Error compensation method and system for industrial robot

Abstract

The invention belongs to the technical field of industrial robot control, and provides an error compensation method and system for an industrial robot, and the method comprises the steps: collecting a composite sensing signal and a pre-tightening force of the industrial robot, and constructing a contribution degree matrix; an error transfer link is extracted, a space-time correlation function is constructed based on the link, and a pretightening force attenuation dominant area is determined through a dynamic cloud atlas algorithm; gradient change features and attenuation trend coefficients of the attenuation dominant region are extracted, a to-be-compensated region is screened, and the saturation compensation risk of the execution mechanism is judged; and if the saturation risk exists, screening pretightening force sensitive joints to construct a compensation distribution model, outputting a segmented compensation demand quantity, and dynamically calibrating the segmented compensation quantity based on a real-time compensation result. The system comprises a contribution analysis module, an attenuation positioning module, a saturation analysis module and a compensation calibration module. The method is beneficial to positioning an error source dominated by pretightening force attenuation, and is beneficial to improving compensation precision and stability.

Description

Technical Field

[0001] This invention belongs to the field of industrial robot control technology, specifically an error compensation method and system for industrial robots. Background Technology

[0002] In high-precision operations such as welding and assembly, the end-effector positioning accuracy of industrial robots directly affects product processing quality, and the attenuation of joint preload is a key factor leading to the accumulation of positioning errors. As the operating time of industrial robots increases, factors such as rolling element impacts, temperature fluctuations, and changes in contact stress in the joint bearings gradually cause preload attenuation, which is then amplified through the joint kinematic chain, ultimately causing the end-effector positioning accuracy to deviate from the allowable threshold and affecting the work performance.

[0003] When determining the error region corresponding to preload decay, existing technologies mostly divide the region based on static error values, without combining the robot joint topology (such as the error propagation characteristics of serial joints) to construct an error propagation model. This fails to present the spatiotemporal distribution law of preload decay in the workspace, which can easily lead to blind selection of compensation objects and overcompensation or undercompensation.

[0004] Therefore, the present invention provides an error compensation method and system for industrial robots. Summary of the Invention

[0005] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0006] The technical solution adopted by this invention to solve its technical problem is: an error compensation method and system for an industrial robot, comprising: An error compensation method for industrial robots includes the following steps: Collect composite sensor signals and preload force of industrial robot, extract time-domain parameters of composite sensor signals and perform correlation analysis with preload force to obtain independent interference parameters, perform weighted contribution quantification on independent interference parameters and construct contribution matrix; Based on the contribution matrix and the robot joint topology, an error propagation model is established to extract the error propagation link. Based on the error propagation link, a spatiotemporal correlation function is constructed, and the preload decay dominance region is determined by inputting the spatiotemporal correlation function with the dynamic cloud map algorithm. Extract the gradient change characteristics and attenuation trend coefficient of the attenuation-dominant region, and select the interval to be compensated based on the gradient change characteristics; perform saturation compensation analysis on the attenuation trend coefficient to determine whether the actuator of the industrial robot is at risk of saturation compensation. If there is a risk of saturation compensation in the actuator, the preload sensitive joint in the error propagation link is extracted, a compensation allocation model and a compensation allocation dataset are constructed, the compensation allocation dataset is input into the compensation allocation model to obtain the segmented compensation requirement of the preload sensitive joint, the real-time segmented compensation result is obtained, and the segmented compensation amount is dynamically calibrated.

[0007] Furthermore, the method for performing the weight contribution quantification is as follows: Collect the values ​​of independent interference parameters for each working condition combination and each preload level, as well as the measured attenuation values ​​for each preload level. Based on the measured attenuation values ​​and independent disturbance parameter values ​​under each operating condition combination, a parameter attenuation dataset is constructed. Construct a multiple linear regression model and solve for the weights of the independent disturbance parameters; The determination coefficient of the multiple regression model is calculated based on the weights. The reliability of the weights is determined based on the determination coefficient. If the weights are reliable, they are retained; otherwise, the weights under the corresponding working conditions are discarded. The contribution of each independent disturbance parameter is obtained by normalizing the credible weights of all independent disturbance parameters under the same operating condition.

[0008] Furthermore, the independent interference parameters are obtained as follows: Extract the time-domain parameters of each signal within the sliding window in the composite sensing signal, and construct a time-domain parameter set; By setting the preload of industrial robot bearings at different gears under the same working conditions, the coefficient of variation of the same time-domain parameter of each signal in the time-domain parameter set with the preload at different gears is calculated. The time-domain parameters of each signal are filtered based on the coefficient of variation to extract signals with strong correlation. A blind source separation model is constructed by inputting signals with strong correlations into the blind source separation model to obtain independent interference parameters.

[0009] Furthermore, the error propagation model is constructed as follows: Extract the mechanical physical quantities corresponding to the independent disturbance parameters and establish a mapping table between independent disturbance parameters and mechanical physical quantities. Obtain the joint topology of the robot and extract the error propagation path diagram by combining the mapping table of independent disturbance parameters and mechanical physical quantities; Based on the contribution matrix, the amplification determination coefficient of each independent interference parameter of each joint is calculated in the error propagation path diagram. Based on the amplification determination coefficient, it is determined whether the independent interference parameter of the joint has an amplification effect. If it does, the error propagation link guided by the preload attenuation is extracted.

[0010] Furthermore, the method for extracting the error propagation link guided by the preload attenuation is as follows: Obtain the current working condition and topology amplification factor of each joint of the industrial robot. If the independent disturbance parameters of the current working condition are temperature drift parameters and vibration harmonic parameters; The contribution of each joint temperature drift parameter and the contribution of each vibration harmonic parameter are multiplied by the topology amplification factor to obtain the amplification determination coefficient of each independent disturbance parameter in each joint. If it is determined that the independent disturbance parameters of the joint have an amplification effect, the mechanical physical quantities corresponding to the independent disturbance parameters dominated by the joint are extracted. By eliminating joints in the error propagation path diagram that do not have an amplification effect, and combining mechanical physical quantities to extract the error propagation link of preload attenuation, the error propagation model is constructed.

