High-rigidity electrically-driven welding robot mechanism based on parallel configuration

Abstract

The invention provides a high-rigidity electrically-driven welding robot mechanism based on a parallel configuration. A controller of the mechanism is configured to execute the following operations that when the multifunctional mechanical arm works, a mechanical arm joint pose data sequence is obtained, decomposition processing is conducted on the mechanical arm joint pose data sequence, a joint average pose data sequence and a joint vibration data sequence are obtained, joint vibration characterization clustering analysis is conducted on a joint vibration dimension set, and joint vibration characterization clustering analysis is conducted on the joint vibration dimension set. Determining a joint vibration time sequence and a joint vibration frequency sequence according to the joint vibration point position, performing time alignment on the joint vibration time sequence and the joint vibration frequency sequence to obtain a mechanical arm joint vibration characteristic value, and correcting the mechanical arm joint vibration proportional gain by taking the mechanical arm joint vibration characteristic value as a feedback signal to obtain a mechanical arm joint vibration proportional gain. And feedback control of the multifunctional mechanical arm can be completed by correcting the proportional gain of the joint vibration of the multifunctional mechanical arm. According to the robot mechanism, through a built-in advanced control algorithm, the control precision and stability of welding operation can be remarkably improved when unstable control parameters exist.

Description

Technical Field

[0001] This application relates to the field of robotic arm technology, and more specifically, to a high-rigidity electrically driven welding robot mechanism based on a parallel configuration. Background Technology

[0002] Parallel configuration robots, as an advanced robotic mechanism characterized by high rigidity, high precision, and high load capacity, have been widely used in the industrial welding field. Compared with traditional serial robots, parallel mechanisms connect the moving and stationary platforms simultaneously through multiple branches, forming a closed kinematic chain structure, thereby significantly improving the overall rigidity and motion accuracy of the mechanism.

[0003] In high-precision welding applications, the motion stability of the robot's end effector directly determines the welding quality. However, in actual operation, due to factors such as the response characteristics of the electrical drive system, the clearance of transmission components, load variations, and external disturbances, parallel welding robots still experience minor joint-level vibrations. Although these vibrations are small in amplitude, they can lead to quality problems such as welding torch trajectory deviation and uneven welds in high-precision welding tasks, severely affecting the welding yield.

[0004] Currently, traditional robot control methods mostly employ fixed PID parameters, which are insufficient to effectively suppress complex time-varying and nonlinear joint chatter. While some advanced control algorithms can theoretically improve control performance, they are often computationally complex, have poor real-time performance, and are difficult to apply in industrial settings. Therefore, there is an urgent need to develop a novel robot mechanism capable of identifying and compensating for joint chatter in real time, while maintaining the high rigidity advantages of parallel mechanisms and further enhancing its control accuracy and stability in welding operations.

[0005] Especially in long-span, long-cycle welding tasks, control parameter drift caused by factors such as mechanical wear and temperature changes can further exacerbate joint tremor. Existing robot mechanisms lack effective online adaptive mechanisms and cannot adjust control strategies according to real-time working conditions, resulting in a gradual deterioration of welding quality as the operation time increases. Summary of the Invention

[0006] This application provides a high-rigidity electrically driven welding robot mechanism based on a parallel configuration to solve the technical problem of decreased control performance of the robotic arm when unstable control parameters exist during operation.

[0007] To solve the above-mentioned technical problems, this application adopts the following technical solution: In a first aspect, this application provides a high-rigidity electrically driven welding robot mechanism based on a parallel configuration, comprising the following steps: When the multi-functional robotic arm is working, the joint pose data sequence of the robotic arm is acquired, and the joint pose data sequence of the robotic arm is decomposed to obtain the average pose data sequence of the joints and the joint vibration data sequence. The joint tremor data sequence is interpolated to obtain a reconstructed tremor data sequence. The average joint pose data sequence is then embedded into the reconstructed tremor data sequence to obtain a joint tremor dimension set. Perform joint tremor characterization cluster analysis on the joint tremor dimension set to obtain tremor identification indicators, determine the joint tremor offset based on the tremor identification indicators, and project the joint tremor offset onto the robotic arm joint pose data sequence to obtain the joint tremor point location. Based on the joint tremor points, determine the joint tremor time sequence and the joint tremor frequency sequence, and align the joint tremor time sequence and the joint tremor frequency sequence in time to obtain the joint tremor characteristic values ​​of the robotic arm; The joint tremor characteristic value of the robotic arm is used as a feedback signal to correct the proportional gain of the joint tremor of the robotic arm, thereby realizing feedback control of the multifunctional robotic arm.

[0008] In some embodiments, decomposing the robotic arm joint pose data sequence to obtain the average joint pose data sequence and the joint vibration data sequence specifically includes: The joint pose data sequence of the robotic arm is uniformly processed to obtain the average joint pose data sequence. The joint average pose data sequence is filtered based on the robotic arm joint pose data sequence to obtain the joint tremor data sequence.

[0009] In some embodiments, interpolating the joint tremor data sequence to obtain a reconstructed tremor data sequence specifically includes: Anomalies are removed from the joint tremor data sequence by pre-setting joint tremor intervals to obtain a joint tremor anomaly-free data sequence; The location of missing data is determined based on the joint tremor de-identified data sequence; Interpolation is performed on the missing data locations to obtain the reconstructed jitter data sequence.

