuracy. For example, too low or too high illumination intensity may reduce the visual recognition accuracy, thus prolonging the visual adjustment time. Through the correction model, the initial visual fine-tuning time can be corrected to obtain the final visual fine-tuning time.

[0064] In step S3, the cycle time model for single socket assembly is constructed based on the position error and the visual fine-tuning time. The corresponding relationship between the pre-positioning time and the pre-positioning error attenuation amount can be established in advance. For example, the longer the pre-positioning time, the larger the pre-positioning error attenuation amount usually is, but at the same time, the pre-positioning process takes longer. The corresponding relationship can be obtained through experimental data, simulation analysis or theoretical calculation. According to the pre-positioning time, the pre-positioning error attenuation amount corresponding to the pre-positioning time can be queried. The visual fine-tuning time can be corrected according to the pre-positioning error attenuation amount. For example, the larger the pre-positioning error attenuation amount, the better the pre-positioning effect, and the visual fine-tuning time can be shortened accordingly. The corrected visual fine-tuning time and the pre-positioning time can be summed up to construct the cycle time model for single socket assembly. The cycle time model can be expressed as the sum of the pre-positioning time and the corrected visual fine-tuning time.

[0065] In step S4, the pre-positioning time is adjusted according to the beat time model, and the conical positioning mechanism pre-positions the socket to be assembled according to the adjusted pre-positioning time to guide the socket assembly and docking. The adjustment range of the pre-positioning time can be determined according to the beat time model. For example, the adjustment range of the pre-positioning time can be set to a range where the change in the pre-positioning time in the beat time model has a relatively small impact on the beat time. Within the adjustment range of the pre-positioning time, multiple adjusted pre-positioning times can be selected. The predicted beat time corresponding to each adjusted pre-positioning time can be calculated according to the beat time model. An optimization algorithm can be adopted to select the optimal pre-positioning time according to the predicted beat time with the goal of minimizing the beat time. The optimization algorithm can be a traversal algorithm, a gradient descent algorithm, a genetic algorithm, etc. After the optimal pre-positioning time is determined, the conical positioning mechanism is controlled to pre-position the socket to be assembled according to the optimal pre-positioning time, thereby guiding the socket assembly and docking.

[0066] Further, step S1 includes:

[0067] S11: Obtain the geometric parameters of the conical positioning mechanism, the initial position error of the socket to be assembled, and the pre-positioning time range;

[0068] S12: Establish an error attenuation model according to the geometric parameters of the conical positioning mechanism;

[0069] S13: Select multiple pre-positioning time points at a preset time interval within the pre-positioning time range, and calculate the position error of the socket after the first pre-positioning corresponding to each pre-positioning time point according to the error attenuation model;

[0070] S14: Establish a mapping relationship between the pre-positioning time and the position error attenuation effect according to the initial position error of the socket to be assembled and the position error of the socket after the first pre-positioning, and output the position error of the socket after the pre-positioning according to the mapping relationship.

[0071] Among them, in step S11, the geometric parameters of the conical positioning mechanism may include structural dimension parameters such as the cone angle, cone surface height, and bottom diameter of the conical positioning mechanism, and these parameters can characterize the shape characteristics of the conical positioning mechanism.

[0072] The initial position error of the socket to be assembled can be obtained by photographing the socket to be assembled before pre-positioning using a vision system and analyzing it through an image processing algorithm. The pre-positioning time range can be preset according to the beat requirements of the actual production line and the performance parameters of the conical positioning mechanism, for example, set to 0.1 second to 0.5 second.

[0073] In step S12, the error attenuation model can be characterized by a mathematical formula. For example, it can be assumed that the error attenuation has an exponential relationship with the pre-positioning time, and the following initial error attenuation model is established: Position error = Initial position error * exp(-k * Pre-positioning time), where k is the attenuation coefficient related to the geometric parameters of the conical positioning mechanism and can be obtained through experimental calibration or simulation analysis.

