A method and system for joint testing of backlash stiffness in robot joint transmission

By converting the input angular displacement into the theoretical output angular displacement, identifying the reversal interval and establishing an elastic reference relationship, the problem of accuracy in testing transmission backlash and stiffness during robot joint reversal is solved, and efficient evaluation of transmission performance is achieved.

CN123111684APending Publication Date: 2026-11-17ZHEJIANG ATTC AUTOMOBILE TECH SERVICE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611578853.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-13
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing technologies struggle to accurately distinguish between clearance transition response and elastic load response during robot joint reversal, resulting in transmission backlash and stiffness test results being affected by changes in load conditions, making it difficult to reflect transmission characteristics under stable load conditions.

Method used

By converting the input angular displacement during the robot joint reversal process into the theoretical output angular displacement, the pre-reversal bearing area, reversal transition area, and post-reversal bearing area are identified, and the elastic reference relationship before and after reversal is established. Using these relationships, the transmission error displacement is converted to a zero transmission torque state, and the transmission backlash and transmission stiffness are determined.

Benefits of technology

This technology enables the differentiation between backlash transition and elastic load response during the same reversing test, improving the accuracy and consistency of transmission backlash and stiffness test results, and enhancing the usability of robot joint transmission performance testing.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123111684A_ABST
    Figure CN123111684A_ABST
Patent Text Reader

Abstract

This invention relates to the field of transmission testing technology, specifically disclosing a method and system for joint testing of robot joint transmission backlash stiffness. The method includes: converting the input end angular displacement during the robot joint reversal process into the theoretical output end angular displacement, and comparing it with the actual output end angular displacement to obtain the transmission error displacement, while simultaneously acquiring the output end transmitted torque; based on the continuous change state of the output end transmitted torque and the transmission error displacement, identifying the pre-reversal load-bearing area, the reversal transition area, and the post-reversal load-bearing area, determining the corresponding load-bearing boundaries, extracting the elastic response interval, and establishing the pre-reversal elastic reference relationship and the post-reversal elastic reference relationship; using the elastic reference relationship to convert the transmission error displacement of the two load-bearing boundaries to a zero transmitted torque state, determining the transmission backlash based on the converted reference boundary displacement, and obtaining the transmission stiffness; this invention can improve the accuracy and consistency of the joint testing results of robot joint transmission backlash stiffness.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of transmission testing technology, and in particular to a method and system for joint backlash stiffness testing of robot joint transmissions. Background Technology

[0002] Robot joints typically transmit motion from the drive end to the output end through reducers, gear pairs, or other transmission mechanisms. Backlash and stiffness directly affect the motion transmission accuracy and load response characteristics during joint reversal. Existing testing methods usually determine backlash and stiffness based on the displacement difference or torque-displacement relationship between the input and output ends. However, during reversal, robot joints simultaneously experience the release of the original load state, the transformation of the transmission contact relationship, and the establishment of load in the opposite direction after reversal. The transmission error displacement includes both gap displacement and elastic deformation displacement, making it difficult to accurately determine the reversal boundary position. The obtained backlash results are easily affected by changes in the load state.

[0003] In existing testing processes, the transmitted torque and elastic deformation levels corresponding to the pre-commutation and post-commutation load states are usually inconsistent. If the transmission backlash is directly determined using the displacement difference before and after commutation, the elastic deformation at the boundary positions will be superimposed on the backlash result. Simultaneously, when extracting the stiffness response within the load transition region, the commutation transition data is easily mixed with the elastic response data, making it difficult for the stiffness test results to accurately reflect the transmission characteristics under stable load conditions. Therefore, how to distinguish between the backlash transition response and the elastic load response during the same commutation test process, and eliminate the influence of load boundary force differences on the backlash and stiffness test results, has become an urgent problem to be solved. Summary of the Invention