[0010] In some embodiments, embedding the average joint pose data sequence into the reconstructed flutter data sequence to obtain a joint flutter dimension set specifically includes: Obtain the start time point of the average pose data sequence of the joint; Obtain the start time point of the reconstructed jitter data sequence; The joint average pose data sequence and the reconstructed tremor data sequence are aligned according to the starting time point to obtain the joint pose aligned data sequence; The joint pose alignment data sequence is subjected to difference extraction to obtain the joint tremor dimension set.

[0011] In some embodiments, performing joint tremor characterization cluster analysis on the joint tremor dimension set to obtain tremor identification indicators specifically includes: The joint tremor dimension data representing the clustering follows the clustering distribution function; Obtain the critical value that determines whether the joint tremor dimension data in the characterization cluster follows the clustering distribution function; The parameter vector characterizing the clustering distribution function is determined based on the clustering distribution function; The determination interval of the clustering distribution function representing the clustering is preset for the joint tremor dimension data; The tremor identification indicators are determined based on the clustering distribution function, the critical value of the clustering distribution function, the parameter vector characterizing the clustering distribution function, and the decision interval of the clustering distribution function. The tremor identification indicators are determined by the following formula: in, This indicates the first dimension of the joint fretting focus. Joint tremor data Flutter identification indicators determined after characterization cluster analysis This represents the set of dimensions for joint flutter, which includes Data on joint tremors , The first dimension of joint fretting is represented by the first dimension. Clustering distribution function of joint tremor data, This represents the data of the joint flutter dimension. Follows the cluster distribution function , This indicates a cluster of joint flutter dimension data. The derivative of the clustering distribution function This represents the probability of joint tremor data that does not conform to the clustering distribution function. This indicates that the data in the joint tremor dimension accepts clustering and follows a clustering distribution function. , This indicates whether the data in the joint tremor dimension follows a clustering distribution function. The critical value, This indicates that the joint tremor dimension data does not accept clustering and does not follow a clustering distribution function. , This represents the parameter vector of the clustering distribution function.

[0012] In some embodiments, projecting the joint scintillation offset onto the robotic arm joint pose data sequence to obtain the joint scintillation point location specifically includes: Determine the time interval between the joint tremor offset and the robotic arm joint pose data sequence; Time alignment is performed on the joint tremor offset and the robotic arm joint pose data sequence; Extract all vibration offset values ​​of the joint vibration offset over a predetermined time. All the jitter offset values ​​extracted at the predetermined time are used as the joint jitter points at the corresponding time in the joint pose data sequence of the robotic arm.

[0013] In some embodiments, using the joint fretting characteristic value of the robotic arm as a feedback signal to correct the proportional gain of the joint fretting of the robotic arm, thereby realizing feedback control of the multifunctional robotic arm, specifically includes: Determine the characteristic values ​​of joint tremor in a standard robotic arm; When the joint fretting characteristic value of the robotic arm is greater than the standard joint fretting characteristic value of the robotic arm, increase the proportional gain; When the joint fretting characteristic value of the robotic arm is less than the standard joint fretting characteristic value of the robotic arm, reduce the proportional gain; Feedback control of the multi-functional robotic arm is achieved by increasing and decreasing the proportional gain.

[0014] Secondly, this application provides a feedback control device for a multifunctional robotic arm, which includes a robotic arm joint posture control unit, the robotic arm joint posture control unit comprising: The robotic arm joint pose data determination module is used to acquire the robotic arm joint pose data sequence when the multifunctional robotic arm is working, and to decompose the robotic arm joint pose data sequence to obtain the average joint pose data sequence and the joint vibration data sequence. The joint tremor dimension set determination module is used to perform interpolation processing on the joint tremor data sequence to obtain a reconstructed tremor data sequence, and to embed the average joint pose data sequence into the reconstructed tremor data sequence to obtain a joint tremor dimension set. The joint tremor point determination module is used to perform joint tremor characterization cluster analysis on the joint tremor dimension set to obtain tremor identification indicators, determine the joint tremor offset based on the tremor identification indicators, and project the joint tremor offset onto the robotic arm joint pose data sequence to obtain the joint tremor point. The robotic arm joint tremor feature value determination module is used to determine the joint tremor time sequence and the joint tremor frequency sequence based on the joint tremor point, and to time-align the joint tremor time sequence and the joint tremor frequency sequence to obtain the robotic arm joint tremor feature value. The feedback correction module is used to correct the proportional gain of the joint tremor of the robotic arm by using the joint tremor characteristic value of the robotic arm as a feedback signal, so as to realize the feedback control of the multifunctional robotic arm.

[0015] Thirdly, this application provides a computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the high-rigidity electrically driven welding robot mechanism based on parallel configuration described above.

[0016] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned high-rigidity electrically driven welding robot mechanism based on a parallel configuration.