[0011] Furthermore, the spatiotemporal correlation function is constructed as follows: Obtain the contribution of the dominant independent disturbance parameters of the joint nodes, the cumulative attenuation of the joint nodes, and the topology amplification factor of the joint nodes. Construct an end-point error accumulation equation, inputting the contribution of the dominant independent interference parameter of the joint node, the cumulative attenuation of the joint node, and the topology amplification factor of the joint node into the end-point error accumulation equation to obtain the cumulative end-point error of the error propagation link. Obtain the topology correction coefficients, multiply the topology correction coefficients and the terminal error accumulation to construct the spatiotemporal correlation function.

[0012] Furthermore, the method to determine whether there is a risk of saturation compensation is as follows: The difference between the maximum positioning error value and the maximum positioning error threshold within the core coordinate range during real-time operation of the industrial robot is obtained to determine the compensation requirement. Based on the dynamic model of industrial robots, equations for joint torque and compensation amount are constructed to obtain the compensation torque. Obtain the transmission ratio of the industrial robot joint reducer, and process the ratio of the compensation requirement to the transmission ratio to obtain the compensation angle. Compensation torque and compensation angle are used as criteria factors for industrial robots; A saturation risk criterion is constructed. If the criterion factor satisfies the saturation risk criterion, it is determined that the industrial robot has a risk of saturation compensation of the actuator.

[0013] Furthermore, the method for determining the coordinate range of the core interval is as follows: Obtain the key working area of ​​the industrial robot, and calculate the angle between the current coordinate point of the industrial robot and the point pointing to the key working area, as the working motion angle; Calculate the absolute deviation between the gradient direction angle and the operation motion angle, and determine whether the gradient direction points to the critical operation area based on the absolute motion deviation; Obtain the gradient rate of change and gradient direction of each coordinate point in the preload decay zone, extract priorities, and obtain the first-level priority points. The first-priority points and their adjacent first-priority points are aggregated into a continuous coordinate range, which is the first-priority interval. All first-priority intervals are treated as intervals to be compensated, and the core interval coordinate range of the intervals to be compensated is output.

[0014] Furthermore, the method for obtaining the segmented compensation demand is as follows: The amplification decision coefficients of the dominant independent interference parameters of each joint in the error propagation link are obtained, and the joints are sorted in descending order based on the amplification decision coefficients. The joint with the highest amplification decision coefficient is selected as the preload sensitive joint. Obtain the total compensation requirement and spatial gradient change rate of the preload-sensitive joint, and combine them with historical compensation allocation data of the preload-sensitive joint; The compensation allocation dataset is constructed by combining the compensation allocation data with the total compensation demand and spatial gradient change rate of the current preload sensitive joint; A compensation allocation model is constructed using the random forest regression algorithm. The compensation allocation dataset is input into the compensation allocation model, and the model outputs the segmented compensation requirements of the preload sensitive joint.

[0015] An error compensation system for an industrial robot includes the following modules: Contribution Analysis Module: Used to collect composite sensor signals and preload of industrial robots, extract time-domain parameters of composite sensor signals and perform correlation analysis with preload to obtain independent interference parameters, perform weighted contribution quantification on independent interference parameters and construct contribution matrix; Attenuation positioning module: Based on the contribution matrix, it is used to establish an error propagation model by combining the robot joint topology to extract the error propagation link, construct a spatiotemporal correlation function based on the error propagation link, and combine the dynamic cloud map algorithm to input the spatiotemporal correlation function to determine the attenuation dominance area of ​​the preload. Saturation Analysis Module: Used to extract the gradient change characteristics and attenuation trend coefficient of the attenuation-dominant region, filter the intervals to be compensated based on the gradient change characteristics, and perform saturation compensation analysis on the attenuation trend coefficient to determine whether the actuator of the industrial robot has a saturation compensation risk. Compensation and calibration module: If there is a risk of saturation compensation of the actuator, the preload sensitive joint in the error transmission link is extracted, a compensation allocation model and a compensation allocation dataset are constructed, the compensation allocation dataset is input into the compensation allocation model to obtain the segmented compensation requirement of the preload sensitive joint, the real-time segmented compensation result is obtained, and the segmented compensation amount is dynamically calibrated.

[0016] The beneficial effects of this invention are as follows: 1. By collecting composite sensor signals such as mechanical vibration, temperature change, and micro-strain of industrial robot joints, and combining them with preload, correlation analysis is performed to extract independent interference parameters. A contribution matrix is ​​constructed through multiple linear regression and normalization, which helps to identify key factors strongly correlated with preload attenuation, quantify the influence weight of different interference parameters on preload attenuation, provide basic data support for subsequent positioning error sources, and adapt to the preload attenuation analysis needs of industrial robots under different load and speed conditions.

[0017] 2. An error propagation model is constructed based on the contribution matrix and the robot joint topology. The error propagation link guided by preload attenuation is extracted. The attenuation-dominant region is determined by combining the spatiotemporal correlation function with the dynamic cloud map algorithm. This is beneficial to combine the mechanical physical quantities corresponding to independent interference parameters with the error propagation characteristics of the joint kinematic chain, reflecting the spatiotemporal distribution law of preload attenuation in the workspace. This makes the identification of the attenuation region more consistent with the actual motion logic of industrial robots and adapts to the error propagation analysis scenarios of common serial topology industrial robot structures.