[0074] In step S13, the preset time interval can be set according to the pre-positioning time range and the calculation accuracy requirement. For example, if the pre-positioning time range is from 0.1 second to 0.5 second, the time interval can be set to 0.05 second, so as to select multiple pre-positioning time points such as 0.1 second, 0.15 second, 0.2 second, 0.25 second, 0.3 second, 0.35 second, 0.4 second, 0.45 second, 0.5 second, etc. For each selected pre-positioning time point, substituting it into the error attenuation model established in step S12, the position error of the corresponding first post-pre-positioning socket can be calculated.

[0075] In step S14, the mapping relationship can be constructed in various forms such as a table, curve fitting, or neural network. For example, a two-dimensional table can be established, where the rows of the table represent different pre-positioning time points and the columns represent the corresponding position errors of the first post-pre-positioning socket. By looking up the table, the corresponding position error can be quickly obtained according to the given pre-positioning time. As a preferred implementation manner, the curve fitting method can be adopted. Based on the data pairs of multiple first pre-positioning time points and position errors calculated in step S13, a curve between the first pre-positioning time and the position error is fitted, and using this curve function relationship as the mapping relationship can more accurately output the position error of the post-pre-positioning socket.

[0076] Furthermore, step S12 includes:

[0077] S121: Establish an initial error attenuation model according to the geometric parameters of the conical positioning mechanism;

[0078] S121: Collect historical wear data of the conical positioning mechanism, vibration data of the robotic arm, workshop ambient temperature data, and manufacturing error data of the socket to be assembled;

[0079] S122: Modify the initial error attenuation model according to the historical wear data of the conical positioning mechanism, vibration data of the robotic arm, workshop ambient temperature data, and manufacturing error data of the socket to be assembled to obtain the error attenuation model.

[0080] Among them, in step S121, the initial error attenuation model can be established in the following way: Assume that the conical positioning mechanism has an ideal geometric shape, such as a right circular cone, and model the mathematical relationship of the position error attenuation of the socket to be assembled by analyzing the geometric structure of the right circular cone.

[0081] In step S121, the historical wear data can be obtained from the maintenance records of the conical positioning mechanism, the vibration data can be collected by installing an acceleration sensor on the robotic arm, the workshop ambient temperature data can be collected by arranging temperature sensors in the workshop environment, and the manufacturing error data can be obtained from the quality inspection report of the socket to be assembled.

[0082] In step S122, the initial error attenuation model can be corrected by using the weighted average method. According to the influence degrees of the historical wear data, vibration data, workshop ambient temperature data, and manufacturing error data on the error attenuation model respectively, different weights are assigned to perform weighted correction on the initial error attenuation model.

[0083] Furthermore, step S122 includes:

[0084] S1221: Respectively establish a wear model of the conical positioning mechanism, a vibration model of the robotic arm, a temperature compensation model, and a manufacturing error model of the socket according to the historical wear data of the conical positioning mechanism, the vibration data of the robotic arm, the workshop ambient temperature data, and the manufacturing error data of the socket to be assembled;

[0085] S1222: Based on the wear model of the conical positioning mechanism, the vibration model of the robotic arm, the temperature compensation model, and the manufacturing error model of the socket, use a weighted fusion algorithm to correct the initial error attenuation model to obtain a corrected error attenuation model.

[0086] Among them, in step S1221, the wear model of the conical positioning mechanism can be established as a function of the usage time. For example, it can be assumed that the wear amount is proportional to the usage time. By recording the historical replacement cycle of the conical positioning mechanism and the wear degree before each replacement, a mathematical model of wear changing with time is fitted. The vibration model of the robotic arm can be established based on the vibration sensor data installed on the robotic arm. By analyzing the vibration frequency and amplitude collected by the sensor, a relationship model between the vibration parameters and the position deviation at the end of the robotic arm is established. The workshop ambient temperature data can be collected in real time by temperature sensors, and the temperature compensation model can be established as a relationship model between the temperature change and the position error caused by the thermal expansion and contraction of components. For example, linear or nonlinear regression methods can be used to fit the functional relationship between the temperature change and the position error. The manufacturing error data of the socket to be assembled can be obtained from the production management system, and the manufacturing error model can be established based on statistical analysis methods. For example, the mean and variance of the size deviation of the sockets in the same batch can be calculated, and a statistical distribution model is used to describe the distribution law of the manufacturing errors.