[0017] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects: In the high-rigidity electrically driven welding robot mechanism and device based on parallel configuration provided in this application, when the multi-functional robotic arm is working, a sequence of joint pose data of the robotic arm is acquired. The sequence of joint pose data of the robotic arm is decomposed to obtain a joint average pose data sequence and a joint vibration data sequence. The joint vibration data sequence is interpolated to obtain a reconstructed vibration data sequence. The joint average pose data sequence is embedded into the reconstructed vibration data sequence to obtain a joint vibration dimension set. The joint vibration dimension set is then subjected to joint vibration characterization clustering. Analysis is performed to obtain tremor identification indicators. Based on the tremor identification indicators, the joint tremor offset is determined. The joint tremor offset is projected onto the joint pose data sequence of the robotic arm to obtain the joint tremor points. Based on the joint tremor points, the joint tremor time sequence and the joint tremor frequency sequence are determined. The joint tremor time sequence and the joint tremor frequency sequence are time-aligned to obtain the joint tremor feature value of the robotic arm. The joint tremor feature value of the robotic arm is used as a feedback signal to correct the joint tremor proportional gain of the robotic arm, thereby realizing feedback control of the multifunctional robotic arm.

[0018] This application reduces the complexity of subsequent data processing and analysis by identifying joint tremor points, which is beneficial for assessing the impact of tremor on the robotic arm's motion. When joint tremor occurs during robotic arm operation, the temporal nature of the tremor is analyzed, and the degree of tremor at the corresponding time point is displayed. In addition, based on the characteristic values ​​of robotic arm joint tremor, the availability and analytical capability of robotic arm joint pose data can be improved, which helps to better control the tremor behavior of the robotic arm, thereby achieving higher quality and more accurate task execution. Furthermore, the characteristic values ​​of robotic arm joint tremor are used as feedback signals to adjust the proportional gain of robotic arm joint tremor to suppress or compensate for tremor to the greatest extent, balance the response speed and stability of robotic arm operation, and improve the motion control performance of the robotic arm. Thus, the control performance of the robotic arm can be improved when there are unstable control parameters during robotic arm operation. Attached Figure Description

[0019] Figure 1 This is an exemplary flowchart of a high-rigidity electrically driven welding robot mechanism based on a parallel configuration, as shown in some embodiments of this application. Figure 2 These are schematic diagrams of exemplary hardware and / or software of a robotic arm joint pose control unit according to some embodiments of this application; Figure 3 This is a structural schematic diagram of a computer device that implements the high-rigidity electrically driven welding robot mechanism based on a parallel configuration, according to some embodiments of this application. Detailed Implementation

[0020] The core of this application is to first acquire the joint pose data sequence of the multi-functional robotic arm during operation, decompose the joint pose data sequence to obtain the average joint pose data sequence and the joint tremor data sequence, perform joint tremor characterization cluster analysis on the joint tremor dimension set, determine the joint tremor time sequence and joint tremor frequency sequence based on the joint tremor points, time-align the joint tremor time sequence and joint tremor frequency sequence to obtain the joint tremor feature value of the robotic arm, and use the joint tremor feature value as a feedback signal to correct the joint tremor proportional gain of the robotic arm, thereby realizing the feedback control of the multi-functional robotic arm. The feedback control of the multi-functional robotic arm can be completed by correcting the joint tremor proportional gain of the multi-functional robotic arm, which can improve the control performance of the robotic arm when there are unstable control parameters during operation.

[0021] To better understand the above technical solutions, a detailed description of the solutions will be provided below in conjunction with the accompanying drawings and specific implementation methods. (Reference) Figure 1 The figure is an exemplary flowchart of a high-rigidity electrically driven welding robot mechanism based on a parallel configuration, according to some embodiments of this application. The high-rigidity electrically driven welding robot mechanism 100 based on a parallel configuration mainly includes the following steps: In step 101, when the multifunctional robotic arm is working, the joint pose data sequence of the robotic arm is acquired, and the joint pose data sequence of the robotic arm is decomposed to obtain the average pose data sequence of the joints and the joint tremor data sequence.

[0022] In practice, when the multifunctional robotic arm is working, the joint pose data sequence of the robotic arm can be read from the data storage unit of the robotic arm's vision sensor.

[0023] In some embodiments, the decomposition of the robotic arm joint pose data sequence to obtain the average joint pose data sequence and the joint vibration data sequence can be achieved by the following steps: The joint pose data sequence of the robotic arm is uniformly processed to obtain the average joint pose data sequence. The joint average pose data sequence is filtered based on the robotic arm joint pose data sequence to obtain the joint tremor data sequence.

[0024] In specific implementation, the robotic arm joint pose data sequence is uniformly processed. When the robotic arm is working, each second of the movement process corresponds to one joint pose data. All the joint pose data in the complete movement process constitute the robotic arm joint pose data sequence. The average value of each movement process in all the robotic arm joint pose data sequences is calculated. The average values ​​of each movement process in all the robotic arm joint pose data sequences are arranged in chronological order to obtain the average joint pose data sequence. The average joint pose data sequence is filtered according to the robotic arm joint pose data sequence. The average pose data sequence is subtracted from the robotic arm joint pose data sequence to obtain the joint tremor data sequence.