[0018] 3. Extract the gradient change characteristics and attenuation trend coefficient of the attenuation-dominant area, screen the compensation interval based on the gradient change characteristics, and perform saturation compensation analysis on the attenuation trend coefficient to determine the risk of the actuator. This is beneficial for analyzing compensation objects with significant error changes and high correlation with key operational areas, and also allows for early assessment of the impact of compensation actions on the actuator's load. This provides a priority basis and safety reference for the formulation of subsequent compensation strategies, reducing resource waste or excessive load on the actuator caused by blind compensation.

[0019] 4. When there is a risk of saturation, select joints sensitive to preload, construct a compensation allocation model using a random forest regression algorithm to output segmented compensation requirements, and then calculate the effect coefficient based on the real-time compensation results to dynamically calibrate the compensation amount. This can concentrate compensation resources on joints that significantly affect the preload attenuation error, generate a segmented scheme that adapts to the actuator's load-bearing capacity, and adjust subsequent actions according to the actual compensation effect. This makes the compensation process more consistent with the actual capacity and error changes of the actuator, which helps to improve the stability of compensation accuracy and reduce actuator wear. Attached Figure Description

[0020] The invention will now be further described with reference to the accompanying drawings.

[0021] Figure 1 This is a flowchart of an error compensation method for an industrial robot according to an embodiment of the present invention; Figure 2 This is a flowchart illustrating the determination process for the strong correlation between the existence of time-domain parameters and the attenuation of preload, as described in this invention. Figure 3This is a module architecture diagram of an error compensation system for an industrial robot according to an embodiment of the present invention. Detailed Implementation

[0022] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0023] Example 1

[0024] Please see Figure 1 As shown in the embodiment of the present invention, an error compensation method for an industrial robot includes the following steps: S1. Collect composite sensor signals and preload force of industrial robot, extract time-domain parameters of composite sensor signals and perform correlation analysis with preload force to obtain independent interference parameters, perform weighted contribution quantification on independent interference parameters and construct contribution matrix.

[0025] The method for collecting composite sensor signals and preload force from industrial robots, extracting time-domain parameters of the composite sensor signals, and performing correlation analysis in conjunction with the preload force is as follows: Preferably, a piezoelectric accelerometer is installed in the joint bearing housing of the industrial robot to collect mechanical vibration signals reflecting the contact impact between the rolling element and the inner and outer rings; Temperature values ​​at different sampling points in the joint area are collected by an infrared array sensor, and the temperature change signal at each sampling point is extracted to reflect the change in thermal preload. By attaching strain gauges to the outer ring of the bearing in the joint area of ​​an industrial robot, micro-strain signals (i.e., contact strain force) reflecting contact pressure fluctuations are collected. Mechanical vibration signal, temperature change signal, and micro-strain signal are collectively referred to as composite sensing signal; Extract the time-domain parameters of each signal within the sliding window in the composite sensing signal, and construct a time-domain parameter set; By setting the preload of industrial robot bearings at different gears under the same working conditions, the coefficient of variation of the same time-domain parameter of each signal in the time-domain parameter set with the preload at different gears is calculated. The time-domain parameters for each signal include: peak value, root mean square value, and kurtosis of the mechanical vibration signal; Temperature change rate and temperature fluctuation variance of the temperature change signal; Peak strain and mean strain of micro-strain signal; If for each signal there is at least one time-domain parameter whose coefficient of variation of preload is higher than or equal to a preset attenuation correlation threshold, and the coefficient of variation increases as the preload attenuates, then it is determined that there is a strong correlation between the time-domain parameter and the preload attenuation. Among them, the preset attenuation correlation threshold corresponds one-to-one with each signal time domain parameter, that is, the coefficient of variation of the preload of a time domain parameter has a corresponding attenuation correlation threshold. like Figure 2 As shown, if the coefficient of variation of the time-domain parameter is lower than the preset decay correlation threshold, or if it shows a decreasing trend with the decay of the preload, no processing is required. It is understandable that the attenuation correlation threshold is preset by those skilled in the art based on historical test data of industrial robots under different load ratios and speed ratios. The minimum coefficient of variation when the time domain parameters and preload attenuation are strongly correlated in the data is statistically analyzed and used as the attenuation correlation threshold to adapt to the determination requirements of strong correlation under actual working conditions. If the time-domain parameters of each signal have a strong correlation with the preload decay, then the signal with the strong correlation is extracted. A blind source separation model is constructed, and signals with strong correlations are input into the blind source separation model to obtain independent interference parameters; For example, if the mechanical vibration signal, temperature change signal, and micro-strain signal are all strongly correlated with the preload attenuation, then the corresponding independent interference parameters are: temperature drift parameter, vibration harmonic parameter, and micro-strain pulse parameter. It should be noted that the blind source separation algorithm adapted to FastICA is selected to construct the blind source separation model. The mechanical vibration, temperature change, and micro-strain signals that are strongly correlated with the preload attenuation are input into the model to separate the independent components corresponding to the preload attenuation in the mixed signal, that is, to obtain independent interference parameters such as temperature drift, vibration harmonics, and micro-strain pulses. Temperature drift parameters are used to reflect the preload offset caused by temperature changes; vibration harmonic parameters are used to reflect the preload fluctuation amplitude caused by periodic impacts of rolling elements; micro-strain pulse parameters are used to characterize the preload pulse change caused by instantaneous contact stress abrupt changes. The method for quantifying the weighted contributions of independent interference parameters and constructing the contribution matrix is ​​as follows: Based on the actual operating scenarios of industrial robots, N combinations of working conditions with different load ratios and speed ratios are designed, and K preload levels are set for each combination of working conditions. It should be noted that the load ratio is the ratio of the actual load to the rated load under operating conditions, and the speed ratio is the ratio of the actual speed to the rated speed under operating conditions. Collect the values ​​of independent interference parameters for each working condition combination and each preload level, as well as the measured attenuation values ​​for each preload level. ; Based on the measured attenuation values ​​and independent disturbance parameter values ​​under each operating condition combination, a parameter attenuation dataset is constructed. Through the formula: Construct a multiple linear regression model and solve for the weights w of the independent disturbance parameters; in, , ... For each independent interference parameter, for example For temperature drift parameters, For vibration harmonic parameters, This is the nth independent interference parameter; , ... The weights to be determined for each independent disturbance parameter. This represents the error term in the multiple linear regression model. Input the parameter decay dataset into the multiple linear regression model, and solve for the weights to be determined under each working condition using the least squares method. The determination coefficient of the multiple regression model is calculated based on the weights to be determined. The reliability of the weights is determined based on the determination coefficient. If the weights are reliable, they are retained; otherwise, the weights under the corresponding working conditions are discarded. For example, calculate the adjusted coefficient of determination. ,like If so, the weights are considered reliable; The contribution of each independent disturbance parameter is obtained by normalizing the credible weights of all independent disturbance parameters under the same working condition. A contribution matrix is ​​constructed based on the contribution of each independent disturbance parameter under each set of operating conditions.