[0087] In step S1222, a weighted fusion algorithm is adopted to integrate the above-mentioned various models. As a preferred implementation, the analytic hierarchy process can be used to determine the weights of the models. First, the influencing factors of each model on the pre-positioning accuracy are determined. For example, the influencing factor of the wear model can be evaluated based on the impact of the wear degree of the conical positioning mechanism on the positioning accuracy. Then, based on the influencing factors, the analytic hierarchy process is used to construct a judgment matrix, and by calculating the eigenvector of the judgment matrix, the weights of the models are obtained. After the weights are determined, the weighted average method can be used to correct the initial error attenuation model. Specifically, the error attenuation amounts predicted by each model can be multiplied by the corresponding weights and then summed to obtain a comprehensive error correction amount. By superimposing the initial error attenuation model and the error correction amount, a corrected error attenuation model is obtained.

[0088] Furthermore, step S1222 includes:

[0089] S12221: Determine the influencing factors of each model on the pre-positioning accuracy;

[0090] S12222: Based on the influencing factors, use the analytic hierarchy process to determine the weights of the models;

[0091] S12223: According to the weights of the models, use the weighted average method to correct the initial error attenuation model to obtain a corrected error attenuation model.

[0092] Among them, in step S12221, the influencing factors of each model on the pre-positioning accuracy are determined. Specifically, the process of determining the influencing factors can include experimental tests, data analysis, or expert evaluation. For example, to determine the influencing factor of the wear model of the conical positioning mechanism on the pre-positioning accuracy, the pre-positioning error data of the conical positioning mechanism at different wear degrees can be collected, and by analyzing the variation law of the pre-positioning error with the wear degree, the influencing factor of the wear model is obtained. Similarly, the influencing factors of the robotic arm vibration model, the temperature compensation model, and the socket manufacturing error model can also be determined by experimental or data analysis methods.

[0093] In step S12222, the weights of the models are determined by the analytic hierarchy process based on the influencing factors determined in step S12221. The analytic hierarchy process is a structured method for dealing with complex decision-making problems. In this step, the pre-positioning accuracy can be regarded as the target layer, and each model (wear model, vibration model, temperature compensation model, manufacturing error model) can be regarded as the criterion layer. By constructing a judgment matrix, the importance of each model relative to the pre-positioning accuracy is compared pairwise and a consistency test is performed, and the weights of the models are calculated. The weights reflect the relative importance of each model in the correction of the error attenuation model.

[0094] In step S12223, the initial error decay model is corrected using the weighted average method according to the model weights determined in step S12222. Specifically, the corrected error decay model can be the result of the weighted sum of the initial error decay model and each model. The weight of each model is used as the coefficient of its corresponding term in the weighted sum, thereby realizing the correction of the initial error decay model.

[0095] Further, step S2 includes:

[0096] S21: Obtain the performance parameters of the visual servo system. According to the position error of the socket after pre-positioning and the performance parameters of the visual servo system, establish a mapping function between the position error and the visual adjustment time;

[0097] S22: Calculate the initial visual fine adjustment time corresponding to the position error according to the mapping function;

[0098] S23: Obtain the illumination intensity of the workshop environment, and according to the relationship between the illumination intensity and the visual recognition accuracy, establish a correction model of the illumination intensity on the visual adjustment time, and correct the initial visual fine adjustment time according to the correction model to obtain the visual fine adjustment time.