[0025] It should be noted that the robotic arm joint pose data sequence in this application includes information such as joint rotation angle, joint extension position, and joint posture at each time point during the execution of the command. The degree of completion of the command information by the robotic arm joint varies due to the influence of various unstable control parameters. By decomposing the robotic arm joint pose data sequence, we can better understand the average state of the robotic arm joint pose and the slight deviations relative to the average pose.

[0026] In step 102, the joint tremor data sequence is interpolated to obtain a reconstructed tremor data sequence. The average pose data sequence of the joint is then embedded into the reconstructed tremor data sequence to obtain a joint tremor dimension set.

[0027] In some embodiments, interpolating the joint tremor data sequence to obtain a reconstructed tremor data sequence can be achieved using the following steps: Anomalies are removed from the joint tremor data sequence by pre-setting joint tremor intervals to obtain a joint tremor anomaly-free data sequence; The location of missing data is determined based on the joint tremor de-identified data sequence; Interpolation is performed on the missing data locations to obtain the reconstructed jitter data sequence.

[0028] In specific implementation, after decomposing the joint pose data sequence of the robotic arm, abnormal data is found in the joint tremor data sequence. Normal operation data of the robotic arm is obtained based on its normal working state. Anomalies are removed from the joint tremor data sequence by setting a preset joint tremor interval. Each data point in the joint tremor data sequence is compared with the preset joint tremor interval. If the data is outside the preset joint tremor interval, it is discarded; if it is within the preset joint tremor interval, it is retained. The joint tremor data sequence is traversed, and all retained data are used as the joint tremor anomaly-free data sequence. If there is missing data at a corresponding time point in the joint tremor anomaly-free data sequence, the position of the missing data at that time point is taken as the missing data location point. The data at the missing data location point is then interpolated. This process yields the reconstructed tremor data sequence. The interpolation of the data at the missing data location point is determined by the following formula: in, For the first Data after interpolation of missing data locations For the first The data preceding the location of the missing data point For the first The data following the location of the missing data.

[0029] In some embodiments, embedding the joint average pose data sequence into the reconstructed flutter data sequence to obtain a joint flutter dimension set can be achieved by the following steps: Obtain the start time point of the average pose data sequence of the joint; Obtain the start time point of the reconstructed jitter data sequence; The joint average pose data sequence and the reconstructed tremor data sequence are aligned according to the starting time point to obtain the joint pose aligned data sequence; The joint pose alignment data sequence is subjected to difference extraction to obtain the joint tremor dimension set.

[0030] In specific implementation, when the robotic arm starts working, the start time point of the obtained robotic arm joint pose sequence is used as the starting time point of the joint average pose data sequence and the reconstructed jitter data sequence. The joint average pose data sequence and the reconstructed jitter data sequence are aligned according to their starting time points to obtain two rows of data with the same starting time. It is ensured that the time points corresponding to the data in the joint average pose data sequence and the data in the reconstructed jitter data sequence at the same position in the two rows of data with the same starting time are consistent. The two rows of data with the same starting time are used as the joint pose aligned data sequence. The data in the reconstructed jitter data sequence and the data in the joint average pose data sequence at the same time point in the joint pose aligned data sequence are subtracted, and the absolute value is taken as the joint jitter dimension data. This process is repeated until the joint jitter dimension data corresponding to all time points in the joint pose aligned data sequence are obtained. All joint jitter dimension data are then combined into a joint jitter dimension set.

[0031] It should be noted that the joint tremor data sequence is interpolated in this application, which can better restore the integrity of the joint tremor data sequence. By embedding the average pose data sequence of the joint into the reconstructed data sequence, a joint tremor dimension set is obtained, which can more detailedly show the degree of joint tremor of the robotic arm.

[0032] In step 103, a joint tremor characterization cluster analysis is performed on the joint tremor dimension set to obtain tremor identification indicators. Based on the tremor identification indicators, the joint tremor offset is determined, and the joint tremor offset is projected onto the robotic arm joint pose data sequence to obtain the joint tremor point.

[0033] In some embodiments, performing joint tremor characterization cluster analysis on the joint tremor dimension set to obtain tremor identification indicators can be achieved through the following steps: The joint tremor dimension data representing the clustering follows the clustering distribution function; Obtain the critical value that determines whether the joint tremor dimension data in the characterization cluster follows the clustering distribution function; The parameter vector characterizing the clustering distribution function is determined based on the clustering distribution function; The determination interval of the clustering distribution function representing the clustering is preset for the joint tremor dimension data; The tremor identification indicators are determined based on the clustering distribution function, the critical value of the clustering distribution function, the parameter vector characterizing the clustering distribution function, and the decision interval of the clustering distribution function.