[0026] S2. Based on the contribution matrix, an error propagation model is established by combining the robot joint topology to extract the error propagation link. Based on the error propagation link, a spatiotemporal correlation function is constructed, and the spatiotemporal correlation function is input by combining the dynamic cloud map algorithm to determine the dominant area of ​​preload attenuation. The error propagation model is constructed as follows: S201. Extract the mechanical physical quantities corresponding to the independent disturbance parameters and establish a mapping table of independent disturbance parameters and mechanical physical quantities. Preferably, the mechanical physical quantities of the independent interference parameters are correlated to establish a mapping table between independent interference parameters and mechanical physical quantities. Those skilled in the art will understand that if the independent disturbance parameter is any one or more of the temperature drift parameter, vibration harmonic parameter, and micro-strain pulse parameter; Temperature drift corresponds to thermal deformation of the joints of an industrial robot. Joint thermal deformation leads to an increase in joint clearance. Therefore, the mechanical physical quantity associated with the temperature drift parameter is the joint clearance value. Vibration harmonics correspond to rolling element impact. Rolling element impact causes changes in bearing stiffness. Therefore, the mechanical and physical quantity associated with the vibration harmonic parameters is the change in bearing stiffness. Micro-strain pulses correspond to contact stress fluctuations, and contact stress fluctuations lead to stiffness fluctuations. Therefore, the mechanical physical quantity associated with the micro-strain pulse parameters is the stiffness fluctuation quantity. S202. Obtain the joint topology of the robot and extract the error propagation path diagram by combining the mapping table of independent disturbance parameters and mechanical physical quantities. Preferably, the topology type of the industrial robot is obtained, and the error propagation path is extracted according to the movement sequence of the industrial robot; For example, if the industrial robot is a serial topology, the error transmission path is obtained according to the joint movement sequence, that is, from joint 1 to joint 2 to the final end effector. The error of the preceding joint is transmitted to the subsequent joint through the force-displacement coupling relationship, amplifying the execution error of the end effector. Mark the joint nodes of the mechanical physical quantities corresponding to the independent disturbance parameters in the error propagation path, and construct the error propagation path diagram; S203. Calculate the amplification determination coefficient of each independent interference parameter in each joint within the error propagation path diagram based on the contribution matrix. Determine whether the independent interference parameter of the joint has an amplification effect based on the amplification determination coefficient. If it does, extract the error propagation link guided by the preload attenuation. Preferably, the current working condition and topology amplification factor of each joint of the industrial robot are obtained, if the independent disturbance parameters of the current working condition are temperature drift parameters and vibration harmonic parameters; The contribution of each joint temperature drift parameter and the contribution of each vibration harmonic parameter are multiplied by the topology amplification factor to obtain the amplification determination coefficient of each independent disturbance parameter in each joint. If the amplification judgment coefficient of the independent interference parameter is higher than or equal to the preset amplification judgment threshold, then the independent interference parameter of the joint is determined to have an amplification effect. It is understandable that the amplification judgment threshold is preset by those skilled in the art based on historical error propagation test data of industrial robots under typical load and speed conditions. This is done by statistically analyzing the product of the minimum contribution of joint independent interference parameters (such as temperature drift and vibration harmonics) to cause significant error amplification and the topological amplification coefficient, which will screen out joints that have a significant impact on end-effector error. If the amplification judgment coefficient of the independent interference parameter is lower than the preset amplification judgment threshold, then the amplification effect of the independent interference parameter of the joint is not within the expected range. If the independent disturbance parameters of the joint have an amplification effect, extract the mechanical physical quantities corresponding to the independent disturbance parameters dominated by the joint. It should be noted that if the amplification determination coefficient of the temperature drift parameter is higher than that of the vibration harmonic parameter, the mechanical physical quantity corresponding to the temperature drift parameter is extracted. Since the higher the amplification determination coefficient, the more significant the amplification effect of the mechanical physical quantity corresponding to the parameter on the end error, the mechanical physical quantity corresponding to the independent interference parameter with the higher amplification determination coefficient is extracted first. After removing joint nodes in the error propagation path diagram where the amplification effect is not expected (i.e., joint nodes where there is no amplification effect), the error propagation link of preload attenuation is extracted by combining mechanical physical quantities, and the error propagation model is constructed. It should be noted that the method for extracting the error propagation link of preload attenuation is as follows: the error propagation link is reconstructed according to the logic of preload attenuation - mechanical physical quantity with amplification effect - error propagation: preload attenuation - increased joint 1 clearance - decreased joint 2 stiffness - error accumulation along the path - end-positioning error; the mechanical physical quantity of each joint node and the contribution ratio of the dominant independent disturbance parameter corresponding to the mechanical physical quantity are obtained from the error propagation link (extracted from the contribution matrix, such as the contribution ratio of joint 1 clearance is 35%, and the contribution ratio of joint 2 stiffness attenuation is 45%). The method for determining the preload attenuation dominance zone by constructing a spatiotemporal correlation function and combining it with a dynamic cloud map algorithm is as follows: Preferably, the terminal error accumulation equation is used: Obtain the cumulative error at the end of the error propagation chain. ; Where i is the number of the joint node in the error propagation path. The contribution of the dominant independent disturbance parameter of the i-th joint node; Let t be the topology scaling factor for the i-th joint node, and t be the running time. This represents the cumulative attenuation of the i-th joint node; It should be noted that the topology amplification factor is used to quantify the amplification effect of joint errors transmitted along the kinematic chain to the end effector, by combining the serial topology characteristics of industrial robots (such as joint 1-joint 2-end effector structure): The error injection method is used to apply a known small error (such as a gap of 0.01 mm) to the target joint. The change in end-effector error is measured, and the ratio of the two is the topology amplification factor of the joint (for example, if a gap of 0.01 mm is injected into joint 1, the end-effector positioning error increases by 0.02 mm, then the topology amplification factor is 2). The cumulative attenuation of the i-th joint node is expressed by the formula: Obtain; is the attenuation value of the initial preload of the i-th joint node, k is the preset cumulative coefficient, m is the attenuation exponent, and t is the running time; It should be noted that the cumulative coefficient k and the attenuation index m are obtained by first collecting measured data of the cumulative preload attenuation of the industrial robot under different load ratios and speed ratios and different running times. Then, based on the formula for the cumulative preload attenuation, the collected running time and the measured data of the attenuation value of the initial preload are substituted into the power function model. The cumulative coefficient and attenuation index suitable for the corresponding working conditions are determined by solving the model using the least squares method. Obtain the cumulative end-effector error and topology correction coefficient A at the (x, y) coordinate points in the industrial robot's workspace; Through the formula: Constructing spatiotemporal correlation functions Calculate each Point error value ; It should be noted that the topology correction factor A was obtained from the industrial robot's factory calibration data; Based on the dynamic cloud map algorithm, the spatiotemporal correlation function is input into the dynamic cloud map algorithm to obtain the preload attenuation dominant region.

