Assembly robot applied to industrial production line

Abstract

According to the assembly robot applied to the industrial production line, workpiece position, vibration frequency and contact force change data are synchronously obtained through an object detection unit, and a high-precision displacement vector is generated in combination with a parallel processing technology; the parameter adjusting unit dynamically calculates a gain correction coefficient based on a sliding mode controller, so that nonlinear fluctuation of workpiece displacement is effectively dealt with, and the robustness of the system is remarkably improved; the path prediction unit analyzes and predicts a future displacement track of a workpiece through historical data, provides a control instruction for the robot, and reduces operation errors caused by lag; the abnormity processing unit monitors a displacement threshold value in real time, triggers a production line to pause and a standby robot to reset, forms a closed-loop emergency mechanism, guarantees equipment safety and production line continuity, and further solves the problems that when an existing industrial robot displaces due to vibration or conveyor belt errors, the robot cannot adjust grabbing or assembling actions in real time, and the working efficiency is high. And the task failure or fault is caused.

Description

Technical Field

[0001] The present invention relates to the technical field of assembly robots, and particularly to an assembly robot applied to an industrial production line. Background Art

[0002] The assembly robot on an industrial production line is a high-precision automation device, composed of a robotic arm, sensors, and a control system, and has the ability to move flexibly with multiple degrees of freedom. It realizes complex assembly tasks through programming, is widely used in industries such as automobiles and electronics, and can complete processes such as welding, handling, and sorting. Its advantages include continuous operation for 24 hours, significantly improving production efficiency and product consistency, while reducing labor costs and safety risks. It is the core equipment of intelligent manufacturing.

[0003] However, current industrial assembly robots rely on preset programs and fixed trajectories and are difficult to cope with the dynamically changing environment on the production line. For example, when the workpiece is displaced due to vibration or conveyor belt error, the robot cannot adjust the grasping or assembly actions in real time, resulting in task failure or malfunction. Therefore, there is an urgent need to design an assembly robot for an industrial production line with reinforcement learning ability. Summary of the Invention

[0004] The purpose of the present invention is to provide an assembly robot applied to an industrial production line to solve the technical problem that when the existing industrial robot faces the displacement of the workpiece due to vibration or conveyor belt error, the robot cannot adjust the grasping or assembly actions in real time, resulting in task failure or malfunction.

[0005] The technical solution of the present invention is realized as follows:

[0006] An assembly robot applied to an industrial production line includes an object detection unit, a parameter adjustment unit, a path prediction unit, and an exception handling unit;

[0007] The object detection unit is used to obtain the original data of the position, vibration frequency, and contact force change of the target workpiece, and process the original data in parallel to generate the displacement vector of the target workpiece;

[0008] The parameter adjustment unit is used to calculate the displacement vector gain correction coefficient by using a sliding mode controller according to the displacement vector of the target workpiece;

[0009] The path prediction unit is used to predict the displacement position of the target workpiece within a preset time according to the historical data of the displacement vector;

[0010] The exception handling unit is used to detect whether the target workpiece exceeds the displacement threshold within a preset number of times. If it exceeds, the task is paused, the production line is notified to pause the conveyor belt, and a spare robot end effector is started to assist in resetting the target workpiece.

[0011] A further technical solution is that the object detection unit includes:

[0012] A multi-angle scanning module, which uses a camera and a vibration sensor to capture the original data of the position, vibration frequency, and contact force of the target workpiece;

[0013] A data alignment module, which is used to unify the original data onto the same time axis for parallel processing to generate unified data;

[0014] A dynamic analysis module, which is used to identify the abnormal vibration mode of the target workpiece and the sudden change force of the contact force according to the unified data;

[0015] A displacement calculation module, which is used to calculate the displacement vector of the target workpiece according to the multi-angle scanning module, the data alignment module, and the dynamic analysis module.

[0016] A further technical solution is that the specific implementation steps of the parameter adjustment unit include:

[0017] Step S11: Compare the actual displacement of the target workpiece with the preset standard displacement to generate an error signal;

[0018] Step S12: Define a virtual sliding surface, where the virtual sliding surface serves as the displacement reference trajectory of the target workpiece;

[0019] Step S13: Determine the displacement speed of the target workpiece according to the deviation between the error signal and the virtual sliding surface;

[0020] Step S14: Integrate the error signal, the preset virtual sliding surface, and the displacement speed data to calculate a displacement vector gain correction coefficient for adjusting the target workpiece.

[0021] A further technical solution is that the specific implementation steps of Step S13 include:

[0022] Step S131: Correlate the deviation between the error signal and the virtual sliding surface to generate a deviation ratio factor;

[0023] Step S132: Set the upper and lower limits of the displacement speed according to the deviation ratio factor;

[0024] Step S133: Obtain the change trend of the error signal and adjust the change range of the displacement speed. Specifically, if the error signal shows a rapid decay trend, lower the upper limit of the displacement speed; if the error signal fluctuates repeatedly, raise the upper limit of the displacement speed;

[0025] Step S134: Calculate the real-time displacement speed of the target workpiece based on the combined scale factor, the change range of the displacement speed, and the dynamic calibration result.