[0034] In practical implementation, joint flutter data in the joint flutter dimension set represent different joint flutter characteristics. Based on these different characteristics, joint flutter data representing various characteristics are selected as the representations of joint flutter, such as... All joint flutter data within the interval belong to the first type of characteristic. All joint tremor data within the interval represent the second type of representation. This process is repeated to obtain joint tremor data representing each representation. Then, based on the properties of the distribution function, a clustering distribution function is obtained. Maximum likelihood estimation is performed on the joint tremor dimension data within the clustering distribution function to obtain the critical values ​​of the joint tremor dimension data that conform to the clustering distribution functions of each representation. The parameter vector of the clustering distribution function is determined based on the clustering distribution function. Based on the representation of the joint tremor dimension data within the clustering distribution function, a judgment interval for accepting the clustering distribution function of the representation cluster is preset, thereby determining the tremor identification indicators. The tremor identification indicators are determined by the following formula: in, This indicates the first dimension of the joint fretting focus. Joint tremor data Flutter identification indicators determined after characterization cluster analysis This represents the set of dimensions for joint flutter, which includes Data on joint tremors , The first dimension of joint fretting is represented by the first dimension. Clustering distribution function of joint tremor data, This represents the data of the joint flutter dimension. Follows the cluster distribution function , This indicates a cluster of joint flutter dimension data. The derivative of the clustering distribution function This represents the probability of joint tremor data that does not conform to the clustering distribution function. This indicates that the data in the joint tremor dimension accepts clustering and follows a clustering distribution function. , This indicates whether the data in the joint tremor dimension follows a clustering distribution function. The critical value, This indicates that the joint tremor dimension data does not accept clustering and does not follow a clustering distribution function. , This represents the parameter vector of the clustering distribution function.

[0035] It should be noted that the clustering distribution function parameter vector is used to describe the distribution of different categories in the dataset. After performing joint tremor characterization analysis on the joint tremor dimension set, the various characterizations can be clustered, which is beneficial for subsequent targeted analysis of joint tremor. After characterization clustering analysis on the joint tremor dimension set, each clustered joint tremor characterization has a specific distribution interval. The average value of the joint tremor data in each specific distribution interval is calculated to obtain the average value of the joint tremor characterization in each interval. The average value of the joint tremor characterization in each interval is used as the tremor identification indicator for this interval.

[0036] In specific implementation, the difference between the vibration identification indicators and the average pose data of the corresponding interval in the average position data sequence of the robotic arm joint is calculated, and the result of the difference is used as the joint vibration offset. The joint vibration offset represents the deviation from the average pose during joint vibration. The joint vibration offset better reflects the smoothness of the robotic arm joint movement and the degree of deviation from the expected trajectory.

[0037] In some embodiments, projecting the joint scintillation offset onto the robotic arm joint pose data sequence to obtain the joint scintillation point can be achieved using the following steps: Determine the time interval between the joint tremor offset and the robotic arm joint pose data sequence; Time alignment is performed on the joint tremor offset and the robotic arm joint pose data sequence; Extract all vibration offset values ​​of the joint vibration offset over a predetermined time. All the jitter offset values ​​extracted at the predetermined time are used as the joint jitter points at the corresponding time in the joint pose data sequence of the robotic arm.

[0038] In practice, the start time when the robotic arm starts working is used as the start time of the joint tremor offset and the robotic arm joint pose data sequence. The joint tremor offset and the robotic arm joint pose data sequence are arranged with the same start time and the data points are arranged in 0.5-second intervals to complete the time alignment. The midpoint of the time interval is taken as the predetermined time, and the joint tremor offset value corresponding to the predetermined time is selected as the joint tremor point at the same time point in the robotic arm joint pose data sequence.

[0039] It should be noted that identifying the joint tremor points and aligning the data at the start time can reduce the complexity of subsequent data processing and analysis, which is beneficial for assessing the impact of tremor on the robotic arm's movement, analyzing the temporal nature of tremor when joint tremor occurs during robotic arm operation, and showing the degree of tremor at the corresponding time point.

[0040] In step 104, the joint tremor time sequence and joint tremor frequency sequence are determined based on the joint tremor points, and the joint tremor time sequence and the joint tremor frequency sequence are time-aligned to obtain the joint tremor characteristic values ​​of the robotic arm.

[0041] In practice, the joint tremor points are decomposed into time-frequency components. Each joint tremor point corresponds to a time point in the robotic arm's joint pose sequence, representing the time when the joint tremor occurs. A 2-second time interval is used to divide the robotic arm's working time into non-overlapping time windows. For each time window, the number of joint tremor points occurring within that window is counted, which serves as the tremor frequency for that window. The window is moved, and the tremor frequency is counted step by step until the entire time range is covered. At the center time point of each window, the joint tremor situation within that time period is recorded, forming a joint tremor time sequence. At the center time point of each time window, the degree of joint tremor within that time period is recorded, i.e., the absolute value of the difference between the joint tremor offset and the average pose of the robotic arm joints at that time point is used as the joint tremor frequency, forming a joint tremor frequency sequence.