[0027] Those skilled in the art will understand that after inputting the spatiotemporal correlation function into the dynamic cloud map algorithm, the algorithm first outputs the end-positioning error values ​​of all coordinate points in the industrial robot's workspace at different running times through the algorithm's analytical function. At the same time, the weight ratio of each independent interference parameter in the contribution matrix to the preload attenuation is called to calculate the proportion of the error component caused by preload attenuation in the total error of each (x,y,t) point error (i.e., the contribution ratio of preload attenuation). The algorithm uses two preset judgment thresholds: an error value threshold (80% of the maximum allowable positioning error under working conditions to filter high error areas) and a contribution ratio threshold (e.g., 60% to ensure that the area error is mainly dominated by preload attenuation, rather than geometric errors, motor errors, or other factors). The algorithm then traverses all spatiotemporal points in the workspace, filters out coordinate points where the error value is greater than or equal to the error threshold and the contribution ratio of preload attenuation is greater than or equal to the contribution ratio threshold, and dynamically maps these points to the workspace visualization interface in a time series. The distribution of attenuation areas is presented with color gradients, ultimately forming a dynamic cloud map that changes over time. The continuously highlighted red area in the dynamic cloud map is the preload attenuation-dominated area.

[0028] Example 2

[0029] Please see Figure 1 As shown in the embodiment of the present invention, an error compensation method for an industrial robot further includes the following steps: S3. Extract the gradient change characteristics and attenuation trend coefficient of the attenuation-dominant region, and select the interval to be compensated based on the gradient change characteristics; perform saturation compensation analysis on the attenuation trend coefficient to determine whether the actuator of the industrial robot has a risk of saturation compensation. The method for extracting the gradient change characteristics and attenuation trend coefficient of the attenuation-dominant region is as follows: Obtain each of the decay-dominant regions Point error value Calculate the partial derivative of the error in the x-direction partial derivatives in the y-direction ; Through formula one: Obtain the gradient rate of change G; Through formula two: Obtain the gradient direction angle ; The gradient rate of change and the gradient direction angle are used as characteristics of gradient change. All within the attenuation-dominant region Point error value An error decay sequence is constructed, and an exponential fit is performed on the error decay sequence to obtain the fitting equation and extract the fitting coefficient as the decay trend coefficient. Preferably, by formula: The error decay sequence is subjected to exponential fitting, and the fitting coefficient d is obtained as the decay trend coefficient. It should be noted that c is the amplitude coefficient solved in the exponential fitting process, f is the constant term solved in the exponential fitting process, e is the natural constant, and t is the running time of the industrial robot. The method for selecting the compensation interval based on gradient change characteristics is as follows: Obtain the coordinates of key operating areas of industrial robots (e.g., welding points, assembly points). ; Calculate the current coordinates of the industrial robot The angle pointing towards the critical area of ​​the operation is used as the operation motion angle; Calculate the absolute deviation between the gradient direction angle and the operation motion angle, and determine whether the gradient direction points to the critical operation area based on the absolute motion deviation; Preferably, the method for determining whether the gradient direction points to the critical area of ​​the operation based on the absolute motion deviation is as follows: if the absolute deviation between the gradient direction angle and the operation motion angle is less than 30 degrees, the gradient direction is determined to point to the critical area (increasing the error will quickly affect the operation accuracy); otherwise, it is considered to deviate from the critical area. Determine the gradient change rate and gradient direction of each (x, y) point within the preload decay zone, whether they point to the critical work area, determine the priority of the compensation interval, and filter the intervals to be compensated. Preferably, the method for determining the priority of compensation intervals and filtering the intervals to be compensated is as follows: If the gradient change rate is higher than the preset high gradient threshold (e.g., 0.5 mm / mm) and the gradient direction points to the critical area of ​​the operation, it corresponds to a first-level priority point. If the gradient change rate is higher than the preset high gradient threshold, or if the gradient direction points away from the critical work area, then it corresponds to a secondary priority point. If the gradient change rate is lower than the preset high gradient threshold and the gradient direction points away from the critical area of ​​the operation, it corresponds to a level three priority point. The first-priority points and their adjacent first-priority points are aggregated into a continuous (x,y) coordinate range, which is the first-priority interval. All first-priority intervals are treated as intervals to be compensated, and the core interval coordinate range of the intervals to be compensated is output. The method for determining whether the actuator of an industrial robot has a risk of saturation compensation by performing saturation compensation analysis on the attenuation trend coefficient is as follows: Based on the attenuation trend coefficient within the compensation interval and the maximum allowable error threshold of the industrial machine, input the fitting equation and solve for the remaining time. By obtaining the difference between the maximum positioning error value and the maximum positioning error threshold within the core coordinate range during real-time operation of the industrial robot, the compensation requirement can be determined. ; The compensation demand rate is obtained by comparing the compensation demand with the remaining time. ; Based on the dynamic model of industrial robots, the equations for joint torque and compensation are constructed as follows: Obtain the compensation torque T; Where ka is the joint force-displacement coefficient and b is the joint damping coefficient; Understandably, the joint force-displacement coefficient ka is determined by collecting measured output torque data of industrial robot joints under different compensation amounts and fitting the linear correlation characteristics between torque and compensation amount. It is used to quantify the correspondence between compensation displacement