[0026] A further technical solution is that the specific implementation steps of step S14 include:

[0027] Step S141: Dynamically weight the deviation between the error signal and the virtual sliding surface to generate the correlation degree between the error signal and the virtual sliding surface.

[0028] Step S142: Combine the displacement speed data and the correlation degree to calculate the coupling coefficient between the displacement speed and the correlation degree.

[0029] Step S143: According to the coupling coefficient, proportionally allocate the gain correction coefficient to different control dimensions.

[0030] A further technical solution is that the path prediction unit includes:

[0031] A feature extraction module, which is used to arrange the displacement vector historical data of the target workpiece in a time series and extract features such as displacement trend, periodic fluctuation, and mutation points.

[0032] A path simulation module, based on the displacement trend, periodic fluctuation, and mutation point feature extraction results, constructs a baseline to simulate the expected displacement path of the target workpiece within a preset time.

[0033] A data compensation module, which is used to combine the real-time monitoring data of the production line and perform disturbance correction on the baseline.

[0034] An interval prediction module, which is used to expand the corrected dynamic baseline into a probability interval to represent the possible displacement range of the target workpiece within a preset time.

[0035] A model parameter adjustment module, which is used to continuously collect the actual displacement data of the target workpiece during the prediction execution process and compare it with the predicted value, and adjust the model parameters through the comparison error feedback.

[0036] A further technical solution is that the specific implementation steps of the path simulation module include:

[0037] Step S21: Separate the displacement trend from the periodic fluctuation.

[0038] Step S22: Based on the mutation point detection result, insert correction points into the baseline.

[0039] Step S23: Fuse the displacement trend component, periodic component, and the correction points according to weights to generate the baseline.

[0040] Step S24: Based on the reference line, simulate the displacement path of the target workpiece in the next N seconds, and compare the actual displacement data in real time. Among them, if the deviation between the displacement path and the actual value exceeds the threshold, trigger the reference line parameter to reduce the weight of the periodic component.

[0041] A further technical solution is that the specific steps executed by the model parameter adjustment module include:

[0042] Step S31: During the prediction execution, synchronously collect the actual displacement data and the predicted value of the target workpiece, and align them according to the time stamp;

[0043] Step S32: Divide the deviation between the predicted value and the actual displacement data into a trend error and a sudden error;

[0044] Step S33: Adjust the sensitivity of the model parameters according to the hierarchical results of the trend error and the sudden error;

[0045] Step S34: Apply the adjusted model parameters to the next round of prediction, and monitor the matching degree between the actual displacement of the target workpiece and the predicted value.

[0046] A further technical solution is that the specific steps executed by Step S33 include:

[0047] Step S331: Set the priority of the model parameter adjustment according to the proportion of the trend error and the sudden error;

[0048] Step S332: Set dynamic response thresholds for the trend error and the sudden error respectively;

[0049] Step S333: Based on the priority and the dynamic response threshold, allocate the adjustment amplitude of the model parameters proportionally.

[0050] A further technical solution is that the exception handling unit includes:

[0051] A threshold trigger detection module, which is used to monitor in real time whether the displacement of the target workpiece exceeds the set threshold within the preset number of times;

[0052] A production line linkage pause module, which is used to send a pause instruction to the production line control system according to the exception signal triggered by the threshold trigger detection module, and synchronously stop the operation of the conveyor belt;

[0053] A standby actuator intervention module, which is used to activate the standby robot end effector and plan a reset path according to the current displacement state of the target workpiece.

[0054] The beneficial effects of the present invention are as follows:

[0055] The workpiece position, vibration frequency, and contact force change data are synchronously acquired by the object detection unit, and high-precision displacement vectors are generated by combining parallel processing technology; the parameter adjustment unit dynamically calculates the gain correction coefficient based on the sliding mode controller to effectively cope with the non-linear fluctuations of the workpiece displacement and significantly improve the system robustness; the path prediction unit predicts the future displacement trajectory of the workpiece through historical data analysis and provides control instructions for the robot to reduce operation errors caused by lag; the anomaly handling unit monitors the displacement threshold in real time, triggers the production line pause and standby robot reset operations, forms a closed-loop emergency mechanism, and ensures the equipment safety and production line continuity, thereby solving the technical problem that when the existing industrial robot faces the displacement of the workpiece due to vibration or conveyor belt error, the robot cannot adjust the grasping or assembly actions in real time, resulting in task failure or malfunction. Brief Description of the Drawings

[0056] Figure 1 It is a block diagram of an assembly robot system applied to an industrial production line according to the present invention;

[0057] Figure 2 It is a block diagram of the object detection unit system in the assembly robot applied to an industrial production line according to the present invention;

[0058] Figure 3 It is a flowchart of the specific implementation steps of the parameter adjustment unit in the assembly robot applied to an industrial production line according to the present invention;