[0099] Among them, in step S21, the performance parameters of the visual servo system can include parameters such as the response time and visual recognition accuracy of the visual servo system.

[0100] In step S22, the establishment of the mapping function can be obtained through the experimental calibration method. For example, a series of position error values are preset in advance, and the adjustment time of the visual servo system at each position error value is actually measured. By fitting these data points, the mapping function between the position error and the visual adjustment time is obtained. As a preferred implementation manner, the mapping function can adopt forms such as a linear function, a polynomial function, or a look-up table.

[0101] In step S23, the environmental illumination intensity can be collected in real time by an illumination sensor. The influence of the illumination intensity on the visual recognition accuracy is as follows: too low or too high illumination intensity will reduce the visual recognition accuracy, thereby affecting the visual adjustment time. The establishment of the correction model can be obtained through experimental analysis. For example, under different illumination intensities, the recognition accuracy and adjustment time of the visual system are tested, and a relationship model between the illumination intensity and the visual adjustment time is established. The correction model can adopt the form of a piecewise function. In the moderate illumination intensity range, the influence of the correction model on the initial visual fine adjustment time is small; in the range of too low or too high illumination intensity, the correction model increases the visual fine adjustment time to ensure the visual recognition accuracy.

[0102] Further, step S3 includes:

[0103] S31: Obtain the pre-positioning time;

[0104] S32: Query the pre-established correspondence between the pre-positioning time and the pre-positioning error attenuation amount according to the pre-positioning time, and obtain the pre-positioning error attenuation amount corresponding to the pre-positioning time;

[0105] S33: Modify the visual fine adjustment time according to the pre-positioning error attenuation amount to obtain the modified visual fine adjustment time;

[0106] S34: Construct a tact time model for a single socket assembly according to the pre-positioning time and the modified visual fine adjustment time.

[0107] Among them, in step S31, the pre-positioning time is the time consumed by the conical positioning mechanism for pre-positioning the socket to be assembled, and this time can be directly read and obtained by the control system.

[0108] In step S32, the correspondence between the pre-positioning time and the pre-positioning error attenuation amount can be pre-established through experiments or simulations. For example, by recording the attenuation of the socket position error under different pre-positioning times through multiple experiments, a correspondence table can be sorted out or a functional relationship can be fitted. In actual applications, by looking up the table or performing function calculations, the corresponding error attenuation amount can be quickly obtained according to the current pre-positioning time.

[0109] In step S33, modifying the visual fine adjustment time takes into account that the pre-positioning link has initially attenuated the position error of the socket, and the amount of error to be processed in the subsequent visual fine adjustment link is reduced. Therefore, the visual fine adjustment time can be correspondingly shortened. The modification method can be to subtract a value related to the pre-positioning error attenuation amount from the visual fine adjustment time. For example, the visual fine adjustment time can be directly multiplied by a correction coefficient related to the error attenuation amount. The larger the error attenuation amount, the smaller the correction coefficient, so as to shorten the visual fine adjustment time.

[0110] In step S34, the tact time model for a single socket assembly is to sum up the pre-positioning time and the modified visual fine adjustment time to obtain the total time consumed for a single assembly. This model provides a basis for subsequent tact time optimization and control.

[0111] Further, step S34 includes:

[0112] S341: According to the pre-positioning time and the modified visual fine adjustment time, use the time series analysis method to predict the tact time for the next N single socket assemblies, where N is an integer greater than 1;

[0113] S342: Calculate the variance of the tact time according to the predicted tact time for the next N single socket assemblies, and determine whether the variance exceeds a preset threshold;

[0114] S343: If the variance exceeds the preset threshold, adjust the pre-positioning time and re-execute steps S31 to S34 to reduce the volatility of the tact time; if the variance does not exceed the preset threshold, sum up the pre-positioning time and the corrected visual fine-tuning time to construct a tact time model for a single socket assembly.