and joint output torque. The joint damping coefficient b is obtained by collecting damping torque data of joints under different compensation demand rates, substituting it into the joint torque and compensation amount equations, and then fitting and solving it using the least squares method. It is used to characterize the degree of influence of joint motion rate on damping torque. Obtain the transmission ratio of the industrial robot joint reducer, and process the ratio of the compensation requirement to the transmission ratio to obtain the compensation angle. Compensation torque and compensation angle are used as criteria factors for industrial robots; A saturation risk criterion is constructed. If the criterion factor satisfies the saturation risk criterion, it is determined that the industrial robot has a risk of saturation compensation of the actuator. Preferably, the method for constructing the saturation risk criterion is as follows: The rated physical limit parameters of the industrial robot actuator (such as the maximum torque of the motor, the maximum angle of the reducer, and the maximum compensation rate of the system) are reserved with a safety margin (the safety margin is usually 20% of the corresponding parameter to avoid damage to the mechanism under extreme conditions), and the safety threshold is obtained, which is the saturation risk criterion. If the compensation demand rate If any one of the compensation torque and compensation angle is higher than or equal to the corresponding preset safety threshold, it is determined that there is a risk of saturation compensation of the actuator; If the compensation demand rate If both the compensation torque and compensation angle are lower than the corresponding preset safety thresholds, then there is no risk of actuator saturation compensation for the time being.

[0030] S4. If there is a risk of saturation compensation of the actuator, extract the preload sensitive joint in the error transmission link, construct a compensation allocation model and a compensation allocation dataset, input the compensation allocation dataset into the compensation allocation model to obtain the segmented compensation requirement of the preload sensitive joint, obtain the real-time segmented compensation result and dynamically calibrate the segmented compensation amount. The method for extracting the preload-sensitive joints in the error propagation link and constructing the compensation allocation model and compensation allocation dataset is as follows: The amplification decision coefficients of the dominant independent interference parameters of each joint in the error propagation link are obtained, and the joints are sorted in descending order based on the amplification decision coefficients. The joint with the highest amplification decision coefficient is selected as the preload sensitive joint. Obtain the total compensation requirement and spatial gradient change rate of the preload-sensitive joint, and combine them with historical compensation allocation data of the preload-sensitive joint; The compensation allocation data includes: the compensation requirement of the preload sensitive joint, the spatial gradient change rate, and the mapping relationship between the safety splitting scheme; The compensation allocation dataset is constructed by combining the compensation allocation data with the total compensation demand and spatial gradient change rate of the current preload sensitive joint; A compensation allocation model is constructed using a random forest regression algorithm. The compensation allocation dataset is input into the compensation allocation model, and the model outputs the segmented compensation requirements of the preload sensitive joint. Those skilled in the art will understand that the way to construct the compensation allocation model is as follows: first, the compensation allocation dataset is preprocessed to remove abnormal compensation records caused by sensor anomalies or actuator failures in historical data (such as data where the single compensation amount exceeds the maximum carrying capacity of the actuator), and then the total compensation demand and the spatial gradient change rate are normalized. Using the total compensation demand and spatial gradient change rate of the current preload sensitive joint as the input labels of the model, the compensation demand of each segment in the historical safety split scheme is output; then the random forest regression model is initialized, the number of decision trees is set to 100 and the maximum tree depth is 8, and the parameters are optimized by 5-fold cross-validation to reduce overfitting. The preprocessed compensation allocation dataset is divided into a training set and a test set in a 7:3 ratio. The training set is used to train the model, and the test set is used to verify the model performance. After training, the total compensation demand of the current preload sensitive joint and the spatial gradient change rate are input into the model. Based on the safety splitting logic corresponding to similar features in historical data, the model outputs the segmented compensation demand that satisfies the condition that the single segmented compensation amount is ≤ the dynamic safety threshold of the actuator, such as 80% of the maximum torque of the motor, 80% of the maximum rotation angle of the reducer, and the sum of the compensation amounts of each segment is equal to the total compensation demand. The method for obtaining real-time segmented compensation results and dynamically calibrating the segmented compensation amount is as follows: Based on the segmented compensation requirements of the preload-sensitive joint, segmented compensation is performed on the preload joint; Acquire the composite sensing signal after each segmented compensation, and extract the actual executed value of the compensation amount after segmented execution. and the error residuals of sensitive joints after execution. ; Through the formula: Obtain the compensation effect coefficient ; It should be noted that, To compensate for the preset output value of the allocation model, The actual rotation angle during segmented compensation is obtained by acquiring the encoder built into the preload sensitive joint, and then the transmission ratio of the joint reducer is called. The actual compensation displacement is obtained by multiplying the transmission ratio and the actual rotation angle. The residual error after segmented compensation is obtained by calculating the current end-point positioning error of the (x,y) coordinate point of the sensitive joint corresponding to the compensation interval after compensation using the spatiotemporal correlation function E(x,y,t). The compensation effect coefficient is used to analyze whether the compensation effect meets the standard. If it does not meet the standard, the ratio of the theoretical compensation demand to the compensation effect coefficient is calculated to obtain the revised compensation demand for the next stage. The compensation amount for the next stage is then adjusted based on the revised compensation demand for the next stage.