[0042] In some embodiments, the joint tremor time series and joint tremor frequency series are time-aligned to obtain the joint tremor feature values ​​of the robotic arm, which can be achieved by the following steps: Obtain the first joint fremitus dimension set Joint tremor data ; Obtain the average pose data of the robotic arm joints within all time windows in the average pose data sequence. ; Obtain the average pose data of the robotic arm joints within the time window. ; Determine the first time window Joint tremor offset ; Determine the first dimension of joint fremitus set Vibration frequency corresponding to each joint vibration data point , ; According to the joint tremor dimension set Joint tremor data Average pose data of robotic arm joints within all time windows Average pose data of robotic arm joints within the time window Within the time window Joint tremor offset Joint tremor dimension focus Vibration frequency corresponding to each joint vibration data point The characteristic values ​​of joint chatter in the robotic arm are determined by the following formula: in, This represents the characteristic value of joint tremor in the robotic arm. This represents a quantification factor for the frequency of joint tremor. This represents the difference between the average pose of the robotic arm joints and the joint jitter offset across all time windows. For the average pose data of the robotic arm joints and the first time window The difference in joint tremor offset This represents the sum of vibration frequencies across the joint vibration dimension set. The first joint in the joint tremor dimension set is the sum of tremor frequencies in the joint tremor dimension set. The deflection rate of joint tremor data.

[0043] It should be noted that the joint tremor frequency quantification factor in this application It represents the frequency of joint vibration of the robotic arm per unit time, within the time window. Joint tremor offset This is achieved by iterating through the joint fretting offsets, dividing the joint fretting offsets into time windows, and then aligning the joint fretting offset values ​​within each time window with the average pose data sequence of the robotic arm joints at the same start time. The time windows are then divided into 2-second windows, and the absolute value of the difference between the joint fretting offset data and the average pose data of the robotic arm joints at the center time point of each time window is used as the joint fretting frequency. This is then used to determine the 1st joint fretting frequency in the joint fretting dimension set. Vibration frequency corresponding to each joint vibration data point Among them, the center time point of the time window corresponding to the tremor frequency is indivual.

[0044] It should be noted that each data point in the joint tremor time series represents the frequency of joint tremor within the corresponding time window, which can more effectively monitor the degree of tremor in the robotic arm. By aligning the joint tremor sequence with time and generating joint tremor feature values ​​for the robotic arm, the availability and analytical capabilities of the robotic arm joint pose data can be improved, which helps to better control the tremor behavior of the robotic arm, thereby achieving higher quality and more accurate task execution.

[0045] In step 105, the joint tremor characteristic value of the robotic arm is used as a feedback signal to correct the proportional gain of the joint tremor of the robotic arm, thereby realizing feedback control of the multifunctional robotic arm.

[0046] In some embodiments, the feedback control of the multifunctional robotic arm is achieved by using the joint fretting characteristic value of the robotic arm as a feedback signal to correct the proportional gain of the joint fretting of the robotic arm, which can be accomplished by the following steps: Determine the characteristic values ​​of joint tremor in a standard robotic arm; When the joint fretting characteristic value of the robotic arm is greater than the standard joint fretting characteristic value of the robotic arm, increase the proportional gain; When the joint fretting characteristic value of the robotic arm is less than the standard joint fretting characteristic value of the robotic arm, reduce the proportional gain; Feedback control of the multi-functional robotic arm is achieved by increasing or decreasing the proportional gain.

[0047] In practice, the standard robotic arm joint chafing characteristic value is obtained through simulation experiments. The standard robotic arm joint chafing characteristic value is used to represent the allowable value of the robotic arm joint chafing frequency. When the robotic arm joint chafing characteristic value is greater than the standard robotic arm joint chafing characteristic value, the proportional gain is increased by manually changing the controller settings to enhance the suppression effect of feedback control on the robotic arm joint chafing. When the robotic arm joint chafing characteristic value is less than the standard robotic arm joint chafing characteristic value, the proportional gain is decreased by manually changing the controller settings to avoid over-response and instability of control parameters.

[0048] It should be noted that the comparison between the standard joint chafing characteristic value and the robotic arm joint chafing characteristic value in this application, and the subsequent adjustment of the proportional gain, is for the purpose of better controlling and optimizing the performance and working behavior of the robotic arm. Increasing the proportional gain will lead to a faster response speed and stronger control force, thereby compensating for chafing more quickly. Decreasing the proportional gain will slow down the response speed of the controller, which helps to reduce oscillation or instability of control parameters. Using the robotic arm joint chafing characteristic value as a feedback signal to adjust the proportional gain of the robotic arm joint chafing is to suppress or compensate for chafing to the greatest extent, balance the response speed and stability of the robotic arm operation, and thus improve the motion control performance of the robotic arm.

[0049] In another aspect, in some embodiments, this application provides a feedback control device for a multifunctional robotic arm, the device including a robotic arm joint pose control unit, referenced... Figure 2 This figure is a schematic diagram of exemplary hardware and / or software of a robotic arm joint pose control unit according to some embodiments of this application. The robotic arm joint pose control unit 200 includes: a robotic arm joint pose data acquisition module 201, a joint tremor dimension set determination module 202, a joint tremor point determination module 203, a robotic arm joint tremor feature value determination module 204, and a feedback correction module 205, which are described below: The robotic arm joint pose data acquisition module 201 in this application is mainly used to acquire the robotic arm joint pose sequence data when the multifunctional robotic arm is working, and to decompose the robotic arm joint pose sequence data to obtain the average joint pose sequence data and joint vibration sequence data. The joint tremor dimension set determination module 202 in this application is mainly used to interpolate the joint tremor sequence data to obtain reconstructed tremor sequence data, and to embed the average pose sequence data of the joint into the reconstructed tremor sequence data to obtain the joint tremor dimension set. The joint tremor point determination module 203 in this application is mainly used to perform joint tremor characterization cluster analysis on the joint tremor dimension set to obtain tremor identification indicators, determine the joint tremor offset based on the tremor identification indicators, and project the joint tremor offset onto the robotic arm joint pose sequence data to obtain the joint tremor point. The robotic arm joint tremor feature value determination module 204 is mainly used to determine the joint tremor time sequence and the robotic arm joint tremor frequency sequence based on the joint tremor point, and to time-align the joint tremor time sequence and the joint tremor frequency sequence to obtain the robotic arm joint tremor feature value. The feedback correction module 205 in this application is mainly used to correct the proportional gain of the joint tremor of the robotic arm by using the joint tremor characteristic value of the robotic arm as a feedback signal, so as to realize the feedback control of the multifunctional robotic arm.