[0059] Figure 4 It is a block diagram of the path prediction unit system in the assembly robot applied to an industrial production line according to the present invention;

[0060] Figure 5 It is a flowchart of the specific implementation steps of the path simulation module in the assembly robot applied to an industrial production line according to the present invention;

[0061] Figure 6 It is a flowchart of the specific implementation steps of the model parameter adjustment module in the assembly robot applied to an industrial production line according to the present invention;

[0062] Figure 7 It is a block diagram of the anomaly handling unit system in the assembly robot applied to an industrial production line according to the present invention. Detailed Embodiments

[0063] To better understand the technical content of the present invention, specific embodiments are provided below, and the present invention will be further described in conjunction with the accompanying drawings.

[0064] See Figures 1 to 7The present invention provides an assembly robot applied to an industrial production line, comprising an object detection unit, a parameter adjustment unit, a path prediction unit and an exception handling unit; the object detection unit is used to obtain raw data of a target workpiece position, a vibration frequency and a contact force change, and to process the raw data in parallel to generate a displacement vector of the target workpiece; the parameter adjustment unit is used to calculate a displacement vector gain correction coefficient using a sliding mode controller according to the displacement vector of the target workpiece; the path prediction unit is used to predict the displacement position of the target workpiece within a preset time according to historical data of the displacement vector; the exception handling unit is used to detect whether the target workpiece exceeds a displacement threshold within a preset number of times, and if so, suspend the task, notify the production line to suspend the conveyor belt, and start the standby robot end effector to assist in resetting the target workpiece.

[0065] It should be noted that the displacement vector includes the vector of the target workpiece's position change in three-dimensional space, which is composed of the displacement components in the X / Y / Z axis directions and the rotation angle, and is used to quantify the dynamic motion state of the target workpiece. The displacement position within the preset time includes the future position of the target workpiece predicted by modeling historical displacement data, which serves as the reference trajectory for the robot to perform operations such as grasping and assembly.

[0066] In the embodiment of the present invention, by integrating four functional modules, namely, an object detection unit, a parameter adjustment unit, a path prediction unit and an exception handling unit, real-time perception, dynamic regulation, prediction and active exception response of the displacement state of the target workpiece are achieved.

[0067] Specifically, the object detection unit synchronously collects the original data such as the position coordinates, vibration frequency and contact force change of the target workpiece through laser ranging, accelerometer and force sensor, and generates high-precision displacement vectors in parallel, so as to realize multi-dimensional modeling of the dynamic state of the workpiece. The parameter adjustment unit calculates the displacement vector gain correction coefficient based on the sliding mode controller, and compensates for the nonlinear fluctuation of the workpiece displacement by dynamically adjusting the control gain, so as to ensure the robustness and real-time performance of the control system. The path prediction unit constructs a time series model using the historical data of the displacement vector, combines trend analysis with the error stratification algorithm, predicts the displacement position of the target workpiece within the preset time in the future, and provides control instructions for the robot end effector; the abnormal handling unit determines whether the workpiece has abnormal displacement through displacement threshold detection within the preset number of times. If the threshold is triggered, the production line control system is linked to suspend the conveyor belt operation, and the standby robot end effector (such as a robotic arm) is started to perform a reset operation, forming a closed-loop emergency mechanism, thereby solving the technical problem that when the existing industrial robot faces the displacement of the workpiece due to vibration or conveyor belt error, the robot cannot adjust the grasping or assembly action in real time, resulting in task failure or malfunction.

[0068] Further, the object detection unit includes: a multi-angle scanning module that uses a camera and a vibration sensor to capture the original data of the position, vibration frequency, and contact force of the target workpiece; a data alignment module that is used to unify the original data onto the same time axis for parallel processing to generate unified data; a dynamic analysis module that is used to identify the abnormal vibration mode of the target workpiece and the sudden change in contact force according to the unified data; and a displacement calculation module that is used to calculate the displacement vector of the target workpiece based on the multi-angle scanning module, the data alignment module, and the dynamic analysis module.