[0115] Among them, in step S341, a time series analysis method is adopted to predict the tact time of multiple future single socket assemblies. As a preferred implementation, an autoregressive moving average model or an exponential smoothing model can be adopted for time series prediction.

[0116] In step S342, the variance of the tact time is calculated to evaluate the degree of fluctuation of the tact time. The preset threshold can be set according to the tact time stability requirements of the actual production line. For example, the preset threshold can be set as the maximum allowable value of the tact time variance.

[0117] In step S343, when the variance exceeds the preset threshold, the pre-positioning time will be adjusted. The adjustment of the pre-positioning time can be based on an optimization algorithm or manual experience. For example, the pre-positioning time can be shortened or extended to observe the change in the volatility of the tact time. After adjusting the pre-positioning time, steps S31 to S34 will be re-executed to form a closed-loop control until the variance of the tact time is reduced below the preset threshold. Thus, the volatility of the tact time is effectively reduced and the stability of production efficiency is improved.

[0118] Further, step S4 includes:

[0119] S41: Determine the pre-positioning time adjustment range according to the tact time model;

[0120] S42: Select multiple adjusted pre-positioning times within the pre-positioning time adjustment range and calculate the predicted tact time corresponding to each adjusted pre-positioning time according to the tact time model;

[0121] S43: Select the optimal pre-positioning time using an optimization algorithm according to the predicted tact time, and the optimization algorithm aims to minimize the tact time;

[0122] S44: Control the conical positioning mechanism to pre-position the socket to be assembled according to the optimal pre-positioning time to guide the socket assembly docking.

[0123] Among them, in step S41, the tact time model is used to determine the pre-positioning time adjustment range, which can be set to plus or minus 10% of the current pre-positioning time, or determined according to the tact time fluctuation range of the actual production line.

[0124] In step S42, within the determined pre-positioning time adjustment range, five pre-positioning time points can be selected at equal intervals. For example, if the adjustment range is from 100 milliseconds to 200 milliseconds, the selected pre-positioning time points can be 100 milliseconds, 125 milliseconds, 150 milliseconds, 175 milliseconds, and 200 milliseconds. For each selected pre-positioning time point, the cycle time model is used to predict the corresponding cycle time. The prediction of the cycle time model can be based on time series analysis methods, such as the ARIMA model, by analyzing historical cycle time data to predict future cycle time.

[0125] In step S43, with the goal of minimizing the cycle time, optimization algorithms can be used, such as the gradient descent method, genetic algorithm, or particle swarm algorithm. The gradient descent method iteratively optimizes the pre-positioning time to make the cycle time decrease along the gradient direction. The genetic algorithm simulates the biological evolution process and searches for the optimal pre-positioning time through operations such as selection, crossover, and mutation. The particle swarm algorithm simulates the foraging behavior of bird flocks and searches for the optimal pre-positioning time through the cooperation and information sharing among particles.

[0126] In step S44, the conical positioning mechanism operates according to the optimal pre-positioning time selected in step S43 to achieve the pre-positioning of the socket to be assembled. The conical positioning mechanism can be a pneumatic conical positioning mechanism or an electric conical positioning mechanism, which guides the socket into the predetermined position through the conical structure.

[0127] Please refer to Figure 2 , on the second aspect, a docking guiding device for a socket assembling machine, which is applied in the steps of a docking guiding method for a socket assembling machine according to any one of the above, the device includes:

[0128] Error attenuation model construction module 201: used to establish an error attenuation model and output the position error of the socket after pre-positioning according to the error attenuation model;

[0129] Visual adjustment time acquisition module 202: used to obtain the visual fine adjustment time according to the position error;

[0130] Cycle time model construction module 203: used to construct a cycle time model for a single socket assembly based on the position error and the visual fine adjustment time;

[0131] Guiding assembly docking module 204: used to adjust the pre-positioning time according to the cycle time model, control the conical positioning mechanism to pre-position the socket to be assembled according to the adjusted pre-positioning time, so as to guide the socket assembly docking.