[0004] This invention provides a method and system for joint testing of backlash stiffness in robot joint transmission, in order to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a method for joint testing of backlash stiffness in robot joint transmissions, comprising: S1. Convert the input end angular displacement during the robot joint reversal process into the theoretical output end angular displacement, compare it with the actual output end angular displacement, obtain the transmission error displacement, and obtain the output end transmitted torque. S2. Based on the continuous change of the output torque and transmission error displacement during the commutation process, identify the pre-commutation load area, commutation transition area, and post-commutation load area. S3. Determine the first load-bearing boundary at the position where the pre-commutation load-bearing area transitions to the commutation transition area, and determine the second load-bearing boundary at the position where the commutation transition area transitions to the post-commutation load-bearing area. S4. Extract the elastic response intervals before and after the commutation from the pre-commutation load area and the post-commutation load area respectively, and establish the pre-commutation elastic reference relationship and the post-commutation elastic reference relationship based on the pre-commutation elastic response interval and the post-commutation elastic response interval. S5. Based on the elastic reference relationship before and after the switching, the transmission error displacement of the first bearing boundary and the second bearing boundary is converted to the zero transmission torque state to obtain the displacement of the first reference boundary and the displacement of the second reference boundary. S6. Based on the first and second reference boundary displacements, determine the transmission backlash of the robot joint, and based on the elastic reference relationship before and after the reversal, determine the transmission stiffness before and after the reversal, and generate the robot joint transmission test results.

[0006] In a preferred embodiment, the step of converting the input angular displacement during the robot joint reversal process into a theoretical output angular displacement, comparing it with the actual output angular displacement to obtain the transmission error displacement, and acquiring the output transmitted torque includes: During the process of the robot joint switching from the forward motion direction to the opposite backward motion direction, the input end angular displacement, the actual output end angular displacement, and the output end transmitted torque are obtained according to the unified sampling timing sequence. The theoretical angular displacement of the output end is determined based on the transmission ratio between the input and output ends of the robot joint and the angular displacement of the input end at each sampling time. The transmission error displacement is obtained by processing the difference between the theoretical angular displacement and the actual angular displacement of the output end at the same sampling time. According to the sampling sequence, the transmission error displacement and the output torque are recorded together.

[0007] In a preferred embodiment, identifying the pre-commutation load-bearing zone, the commutation transition zone, and the post-commutation load-bearing zone based on the continuous change in the output torque and transmission error displacement during the commutation process includes: Set continuous sampling windows along the sampling time sequence to obtain the output torque and transmission error displacement within each continuous sampling window; Using the transmission error displacement as the independent variable and the output torque as the dependent variable, the output torque and transmission error displacement within each continuous sampling window are linearly fitted to obtain the response slope and fitting residual of the continuous sampling window. Among them, when the range of transmission error displacement within the continuous sampling window does not exceed the angular displacement detection resolution, the continuous sampling window is not used for load state determination. The allowable deviation of the response slope for each consecutive sampling window is determined based on the fitting residual, and the difference in response slope between adjacent consecutive sampling windows is compared with the corresponding allowable deviation of the response slope. Based on the output torque sign that exceeds the output torque detection resolution and appears continuously along the sampling time sequence in the continuous sampling data before commutation, the direction of the transmission torque before commutation is determined, and the transmission direction opposite to the direction of the transmission torque before commutation is determined as the direction of the transmission torque after commutation; from the sampling data of the output torque along the direction of the transmission torque before commutation, the sampling segment that appears continuously along the sampling time sequence and whose response slope difference between adjacent continuous sampling windows is within the allowable deviation of the corresponding response slope is extracted to obtain the bearing area before commutation; From the sampled data of the torque transmitted from the output end along the direction of the torque transmitted after commutation, extract the sampling segment that appears continuously along the sampling time sequence and whose response slope difference between adjacent consecutive sampling windows is within the allowable deviation of the corresponding response slope, and obtain the bearing area after commutation; The sampling section between the end position of the pre-commutation bearing area and the beginning position of the post-commutation bearing area is identified as the commutation transition area.

[0008] In a preferred embodiment, determining the first load-bearing boundary at the position where the pre-commutation load-bearing area transitions to the commutation transition area, and determining the second load-bearing boundary at the position where the commutation transition area transitions to the post-commutation load-bearing area, includes: Read the continuous sampling windows adjacent to the pre-commutation load area and the commutation transition area, and take the position where the response slope difference between the adjacent continuous sampling windows changes from being within the corresponding response slope allowable deviation to exceeding the corresponding response slope allowable deviation as the first load boundary; Read the continuous sampling windows adjacent to the commutation transition zone and the load-bearing zone after commutation, and use the position where the difference in response slope between adjacent continuous sampling windows changes from exceeding the allowable deviation of the corresponding response slope to being within the allowable deviation of the corresponding response slope as the second load-bearing boundary.