[0069] In the embodiment of the present invention, the multi-angle scanning module cooperates with an industrial camera, a triaxial accelerometer, and a force sensor to respectively collect the position coordinates of the target workpiece, the harmonic component vibration frequency in the range of 0 - 2000 Hz, and the instantaneous fluctuation contact force change of the clamping force or friction force, generating three types of original data; among them, the camera extracts the contour and center coordinates of the target workpiece through image recognition technology, and the vibration sensor quantifies the vibration signal and contact force data through the piezoelectric effect or strain gauge principle. The camera provides high-resolution visual positioning, and the vibration sensor captures microscopic mechanical characteristics, and the two complement each other to comprehensively perceive the state of the target workpiece. The data alignment module uses a timestamp hardware trigger synchronization algorithm to eliminate the time sequence offset of multiple sensors through hardware triggering or software interpolation, ensuring that subsequent analysis is based on synchronized data and avoiding misjudgment caused by timestamp misalignment. Then, the image frame, vibration signal, and force data are mapped to the unified time axis to eliminate the time sequence offset caused by the difference in sensor sampling rates, forming a unified data set, providing a benchmark for subsequent parallel processing. The dynamic analysis module extracts the periodic characteristics of the vibration frequency based on the unified data set through the sliding window Fourier transform, and combines the sliding mode control theory to identify abnormal vibration modes such as high-frequency chatter and non-linear harmonics; at the same time, the differential threshold algorithm is used to detect the sudden change in contact force to determine whether the workpiece has abnormal contacts such as jamming and slipping. The displacement calculation module comprehensively considers the workpiece position data scanned at multiple angles, the aligned vibration and force data, and uses the three-dimensional coordinate difference formula: ΔX = X2 - X1, ΔY = Y1 - Y2, ΔZ = Z1 - Z2 to calculate the displacement vector, and corrects the displacement error in combination with the dynamic analysis result, and finally outputs a high-precision target workpiece displacement vector including direction and amplitude.

[0070] Preferably, the specific steps executed by the parameter adjustment unit include:

[0071] Step S11: Compare the actual displacement of the target workpiece with the preset standard displacement to generate an error signal;

[0072] Step S12: Define a virtual sliding surface, where the virtual sliding surface serves as the displacement reference trajectory of the target workpiece;

[0073] Step S13: Determine the displacement speed of the target workpiece according to the deviation between the error signal and the virtual sliding surface;

[0074] Step S14: Integrate the error signal, the preset virtual sliding surface, and the displacement speed data to calculate the displacement vector gain correction coefficient for adjusting the target workpiece.

[0075] It should be noted that the virtual sliding surface includes the mathematical reference surface defined in sliding mode control, which characterizes the dynamic characteristics of the ideal displacement trajectory. Through the reaching law design, it ensures that the actual displacement quickly converges to this surface and maintains stability. The preset standard displacement includes the theoretical trajectory coordinates.

[0076] In the embodiment of the present invention, the actual displacement of the target workpiece is compared with the preset standard displacement in real time, and the deviation value in the spatial dimension between the two is calculated to generate an error signal, which is used to quantify the deviation degree of the current displacement of the target workpiece from the target trajectory. For example, if the actual displacement is 5.2mm, -0.3mm, 1.8mm, and the preset standard displacement is 5.0mm, 0.0mm, 2.0mm, then the error signal is 0.2mm, -0.3mm, -0.2mm. Then, a virtual sliding surface is defined. The virtual sliding surface is represented by the mathematical function S(t) = e(t) + λ∫e(t)dt, where e(t) is the error signal and λ is a positive definite constant, which characterizes the ideal displacement reference trajectory of the target workpiece. The design of the sliding surface follows the sliding mode control theory, and through the reaching law, it ensures that the actual displacement quickly converges to the sliding surface and maintains stable motion. According to the deviation between the error signal and the virtual sliding surface (such as the sliding surface function value S(t)), combined with the time difference calculation Δt, the displacement speed of the target workpiece is deduced, that is, the rate of displacement change per unit time (dx / dt, dy / dt, dz / dt). For example, if the error signal increases linearly with time, the displacement speed v(t) can be estimated and quantified by the slope Δe / Δt, which reflects the intensity of the trend of the target workpiece deviating from the trajectory. Finally, integrate the error signal, the virtual sliding surface function value, and the displacement speed data, and use the sliding mode reaching law to design the gain correction coefficient K. The specific formula is as follows: Among them, η is the switching gain, which is used to suppress chattering, and κ is the reaching speed coefficient, which determines the system convergence speed. Substitute the above formula into the sliding surface dynamic equation to deduce the gain correction coefficient: Where ε is a small positive number to prevent the denominator from being zero. This coefficient is used to dynamically adjust the output gain of the controller to compensate for the displacement deviation of the target workpiece caused by disturbances such as vibration and sudden change of contact force. For example, when the error signal increases and the sliding surface deviation exceeds the threshold, the gain correction coefficient is adjusted to K = 1.2 to enhance the response intensity of the controller.

[0077] Further, the specific implementation steps of step S13 include:

[0078] Step S131: Correlate the error signal with the deviation of the virtual sliding surface to generate a deviation proportionality factor;

[0079] Step S132: Set the upper and lower limits of the displacement speed according to the deviation proportionality factor;

[0080] Step S133: Obtain the change trend of the error signal and adjust the change range of the displacement speed. Specifically, if the error signal shows a rapid decay trend, lower the upper limit of the displacement speed; if the error signal fluctuates repeatedly, raise the upper limit of the displacement speed;

[0081] Step S134: Calculate the real-time displacement speed of the target workpiece by synthesizing the proportionality factor, the change range of the displacement speed, and the dynamic calibration result.