[0050] In addition, this application also provides a computer device, the computer device including a memory and a processor, the memory storing code, the processor being configured to acquire the code and execute the high-rigidity electrically driven welding robot mechanism based on parallel configuration described above.

[0051] In some embodiments, reference Figure 3 This figure is a structural schematic diagram of a computer device implementing the high-rigidity electrically driven welding robot mechanism based on a parallel configuration, according to some embodiments of this application. The high-rigidity electrically driven welding robot mechanism based on a parallel configuration described in the above embodiments can be... Figure 3 The computer device shown is used to implement this, and the computer device includes at least one processor 301, a communication bus 302, a memory 303, and at least one communication interface 304.

[0052] The processor 301 may be a general-purpose central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more for controlling the execution of the high-rigidity electrically driven welding robot mechanism based on parallel configuration described in this application.

[0053] The communication bus 302 may include a path for transmitting information between the aforementioned components.

[0054] The memory 303 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or it may be an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital versatile optical discs, Blu-ray discs, etc.), a magnetic disk or other magnetic storage device, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto. The memory 303 may exist independently and be connected to the processor 301 via a communication bus 302. The memory 303 may also be integrated with the processor 301.

[0055] The memory 303 stores program code for executing the scheme of this application, and its execution is controlled by the processor 301. The processor 301 executes the program code stored in the memory 303. The program code may include one or more software modules. In the above embodiments, the joint pose sequence data of the robotic arm can be implemented by the processor 301 and one or more software modules in the program code in the memory 303.

[0056] Communication interface 304 uses any transceiver-like device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0057] In a specific implementation, as one example, a computer device may include multiple processors, each of which may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores used to process data (e.g., computer program instructions).

[0058] The aforementioned computer device can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer device can be a desktop computer, a portable computer, a network server, a handheld digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer device.

[0059] In addition, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned high-rigidity electrically driven welding robot mechanism based on a parallel configuration.

[0060] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0061] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of the invention. Therefore, if these modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include these modifications and variations.

Claims

1. A high-rigidity electrically driven welding robot mechanism based on a parallel configuration, characterized in that, The controller is configured to perform the following steps: When the multi-functional robotic arm is working, the joint pose data sequence of the robotic arm is acquired, and the joint pose data sequence of the robotic arm is decomposed to obtain the average pose data sequence of the joints and the joint vibration data sequence. The joint tremor data sequence is interpolated to obtain a reconstructed tremor data sequence. The average joint pose data sequence is then embedded into the reconstructed tremor data sequence to obtain a joint tremor dimension set. Perform joint tremor characterization cluster analysis on the joint tremor dimension set to obtain tremor identification indicators, determine the joint tremor offset based on the tremor identification indicators, and project the joint tremor offset onto the robotic arm joint pose data sequence to obtain the joint tremor point location. Based on the joint tremor points, determine the joint tremor time sequence and the joint tremor frequency sequence, and align the joint tremor time sequence and the joint tremor frequency sequence in time to obtain the joint tremor characteristic values ​​of the robotic arm; The joint tremor characteristic value of the robotic arm is used as a feedback signal to correct the proportional gain of the joint tremor of the robotic arm, thereby realizing feedback control of the multifunctional robotic arm.

2. The high-rigidity electrically driven welding robot mechanism based on a parallel configuration as described in claim 1, characterized in that, The decomposition of the robotic arm joint pose data sequence to obtain the average joint pose data sequence and joint tremor data sequence specifically includes: The joint pose data sequence of the robotic arm is uniformly processed to obtain the average joint pose data sequence. The joint average pose data sequence is filtered based on the robotic arm joint pose data sequence to obtain the joint tremor data sequence.

3. The high-rigidity electrically driven welding robot mechanism based on a parallel configuration as described in claim 1, characterized in that, The process of interpolating the joint tremor data sequence to obtain the reconstructed tremor data sequence specifically includes: Anomalies are removed from the joint tremor data sequence by pre-setting joint tremor intervals to obtain a joint tremor anomaly-free data sequence; The location of missing data is determined based on the joint tremor de-identified data sequence; Interpolation is performed on the missing data locations to obtain the reconstructed jitter data sequence.