[0031] Example 3

[0032] Please see Figure 3 As shown in the figure, an error compensation system for an industrial robot according to an embodiment of the present invention includes the following modules: Contribution Analysis Module: Used to collect composite sensor signals and preload of industrial robots, extract time-domain parameters of composite sensor signals and perform correlation analysis with preload to obtain independent interference parameters, perform weighted contribution quantification on independent interference parameters and construct contribution matrix; Attenuation positioning module: Based on the contribution matrix, it is used to establish an error propagation model by combining the robot joint topology to extract the error propagation link, construct a spatiotemporal correlation function based on the error propagation link, and combine the dynamic cloud map algorithm to input the spatiotemporal correlation function to determine the attenuation dominance area of ​​the preload. Saturation Analysis Module: Used to extract the gradient change characteristics and attenuation trend coefficient of the attenuation-dominant region, filter the intervals to be compensated based on the gradient change characteristics, and perform saturation compensation analysis on the attenuation trend coefficient to determine whether the actuator of the industrial robot has a saturation compensation risk. Compensation and calibration module: If there is a risk of saturation compensation of the actuator, the preload sensitive joint in the error transmission link is extracted, a compensation allocation model and a compensation allocation dataset are constructed, the compensation allocation dataset is input into the compensation allocation model to obtain the segmented compensation requirement of the preload sensitive joint, the real-time segmented compensation result is obtained, and the segmented compensation amount is dynamically calibrated.

[0033] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. An error compensation method for an industrial robot, characterized in that: Includes the following steps: Collect composite sensor signals and preload force of industrial robot, extract time-domain parameters of composite sensor signals and perform correlation analysis with preload force to obtain independent interference parameters, perform weighted contribution quantification on independent interference parameters and construct contribution matrix; Based on the contribution matrix and the robot joint topology, an error propagation model is established to extract the error propagation link. Based on the error propagation link, a spatiotemporal correlation function is constructed, and the preload decay dominance region is determined by inputting the spatiotemporal correlation function with the dynamic cloud map algorithm. Extract the gradient change characteristics and attenuation trend coefficient of the attenuation-dominant region, and select the interval to be compensated based on the gradient change characteristics. Saturation compensation analysis is performed on the attenuation trend coefficient to determine whether there is a risk of saturation compensation in the actuator of the industrial robot; If there is a risk of saturation compensation in the actuator, the preload sensitive joint in the error propagation link is extracted, a compensation allocation model and a compensation allocation dataset are constructed, the compensation allocation dataset is input into the compensation allocation model to obtain the segmented compensation requirement of the preload sensitive joint, the real-time segmented compensation result is obtained, and the segmented compensation amount is dynamically calibrated.

2. The error compensation method for an industrial robot according to claim 1, characterized in that: The method for performing the weight contribution quantification is as follows: Collect the values ​​of independent interference parameters for each working condition combination and each preload level, as well as the measured attenuation values ​​for each preload level. Based on the measured attenuation values ​​and independent disturbance parameter values ​​under each operating condition combination, a parameter attenuation dataset is constructed. Construct a multiple linear regression model and solve for the weights of the independent disturbance parameters; The determination coefficient of the multiple regression model is calculated based on the weights. The reliability of the weights is determined based on the determination coefficient. If the weights are reliable, they are retained; otherwise, the weights under the corresponding working conditions are discarded. The contribution of each independent disturbance parameter is obtained by normalizing the credible weights of all independent disturbance parameters under the same operating condition.

3. The error compensation method for an industrial robot according to claim 2, characterized in that: The method for obtaining independent interference parameters is as follows: Extract the time-domain parameters of each signal within the sliding window in the composite sensing signal, and construct a time-domain parameter set; By setting the preload of industrial robot bearings at different gears under the same working conditions, the coefficient of variation of the same time-domain parameter of each signal in the time-domain parameter set with the preload at different gears is calculated. The time-domain parameters of each signal are filtered based on the coefficient of variation to extract signals with strong correlation. A blind source separation model is constructed by inputting signals with strong correlations into the blind source separation model to obtain independent interference parameters.