[0082] In the invention embodiment, the error signal of the target workpiece is correlated with the deviation of the virtual sliding surface, and the deviation proportionality factor K is generated through the PI algorithm in the sliding mode control P . For example, if the error signal is e(t) and the deviation of the virtual sliding surface is then the proportionality factor can be expressed as: Set the upper limit V max and the lower limit V min of the displacement speed according to the deviation proportionality factor. The specific formula is:

[0083] V max = V0 + α·K P , V min = V0 - β·K p

[0084] where V0 is the reference speed, and α and β are adjustment coefficients. When K P increases, the system expands the speed regulation range by increasing the upper limit V max and decreasing the lower limit V min to cope with larger deviations. Analyze the change trend of the error signal e(t) through the sliding window Fourier transform or the moving average algorithm: A rapid decay trend indicates that the actual displacement quickly converges to the virtual sliding surface. At this time, lower the upper limit V max to V' max = V max ·(1 - γ) to avoid overshoot. A repeated fluctuation trend and e(t) frequently changes sign, indicating that there are disturbances or chattering in the actual displacement. At this time, raise the upper limit V max to V' max = V max ·(1 + δ) to enhance the response ability. Here, θ, φ, γ, and δ are preset threshold coefficients. Synthesize the deviation proportionality factor K P , the adjusted speed range [V' min, V' max and the dynamic calibration results (such as the sensor noise compensation value η), calculate the real-time displacement velocity v(t) through the sliding mode reaching law formula:

[0085]

[0086] Among them, sat(·) is the boundary layer saturation function used to suppress high-frequency chattering. This formula ensures that the speed is dynamically adjusted within the set range and combines the calibration data to eliminate measurement errors.

[0087] Furthermore, the specific implementation steps of step S14 include:

[0088] Step S141: Dynamically weight the deviation between the error signal and the virtual sliding surface to generate the correlation degree between the error signal and the virtual sliding surface;

[0089] Step S142: Combine the displacement velocity data with the correlation degree to calculate the coupling coefficient between the displacement velocity and the correlation degree;

[0090] Step S143: According to the coupling coefficient, proportionally allocate the gain correction coefficient to different control dimensions.

[0091] In the embodiment of the present invention, the error signal of the target workpiece and the deviation of the virtual sliding surface are dynamically weighted in real time to generate the correlation degree between the two. The dynamic weighting adopts the sliding window exponential decay algorithm, and the weight w(t) changes with time:

[0092]

[0093] Among them, α > 0, and the calculation formula of the correlation degree C(t) is:

[0094] C(t) = w(t)·e(t) + (1 - w(t))·s(t)

[0095] Among them, e(t) and s(t) are respectively the above-mentioned error signal and the deviation of the virtual sliding surface. The weight w(t) is dynamically adjusted according to the amplitude of the sliding surface deviation s(t): when s(t) is larger, the weight w(t) increases, strengthening the influence of the error signal on the correlation degree; when s(t) is smaller, the weight w(t) decreases, weakening the role of the error signal. The dynamic weighting ensures that the correlation degree C(t) can reflect the coupling strength between the error signal and the sliding surface deviation in real time, providing a quantitative basis for subsequent gain allocation. Combining the displacement velocity data v(t) with the correlation degree C(t), calculate the coupling coefficient between the two through covariance analysis, and the formula is as follows:

[0096]

[0097] Among them, Cov(·,·) represents covariance, and Var(·) represents variance. If KC Close to 1, indicating a strong positive correlation between the displacement speed and the correlation degree, and it is necessary to increase the gain correction coefficient to suppress error diffusion; K C Close to -1, indicating a negative correlation between the two, and it is necessary to reduce the gain correction coefficient to avoid overshoot. According to the coupling coefficient K C , the gain correction coefficient K is proportionally allocated to different control dimensions (such as X / Y / Z axes).

[0098] Furthermore, the path prediction unit includes: a feature extraction module for arranging the historical displacement vector data of the target workpiece in a time series and extracting features such as displacement trend, periodic fluctuation, and mutation point; a path simulation module for constructing a baseline based on the displacement trend, periodic fluctuation, and mutation point feature extraction results to simulate the expected displacement path of the target workpiece within a preset time; a data compensation module for combining the real-time monitoring data of the production line to correct the perturbation of the baseline; an interval prediction module for expanding the corrected dynamic baseline into a probability interval to represent the possible displacement range of the target workpiece within a preset time; a model parameter adjustment module for continuously collecting the actual displacement data of the target workpiece during the prediction execution and comparing it with the predicted value, and adjusting the model parameters through the comparison error feedback.

[0099] It should be noted that the baseline includes the expected displacement path constructed by combining trends, periodicity, and mutation characteristics, reflecting the law of the target workpiece's movement. The probability interval includes the prediction boundary expanded based on statistical methods, representing the possible range of the target workpiece's displacement. The model can be a neural network model or an AI model, and the adjustment of the model parameters includes the process of optimizing the model parameters through error feedback to ensure that the prediction model adapts to the actual working conditions.