[0132] Among them, the error attenuation model construction module 201 is configured to establish an error attenuation model by analyzing the geometric parameters of the conical positioning mechanism, the initial position error of the socket to be assembled, and the pre-positioning time range. The establishment of this error attenuation model can use methods such as historical data, experimental data, or simulation to predict the remaining position error of the socket after pre-positioning at different pre-positioning times.

[0133] The visual adjustment time acquisition module 202 is configured to determine the time required for the visual system to perform fine adjustment according to the position error output by the error attenuation model and in combination with the performance parameters of the visual servo system. The visual adjustment time acquisition module 202 can pre-establish a mapping relationship between the position error and the visual adjustment time, for example, through function fitting or look-up table to achieve fast query.

[0134] The tact time model construction module 203 is configured to integrate the pre-positioning time and the visual fine adjustment time to construct a tact time model for single socket assembly. This model can simply add the pre-positioning time and the visual fine adjustment time, or adopt a more complex model, such as considering time series analysis methods to predict the tact time fluctuation.

[0135] The guided assembly docking module 204 is configured to use the tact time model to analyze the influence of different pre-positioning times on the tact time, and adopt an optimization algorithm, such as the gradient descent method or the genetic algorithm, to select the optimal pre-positioning time. The selected optimal pre-positioning time is then used to control the conical positioning mechanism to achieve precise pre-positioning of the socket to be assembled, and finally guide the socket to complete the assembly docking.

[0136] In this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations.

[0137] The above are only embodiments of the present application and are not used to limit the protection scope of the present application. For those skilled in the art, the present application can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A docking guidance method for a socket assembly machine, used in the visually guided socket assembly process in a high-speed production workshop of electronic products, to improve the socket assembly docking speed, characterized in that: The method comprises the steps of: S1: establishing an error attenuation model, and outputting a position error of the socket after pre-positioning according to the error attenuation model; S2: Obtaining visual fine adjustment time according to the position error; S3: constructing a beat time model for a single socket assembly based on the position error and the visual fine adjustment time; S4: adjusting the pre-positioning time according to the beat time model, and controlling the conical positioning mechanism to pre-position the socket to be assembled according to the adjusted pre-positioning time to guide the socket assembly and docking.

2. A docking guide method for a socket assembly machine according to claim 1, characterized in that: Step S1 includes: S11: obtaining geometric parameters of the conical positioning mechanism, an initial position error of the socket to be assembled, and a pre-positioning time range; S12: establishing an error attenuation model according to the geometric parameters of the conical positioning mechanism; S13: selecting a plurality of pre-positioning time points at preset time intervals within the pre-positioning time range, and calculating the position error of the socket after the first pre-positioning corresponding to each of the pre-positioning time points according to the error attenuation model; S14: establishing a mapping relationship between the pre-positioning time and the position error attenuation effect according to the initial position error of the socket to be assembled and the position error of the socket after the first pre-positioning, and outputting the position error of the socket after the pre-positioning according to the mapping relationship.

3. A docking guide method for a socket assembly machine according to claim 2, characterized in that: Step S12 includes: S121: establishing an initial error attenuation model according to the geometric parameters of the conical positioning mechanism; S121: collecting historical wear data of the conical positioning mechanism, vibration data of the robot arm, workshop environment temperature data, and manufacturing error data of the socket to be assembled; S122: Correcting the initial error attenuation model according to the historical wear data of the conical positioning mechanism, the vibration data of the robot arm, the workshop environment temperature data, and the manufacturing error data of the socket to be assembled to obtain the error attenuation model.

4. A docking guide method for a socket assembly machine according to claim 3, characterized in that: Step S122 includes: S1221: Establishing a wear model of the conical positioning mechanism, a vibration model of the mechanical arm, a temperature compensation model, and a socket manufacturing error model according to the historical wear data of the conical positioning mechanism, the vibration data of the mechanical arm, the workshop environment temperature data, and the manufacturing error data of the socket to be assembled; S1222: Based on the wear model of the conical positioning mechanism, the robot arm vibration model, the temperature compensation model and the socket manufacturing error model, a weighted fusion algorithm is used to correct the initial error attenuation model to obtain the corrected error attenuation model.