4. The high-rigidity electrically driven welding robot mechanism based on a parallel configuration as described in claim 1, characterized in that, Embedding the average joint pose data sequence into the reconstructed flutter data sequence to obtain the joint flutter dimension set specifically includes: Obtain the start time point of the average pose data sequence of the joint; Obtain the start time point of the reconstructed jitter data sequence; The joint average pose data sequence and the reconstructed tremor data sequence are aligned according to the starting time point to obtain the joint pose aligned data sequence; The joint pose alignment data sequence is subjected to difference extraction to obtain the joint tremor dimension set.

5. The high-rigidity electrically driven welding robot mechanism based on a parallel configuration as described in claim 1, characterized in that, Performing joint tremor characterization cluster analysis on the aforementioned joint tremor dimension set yields tremor identification indicators, specifically including: The joint tremor dimension data representing the clustering follows the clustering distribution function; Obtain the critical value that determines whether the joint tremor dimension data in the characterization cluster follows the clustering distribution function; The parameter vector characterizing the clustering distribution function is determined based on the clustering distribution function; The determination interval of the clustering distribution function representing the clustering is preset for the joint tremor dimension data; The tremor identification indicators are determined based on the clustering distribution function, the critical value of the clustering distribution function, the parameter vector characterizing the clustering distribution function, and the decision interval of the clustering distribution function. The tremor identification indicators are determined by the following formula: in, This indicates the first dimension of the joint fretting focus. Joint tremor data Flutter identification indicators determined after characterization cluster analysis This represents the set of dimensions for joint flutter, which includes Data on joint tremors , The first dimension of joint fretting is represented by the first dimension. Clustering distribution function of joint tremor data, This represents the data of the joint flutter dimension. Follows the cluster distribution function , This indicates a cluster of joint flutter dimension data. The derivative of the clustering distribution function This represents the probability of joint tremor data that does not conform to the clustering distribution function. This indicates that the data in the joint tremor dimension accepts clustering and follows a clustering distribution function. , This indicates whether the data in the joint tremor dimension follows a clustering distribution function. The critical value, This indicates that the joint tremor dimension data does not accept clustering and does not follow a clustering distribution function. , This represents the parameter vector of the clustering distribution function.

6. The high-rigidity electrically driven welding robot mechanism based on a parallel configuration as described in claim 1, characterized in that, Projecting the joint scintillation offset onto the robotic arm joint pose data sequence to obtain the joint scintillation points specifically includes: Determine the time interval between the joint tremor offset and the robotic arm joint pose data sequence; Time alignment is performed on the joint tremor offset and the robotic arm joint pose data sequence; Extract all vibration offset values ​​of the joint vibration offset over a predetermined time. All the jitter offset values ​​extracted at the predetermined time are used as the joint jitter points at the corresponding time in the joint pose data sequence of the robotic arm.

7. The high-rigidity electrically driven welding robot mechanism based on a parallel configuration as described in claim 1, characterized in that, Using the joint fretting characteristic value of the robotic arm as a feedback signal to correct the proportional gain of the joint fretting of the robotic arm, the feedback control of the multifunctional robotic arm specifically includes: Determine the characteristic values ​​of joint tremor in a standard robotic arm; When the joint fretting characteristic value of the robotic arm is greater than the standard joint fretting characteristic value of the robotic arm, increase the proportional gain; When the joint fretting characteristic value of the robotic arm is less than the standard joint fretting characteristic value of the robotic arm, reduce the proportional gain; Feedback control of the multi-functional robotic arm is achieved by increasing or decreasing the proportional gain.

8. A feedback control device for a multifunctional robotic arm, characterized in that, The device is integrated into the controller of the high-rigidity electrically driven welding robot mechanism based on a parallel configuration as described in any one of claims 1 to 7. The device includes a robotic arm joint posture control unit, which includes: The robotic arm joint pose data determination module is used to acquire the robotic arm joint pose data sequence when the multifunctional robotic arm is working, and to decompose the robotic arm joint pose data sequence to obtain the average joint pose data sequence and the joint vibration data sequence. The joint tremor dimension set determination module is used to perform interpolation processing on the joint tremor data sequence to obtain a reconstructed tremor data sequence, and to embed the average joint pose data sequence into the reconstructed tremor data sequence to obtain a joint tremor dimension set. The joint tremor point determination module is used to perform joint tremor characterization cluster analysis on the joint tremor dimension set to obtain tremor identification indicators, determine the joint tremor offset based on the tremor identification indicators, and project the joint tremor offset onto the robotic arm joint pose data sequence to obtain the joint tremor point. The robotic arm joint tremor feature value determination module is used to determine the joint tremor time sequence and the joint tremor frequency sequence based on the joint tremor point, and to time-align the joint tremor time sequence and the joint tremor frequency sequence to obtain the robotic arm joint tremor feature value. The feedback correction module is used to correct the proportional gain of the joint tremor of the robotic arm by using the joint tremor characteristic value of the robotic arm as a feedback signal, so as to realize the feedback control of the multifunctional robotic arm.

9. A computer device, characterized in that, The computer device includes a memory and a processor, the memory storing code, and the processor being configured to retrieve the code and execute the high-rigidity electrically driven welding robot mechanism based on a parallel configuration as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the high-rigidity electrically driven welding robot mechanism based on a parallel configuration as described in any one of claims 1 to 7.

Images