[0100] In the embodiment of the present invention, the feature extraction module performs time series analysis on the historical displacement vector data of the target workpiece, and extracts features such as uniform, accelerating, or decelerating displacement trends, mechanical resonance or interference periodic fluctuations, and sudden pauses or sudden acceleration mutation points; subsequently, the path simulation module constructs a baseline based on the above features and simulates the expected displacement path of the workpiece through trend fitting, periodic compensation, and mutation correction; then, the data compensation module combines the real-time monitoring data of the production line (such as vibration noise, external disturbance) and dynamically corrects the baseline through a weighted algorithm to adapt to environmental changes; on this basis, the interval prediction module uses statistical methods to expand the corrected baseline into a probability interval, quantifying the distribution of prediction errors and forming a confidence range for the workpiece displacement; finally, the model parameter adjustment module continuously collects the actual displacement data, and dynamically optimizes the model parameters through the error feedback-driven gradient descent method, enabling the prediction model to adapt to the actual working condition changes. This technology realizes high-precision prediction and robust control of the target workpiece displacement path through feature-driven modeling, dynamic compensation, probability boundary expansion, and online parameter optimization.

[0101] Furthermore, the specific implementation steps of the path simulation module include:

[0102] Step S21: Separate the displacement trend from the periodic fluctuations.

[0103] Step S22: Insert correction points into the baseline based on the mutation point detection results.

[0104] Step S23: Fuse the displacement trend component, the periodic component, and the correction points according to weights to generate a baseline.

[0105] Step S24: Based on the baseline, simulate the displacement path of the target workpiece in the next 3 - 5 seconds and compare it with the actual displacement data in real time. If the deviation between the displacement path and the actual value exceeds the threshold, trigger a reduction in the weight of the periodic component in the baseline parameters.

[0106] In the embodiments of the present invention, the displacement trend and periodic fluctuations of the target workpiece are decoupled through time - frequency analysis (such as separating low - frequency and high - frequency signals), and their characteristics are extracted respectively. Subsequently, based on the mutation point detection results (such as sudden pauses, rapid accelerations, or rapid decelerations), the abnormal positions are identified through a threshold detection algorithm, and correction points are inserted into the baseline to compensate for the mutation disturbances. Then, the separated displacement trend component, periodic component, and correction points are fused according to dynamic weights to generate a baseline, where the weights are dynamically adjusted according to trend stability, periodic intensity, and mutation frequency to balance the contributions of each component. Finally, based on the baseline, the displacement path of the workpiece in the next N seconds is simulated, and the actual displacement data is compared in real time. If the deviation between the predicted path and the actual value exceeds the preset threshold, the weight of the periodic component is automatically reduced, and the influence of the trend component is preferentially strengthened, thereby enhancing the robustness of the model to environmental disturbances. This method realizes high - precision prediction of the displacement path of the target workpiece and real - time optimization of model parameters through multi - component decoupling, dynamic correction, and adaptive adjustment.

[0107] Furthermore, the specific implementation steps of the model parameter adjustment module include:

[0108] Step S31: During the prediction execution, synchronously collect the actual displacement data and predicted values of the target workpiece and align them according to timestamps.

[0109] Step S32: Divide the deviation between the predicted value and the actual displacement data into trend errors and sudden errors.

[0110] Step S33: Adjust the sensitivity of the model parameters according to the hierarchical results of trend errors and sudden errors.

[0111] Step S34: Apply the adjusted model parameters to the next round of prediction and monitor the matching degree between the actual displacement and the predicted value of the target workpiece.

[0112] In an embodiment of the present invention, during predictive execution, actual displacement data of a target workpiece and predicted values (output by the model) are synchronously collected, and the time difference in data collection is eliminated through timestamp alignment to ensure that the two are compared under the same time reference; the deviation between the predicted value and the actual displacement is divided into two categories: trend errors (such as the prediction deviation of the model for the long-term acceleration / deceleration trend of the workpiece) and sudden errors (such as short-term severe deviations caused by external shocks or sensor noise). The former extracts long-term deviation features through low-pass filtering, and the latter identifies mutation events through threshold detection; the sensitivity of the model parameters is adjusted according to the hierarchical results of the error types: if the trend error is significant, the response weight of the model to the displacement trend (such as the linear regression slope) is enhanced, and the sensitivity to high-frequency disturbances is reduced; if sudden errors are frequent, the sensitivity of the mutation point detection is increased and the threshold for inserting correction points is dynamically adjusted to compensate for short-term disturbances; the adjusted model parameters are applied to the next round of prediction, and the matching degree between the actual displacement and the predicted value is continuously monitored (such as quantifying the accuracy through the root mean square error). If the matching degree is lower than the preset threshold, a new round of error analysis and parameter adjustment is triggered to form a closed-loop feedback mechanism. This method enables the model to maintain high-precision prediction ability under complex working conditions through synchronous data alignment, error hierarchical analysis, and dynamic parameter optimization.