[0009] In a preferred embodiment, extracting the pre-commutation elastic response interval and the post-commutation elastic response interval from the pre-commutation load-bearing region and the post-commutation load-bearing region, respectively, includes: A continuous sampling window is set along the sampling timing of the pre-commutation and post-commutation load-bearing areas to obtain the adjacent sampling changes of the output torque and transmission error displacement within each continuous sampling window; According to the sampling order, the output torque change and transmission error displacement change within the same continuous sampling window are used as vector components to construct torque change vector and displacement change vector respectively. The direction consistency coefficient of each continuous sampling window is determined based on the cosine of the angle between the torque change vector and the displacement change vector. The limit value of the direction consistency coefficient is determined based on the detection resolution of the output torque and transmission error displacement, as well as the magnitude of the torque change vector and displacement change vector. Continuous sampling segments with a direction consistency coefficient not lower than the corresponding limit value are extracted along the sampling time sequence to obtain the elastic response interval before commutation and the elastic response interval after commutation.

[0010] In a preferred embodiment, determining the limiting value of the direction consistency coefficient based on the detection resolution of the output transmitted torque and transmission error displacement, as well as the magnitudes of the torque change vector and displacement change vector, includes: Based on the relationship between the detection resolution of the torque transmitted at the output end and the magnitude of the torque change vector, the allowable deviation of the direction corresponding to the torque change vector is determined; Based on the relationship between the detection resolution of the transmission error displacement and the magnitude of the displacement change vector, the allowable deviation of the direction corresponding to the displacement change vector is determined. The allowable included angle is determined based on the permissible deviation of the directions corresponding to the torque change vector and the displacement change vector, and the cosine value of the included angle corresponding to the allowable included angle is used as the limiting value of the direction consistency coefficient.

[0011] In a preferred embodiment, establishing the pre-commutation elastic reference relationship and the post-commutation elastic reference relationship includes: Obtain the changes in output transmitted torque and transmission error displacement at adjacent sampling times within the elastic response interval before and after commutation. Based on the ratio between the change in transmission error displacement and the change in the corresponding output torque, the unit torque displacement change in the elastic response range before and after commutation is determined respectively. By selecting the sampling time in the middle of the pre-commutation elastic response interval and the post-commutation elastic response interval along the sampling time sequence, the corresponding output end transmission torque and transmission error displacement are obtained. Based on the unit torque displacement change, the corresponding change of transmission error displacement when the output end transmission torque changes is determined, thus obtaining the pre-commutation elastic reference relationship and the post-commutation elastic reference relationship.

[0012] In a preferred embodiment, the step of converting the transmission error displacement of the first bearing boundary and the second bearing boundary to a zero transmission torque state based on the pre-commutation elastic reference relationship and the post-commutation elastic reference relationship includes: Obtain the output torque and transmission error displacement of the first and second load-bearing boundaries; Based on the pre-transformation elastic reference relationship and the output end transmitted torque and transmission error displacement of the first bearing boundary, the transmission error displacement corresponding to the zero transmitted torque state is determined, and the first reference boundary displacement is obtained. Based on the elastic reference relationship after reversal and the output end transmitted torque and transmission error displacement of the second bearing boundary, the transmission error displacement corresponding to the zero transmitted torque state is determined, and the displacement of the second reference boundary is obtained.

[0013] In a preferred embodiment, the process of determining the transmission backlash of the robot joint based on the first and second reference boundary displacements, and determining the transmission stiffness before and after the commutation based on the pre-commutation and post-commutation elastic reference relationships, and generating robot joint transmission test results, includes: Determine the displacement interval between the first reference boundary displacement and the second reference boundary displacement, and use the displacement interval as the transmission backlash of the robot joint; Based on the proportional relationship between the change in output torque and the change in transmission error displacement in the elastic reference relationship before and after commutation, the change in output torque corresponding to a unit change in transmission error displacement is determined, and the transmission stiffness before and after commutation is obtained. The transmission backlash, transmission stiffness before reversal, and transmission stiffness after reversal are correlated and recorded to obtain the test results of the robot joint transmission.