4. The error compensation method for an industrial robot according to claim 1, characterized in that: The error propagation model is constructed as follows: Extract the mechanical physical quantities corresponding to the independent disturbance parameters and establish a mapping table between independent disturbance parameters and mechanical physical quantities. Obtain the joint topology of the robot and extract the error propagation path diagram by combining the mapping table of independent disturbance parameters and mechanical physical quantities; Based on the contribution matrix, the amplification determination coefficient of each independent interference parameter of each joint is calculated in the error propagation path diagram. Based on the amplification determination coefficient, it is determined whether the independent interference parameter of the joint has an amplification effect. If it does, the error propagation link guided by the preload attenuation is extracted.

5. The error compensation method for an industrial robot according to claim 4, characterized in that: The method for extracting the error propagation link of the preload attenuation guide is as follows: Obtain the current working condition and topology amplification factor of each joint of the industrial robot. If the independent disturbance parameters of the current working condition are temperature drift parameters and vibration harmonic parameters; The contribution of each joint temperature drift parameter and the contribution of each vibration harmonic parameter are multiplied by the topology amplification factor to obtain the amplification determination coefficient of each independent disturbance parameter in each joint. If it is determined that the independent disturbance parameters of the joint have an amplification effect, the mechanical physical quantities corresponding to the independent disturbance parameters dominated by the joint are extracted. By eliminating joints in the error propagation path diagram that do not have an amplification effect, and combining mechanical physical quantities to extract the error propagation link of preload attenuation, the error propagation model is constructed.

6. The error compensation method for an industrial robot according to claim 1, characterized in that: The spatiotemporal correlation function is constructed as follows: Obtain the contribution of the dominant independent disturbance parameters of the joint nodes, the cumulative attenuation of the joint nodes, and the topology amplification factor of the joint nodes. Construct an end-point error accumulation equation, inputting the contribution of the dominant independent interference parameter of the joint node, the cumulative attenuation of the joint node, and the topology amplification factor of the joint node into the end-point error accumulation equation to obtain the cumulative end-point error of the error propagation link. Obtain the topology correction coefficients, multiply the topology correction coefficients and the terminal error accumulation to construct the spatiotemporal correlation function.

7. The error compensation method for an industrial robot according to claim 1, characterized in that: The method to determine whether there is a risk of saturation compensation is as follows: The difference between the maximum positioning error value and the maximum positioning error threshold within the core coordinate range during real-time operation of the industrial robot is obtained to determine the compensation requirement. Based on the dynamic model of industrial robots, equations for joint torque and compensation amount are constructed to obtain the compensation torque. Obtain the transmission ratio of the industrial robot joint reducer, and process the ratio of the compensation requirement to the transmission ratio to obtain the compensation angle. Compensation torque and compensation angle are used as criteria factors for industrial robots; A saturation risk criterion is constructed. If the criterion factor satisfies the saturation risk criterion, it is determined that the industrial robot has a risk of saturation compensation of the actuator.

8. The error compensation method for an industrial robot according to claim 7, characterized in that: The method for determining the coordinate range of the core interval is as follows: Obtain the key working area of ​​the industrial robot, and calculate the angle between the current coordinate point of the industrial robot and the point pointing to the key working area, as the working motion angle; Calculate the absolute deviation between the gradient direction angle and the operation motion angle, and determine whether the gradient direction points to the critical operation area based on the absolute motion deviation; Obtain the gradient rate of change and gradient direction of each coordinate point in the preload decay zone, extract priorities, and obtain the first-level priority points. The first-priority points and their adjacent first-priority points are aggregated into a continuous coordinate range, which is the first-priority interval. All first-priority intervals are treated as intervals to be compensated, and the core interval coordinate range of the intervals to be compensated is output.

9. The error compensation method for an industrial robot according to claim 1, characterized in that: The method for obtaining the segmented compensation demand is as follows: The amplification decision coefficients of the dominant independent interference parameters of each joint in the error propagation link are obtained, and the joints are sorted in descending order based on the amplification decision coefficients. The joint with the highest amplification decision coefficient is selected as the preload sensitive joint. Obtain the total compensation requirement and spatial gradient change rate of the preload-sensitive joint, and combine them with historical compensation allocation data of the preload-sensitive joint; The compensation allocation dataset is constructed by combining the compensation allocation data with the total compensation demand and spatial gradient change rate of the current preload sensitive joint; A compensation allocation model is constructed using the random forest regression algorithm. The compensation allocation dataset is input into the compensation allocation model, and the model outputs the segmented compensation requirements of the preload sensitive joint.

10. An error compensation system for an industrial robot, used to implement the error compensation method for an industrial robot according to any one of claims 1-9, characterized in that: Includes the following modules: Contribution Analysis Module: Used to collect composite sensor signals and preload of industrial robots, extract time-domain parameters of composite sensor signals and perform correlation analysis with preload to obtain independent interference parameters, perform weighted contribution quantification on independent interference parameters and construct contribution matrix; Attenuation positioning module: Based on the contribution matrix, it is used to establish an error propagation model by combining the robot joint topology to extract the error propagation link, construct a spatiotemporal correlation function based on the error propagation link, and combine the dynamic cloud map algorithm to input the spatiotemporal correlation function to determine the attenuation dominance area of ​​the preload. Saturation analysis module: used to extract the gradient change characteristics and attenuation trend coefficient of the attenuation-dominant region, and to select the interval to be compensated based on the gradient change characteristics; Saturation compensation analysis is performed on the attenuation trend coefficient to determine whether there is a risk of saturation compensation in the actuator of the industrial robot; Compensation and calibration module: If there is a risk of saturation compensation of the actuator, the preload sensitive joint in the error transmission link is extracted, a compensation allocation model and a compensation allocation dataset are constructed, the compensation allocation dataset is input into the compensation allocation model to obtain the segmented compensation requirement of the preload sensitive joint, the real-time segmented compensation result is obtained, and the segmented compensation amount is dynamically calibrated.

Images