[0113] Preferably, the specific implementation steps of step S33 include:

[0114] Step S331: Set the priority of model parameter adjustment according to the proportion of trend errors and sudden errors.

[0115] Step S332: Set dynamic response thresholds for trend errors and sudden errors respectively.

[0116] Step S333: Based on the priority and dynamic response thresholds, allocate the adjustment amplitude of the model parameters proportionally.

[0117] In an embodiment of the present invention, according to the proportion of the potential error generated by the prediction deviation of the model for the long-term acceleration / deceleration trend of the target workpiece and the sudden error generated by the short-term severe deviation caused by external shock or sensor noise (such as the proportion P of the trend error t and the proportion P of the sudden error s ), the priority of model parameter adjustment is set through a weighting rule: if P t > P S , then the sensitivity of the model to trend features is adjusted first (such as enhancing the weight of the linear regression slope); if p S > p t , then the correction ability of the model to mutation points is optimized first (such as increasing the sensitivity of the mutation point detection threshold). Dynamic response thresholds are set for trend errors and sudden errors respectively: the trend error threshold θ tCalculate the long-term fluctuation range of historical deviation through moving window averaging or low-pass filtering, i.e.:

[0118] θ e = μ e + k·σ e

[0119] where μ is the mean value, σ is the standard deviation, and k is the amplification factor. And the sudden error threshold θ s is dynamically adjusted by real-time monitoring of the frequency and amplitude of mutation events, i.e.:

[0120] θ s = θ b ·(1 + α·f j )

[0121] where f j is the mutation frequency and α is the adjustment coefficient. Based on the priority and dynamic response threshold, allocate the adjustment amplitude of model parameters proportionally: if the priority of the trend error is high, the parameter adjustment amplitude Δw is allocated according to the proportion P t of the trend error, Δw tr = Δw to ·P t , and limit the adjustment amplitude Δw s of the sudden error not to exceed the remaining proportion; conversely, if the priority of the sudden error is high, allocate the adjustment amplitude in the reverse direction and dynamically compress the adjustment weight of the trend error. This method enables the model to achieve precise and adaptive optimization of parameter adjustment under complex working conditions through the priority setting driven by error proportion, dynamic threshold response, and proportional allocation strategy.

[0122] Furthermore, the exception handling unit includes: a threshold trigger detection module for real-time monitoring whether the displacement of the target workpiece exceeds the set threshold within a preset number of times; a production line linkage pause module for sending a pause instruction to the production line control system according to the exception signal triggered by the threshold trigger detection module to synchronously stop the conveyor belt operation; a standby actuator intervention module for activating the standby robot end effector and planning a reset path according to the current displacement state of the target workpiece.

[0123] In an embodiment of the present invention, the threshold trigger detection module continuously obtains the displacement of the target workpiece through a sensor or a vision system within a preset number of times (such as multiple consecutive sampling periods), and compares it with the threshold set by the process; when the displacement exceeds the threshold range, this module triggers an abnormal signal and starts the subsequent process. After receiving the abnormal signal, the production line linkage pause module immediately sends a pause instruction to the production line control system through an industrial communication protocol, synchronously stops the operation of the conveyor belt and other associated devices, ensures that the position of the target workpiece does not change when an abnormality occurs, and avoids secondary failures. The standby actuator intervention module activates the standby robot end effector (such as a robotic arm or a fixture) after the production line pauses, and dynamically plans a reset path according to the current displacement state of the target workpiece (such as the deviation direction and amplitude): if the deviation is small, the grasping path is calculated through a kinematic algorithm; if the deviation is serious, a preset safe path (such as retreating to the initial position) is called. The whole process realizes the closed-loop control from abnormal detection to reset through the coordinated cooperation of sensor data acquisition, industrial control signal transmission and intelligent path planning, ensuring the stability and safety of the production line operation.

[0124] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An assembly robot applied to an industrial production line, characterized in that, It includes an object detection unit, a parameter adjustment unit, a path prediction unit, and an exception handling unit; The object detection unit is used to obtain the original data of the target workpiece position, vibration frequency, and contact force change, and parallel process the original data to generate the displacement vector of the target workpiece; The parameter adjustment unit is used to calculate the displacement vector gain correction coefficient by using a sliding mode controller according to the displacement vector of the target workpiece; The path prediction unit is used to predict the displacement position of the target workpiece within a preset time according to the historical data of the displacement vector; The exception handling unit is used to detect whether the target workpiece exceeds the displacement threshold within a preset number of times. If it exceeds, the task is paused, the production line is notified to pause the conveyor belt, and a spare robot end effector is started to assist in resetting the target workpiece.

2. The assembly robot applied to an industrial production line according to claim 1, characterized in that, The object detection unit includes: The multi-angle scanning module, which uses a camera and a vibration sensor to capture the original data of the position, vibration frequency, and contact force of the target workpiece; The data alignment module is used to unify the original data onto the same time axis, perform parallel processing, and generate unified data; The dynamic analysis module is used to identify the abnormal vibration mode of the target workpiece and the mutation intensity of the contact force according to the unified data; The displacement calculation module is used to calculate the displacement vector of the target workpiece according to the multi-angle scanning module, the data alignment module, and the dynamic analysis module.