5. A docking guide method for a socket assembly machine according to claim 4, characterized in that: Step S1222 includes: S12221: Determine the influence factor of each model on the pre-positioning accuracy; S12222: Based on the influencing factors, determine the weight of each model using the hierarchical analysis method; S12223: According to the weights of each model, the initial error attenuation model is corrected by using a weighted average method to obtain a corrected error attenuation model.

6. A docking guiding method for a socket assembly machine according to claim 1, characterized in that: Step S2 includes: S21: Acquire performance parameters of the visual servo system, and establish a mapping function between the position error and the visual adjustment time according to the position error of the socket after the pre-positioning and the performance parameters of the visual servo system; S22: calculating, according to the mapping function, an initial visual fine adjustment time corresponding to the position error; S23: Obtain the ambient light intensity of the workshop, and establish a correction model of light intensity to visual adjustment time according to the relationship between light intensity and visual recognition accuracy, and correct the initial visual fine adjustment time according to the correction model to obtain the visual fine adjustment time.

7. A docking guide method for a socket assembly machine according to claim 1, characterized in that: Step S3 includes: S31: Acquire the pre-positioning time; S32: According to the predetermined positioning time, querying a pre-established correspondence relationship between the predetermined positioning time and the predetermined positioning error attenuation amount, and obtaining the predetermined positioning error attenuation amount corresponding to the predetermined positioning time; S33: correcting the visual fine adjustment time according to the predetermined positioning error attenuation amount to obtain a corrected visual fine adjustment time; S34: Constructing a tact time model for a single socket assembly according to the pre-positioning time and the corrected visual fine adjustment time.

8. A docking guiding method for a socket assembly machine according to claim 7, characterized in that: Step S34 includes: S341: predicting the takt time of N single socket assemblies in the future according to the pre-positioning time and the corrected visual fine adjustment time by using a time series analysis method, where N is an integer greater than 1; S342: Calculate the variance of the takt time according to the predicted takt time of the future N single socket assemblies, and determine whether the variance exceeds a preset threshold; S343: If the variance exceeds the preset threshold, the pre-positioning time is adjusted and steps S31 to S34 are re-executed to reduce the volatility of the beat time; if the variance does not exceed the preset threshold, the pre-positioning time and the corrected visual fine-adjustment time are added together to construct a beat time model for a single socket assembly.

9. A docking guiding method for a socket assembly machine according to claim 1, characterized in that: Step S4 includes: S41: Determine a pre-positioning time adjustment range according to the beat time model; S42: selecting a plurality of adjusted pre-positioning times within the pre-positioning time adjustment range, and calculating a predicted tact time corresponding to each adjusted pre-positioning time according to the tact time model; S43: According to the predicted takt time, an optimization algorithm is used to select an optimal pre-positioning time, wherein the optimization algorithm aims to minimize the takt time; S44: Control the conical positioning mechanism to pre-position the socket to be assembled according to the optimal pre-positioning time to guide the socket assembly and docking.

10. A docking guide device for a socket assembly machine, characterized in that: In the steps of a docking guiding method for a socket assembly machine as applied to any one of claims 1 to 9, the device comprises: Error attenuation model building module: used to establish an error attenuation model and output the position error of the socket after pre-positioning according to the error attenuation model; A visual adjustment time acquisition module is used to acquire the visual fine adjustment time according to the position error; A beat time model building module: used to build a beat time model of a single socket assembly based on the position error and the visual fine adjustment time; Guiding assembly docking module: used to adjust the pre-positioning time according to the beat time model, and control the conical positioning mechanism to pre-position the socket to be assembled according to the adjusted pre-positioning time to guide the socket assembly docking.

Images