3. The assembly robot applied to an industrial production line according to claim 1, wherein The specific execution steps of the parameter adjustment unit include: Step S11: Compare the actual displacement of the target workpiece with the preset standard displacement to generate an error signal; Step S12: Define a virtual sliding surface, where the virtual sliding surface serves as the displacement reference trajectory of the target workpiece; Step S13: Determine the displacement speed of the target workpiece according to the deviation between the error signal and the virtual sliding surface; Step S14: Integrate the error signal, the preset virtual sliding surface, and the displacement speed data to calculate the displacement vector gain correction coefficient for adjusting the target workpiece.

4. The assembly robot applied to an industrial production line according to claim 3, wherein, The specific execution steps of Step S13 include: Step S131: Correlate the deviation between the error signal and the virtual sliding surface to generate a deviation ratio factor; Step S132: Set the upper and lower limits of the displacement speed according to the deviation ratio factor; Step S133: Obtain the change trend of the error signal and adjust the change range of the displacement speed. Among them, if the error signal shows a rapid decay trend, the upper limit of the displacement speed is reduced; if the error signal fluctuates repeatedly, the upper limit of the displacement speed is increased; Step S134: Calculate the real-time displacement speed of the target workpiece by synthesizing the ratio factor, the change range of the displacement speed, and the dynamic calibration result.

5. The assembly robot applied to an industrial production line according to claim 4, wherein, The specific execution steps of Step S14 include: Step S141: Dynamically weight the deviation between the error signal and the virtual sliding surface to generate the correlation degree between the error signal and the virtual sliding surface; Step S142: Combine the displacement speed data with the correlation degree to calculate the coupling coefficient between the displacement speed and the correlation degree; Step S143: Allocate the gain correction coefficient to different control dimensions proportionally according to the coupling coefficient.

6. The assembly robot applied to an industrial production line according to claim 1, wherein The path prediction unit includes: A feature extraction module, configured to arrange the displacement vector historical data of the target workpiece in a time series, and extract displacement trends, periodic fluctuations, and feature of mutation points; A path simulation module, which constructs a baseline based on the extraction results of the displacement trend, periodic fluctuations, and feature of mutation points to simulate the expected displacement path of the target workpiece within a preset time; A data compensation module, configured to correct the perturbation of the baseline by combining the real-time monitoring data of the production line; An interval prediction module, configured to expand the corrected dynamic baseline into a probability interval to represent the possible displacement range of the target workpiece within a preset time; A model parameter adjustment module, which is configured to continuously collect the actual displacement data of the target workpiece during the prediction execution process and compare it with the predicted value, and adjust the model parameters through the comparison error feedback.

7. The assembly robot applied to an industrial production line according to claim 6, characterized in that, The specific execution steps of the path simulation module include: Step S21: Separate the displacement trend from the periodic fluctuations; Step S22: Insert correction points into the baseline based on the mutation point detection results; Step S23: Fuse the displacement trend component, the periodic component, and the correction points according to weights to generate the baseline; Step S24: Based on the baseline, simulate the displacement path of the target workpiece in the next N seconds, and compare it with the actual displacement data in real time. If the deviation between the displacement path and the actual value exceeds the threshold, trigger the baseline parameters to reduce the weight of the periodic component.

8. The assembly robot applied to an industrial production line according to claim 6, characterized in that, The specific execution steps of the model parameter adjustment module include: Step S31: During the prediction execution, synchronously collect the actual displacement data and the predicted value of the target workpiece, and align them according to the time stamp; Step S32: Divide the deviation between the predicted value and the actual displacement data into a trend error and a sudden error; Step S33: Adjust the sensitivity of the model parameters according to the stratification results of the trend error and the sudden error; Step S34: Apply the adjusted model parameters to the next round of prediction, and monitor the matching degree between the actual displacement of the target workpiece and the predicted value.

9. The assembly robot applied to an industrial production line according to claim 8, characterized in that, The specific execution steps of Step S33 include: Step S331: Set the priority of the model parameter adjustment according to the ratio of the trend error to the sudden error; Step S332: Set dynamic response thresholds for the trend error and the sudden error respectively; Step S333: Allocate the model parameter adjustment amplitude proportionally based on the priority and the dynamic response threshold.

10. The assembly robot applied to an industrial production line according to claim 1, characterized in that, The exception handling unit includes: A threshold trigger detection module, configured to monitor in real time whether the displacement of the target workpiece exceeds a set threshold within the preset number of times; A production line linkage pause module, configured to send a pause command to the production line control system according to the exception signal triggered by the threshold trigger detection module, and synchronously stop the conveyor belt; A standby actuator intervention module, configured to activate the standby robot end effector and plan a reset path according to the current displacement state of the target workpiece.

Images