[0014] To address the above problems, the present invention also provides a joint testing system for robot joint transmission backlash stiffness, the system comprising: The data acquisition module is used to convert the input end angular displacement during the robot joint reversal process into the theoretical output end angular displacement, compare it with the actual output end angular displacement, obtain the transmission error displacement, and obtain the output end transmitted torque. The load identification module is used to identify the pre-commutation load area, commutation transition area, and post-commutation load area based on the continuous change of the output torque and transmission error displacement during the commutation process. The boundary determination module is used to determine the first bearing boundary at the position where the bearing area before the changeover transitions to the bearing area after the changeover, and to determine the second bearing boundary at the position where the bearing area after the changeover transitions to the bearing area after the changeover. The benchmark construction module is used to extract the pre-commutation elastic response interval and the post-commutation elastic response interval from the pre-commutation load area and the post-commutation load area, respectively, and to establish the pre-commutation elastic benchmark relationship and the post-commutation elastic benchmark relationship based on the pre-commutation elastic response interval and the post-commutation elastic response interval. The boundary conversion module is used to convert the transmission error displacement of the first bearing boundary and the second bearing boundary to a zero transmission torque state based on the elastic reference relationship before and after the conversion, so as to obtain the first reference boundary displacement and the second reference boundary displacement. The test output module is used to determine the transmission backlash of the robot joint based on the first reference boundary displacement and the second reference boundary displacement, and to determine the transmission stiffness before and after the commutation based on the elastic reference relationship before and after the commutation, and to generate the robot joint transmission test results.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention converts the input angular displacement into the theoretical output angular displacement and combines it with the actual output angular displacement to obtain the transmission error displacement, enabling the output torque and transmission error displacement to be analyzed in a unified commutation process. Furthermore, based on the continuous change of the two, it divides the pre-commutation load area, the commutation transition area, and the post-commutation load area, and extracts the elastic response interval from the pre-commutation load area and the post-commutation load area, so that the gap conversion process can be distinguished from the stable elastic load process, improving the accuracy of load state identification and elastic response data selection.

[0016] 2. This invention establishes elastic reference relationships based on the elastic response intervals before and after reversal, and uses the corresponding elastic reference relationships to convert the transmission error displacement of the bearing boundaries on both sides of the reversal to a unified zero transmission torque state, thereby correcting the influence of elastic deformation contained in the bearing boundaries. Based on this, the transmission backlash is determined, and the transmission stiffness is obtained using the elastic response relationship. This achieves joint characterization of backlash and stiffness in the same reversal test process, improving the consistency, accuracy, and usability of robot joint transmission performance test results. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a method for jointly testing the backlash stiffness of a robot joint transmission according to an embodiment of the present invention. Figure 2 A functional block diagram of a robot joint transmission backlash stiffness joint testing system provided in an embodiment of the present invention; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0019] This application provides a method for jointly testing the backlash and stiffness of robot joint transmission. The method is executed by an electronic device communicatively connected to a robot joint testing device. The electronic device receives, via a data acquisition interface, the input angular displacement, the actual output angular displacement, and the output transmitted torque, obtained according to a unified sampling sequence. It then executes the steps described in this application through a processor, outputting the transmission backlash, the transmission stiffness before commutation, and the transmission stiffness after commutation. The electronic device can be an industrial control computer, controller, or a terminal device with corresponding data processing capabilities, installed at the testing site.

[0020] Reference Figure 1The diagram shown is a flowchart illustrating a method for jointly testing the backlash stiffness of a robot joint transmission according to an embodiment of the present invention. In this embodiment, the method for jointly testing the backlash stiffness of a robot joint transmission includes: S1. Convert the input end angular displacement during the robot joint reversal process into the theoretical output end angular displacement, compare it with the actual output end angular displacement, obtain the transmission error displacement, and obtain the output end transmitted torque. In this embodiment of the invention, the step of converting the input end angular displacement during the robot joint reversal process into the theoretical output end angular displacement, comparing it with the actual output end angular displacement to obtain the transmission error displacement, and obtaining the output end transmitted torque includes: During the process of the robot joint switching from the forward motion direction to the opposite backward motion direction, the input end angular displacement, the actual output end angular displacement, and the output end transmitted torque are obtained according to the unified sampling timing sequence. The theoretical angular displacement of the output end is determined based on the transmission ratio between the input and output ends of the robot joint and the angular displacement of the input end at each sampling time. The transmission error displacement is obtained by processing the difference between the theoretical angular displacement and the actual angular displacement of the output end at the same sampling time. According to the sampling sequence, the transmission error displacement and the output torque are recorded together.

[0021] During the complete reversal process of a robot joint transitioning from the forward motion direction to the opposite backward motion direction, the input angular displacement detection device, the output angular displacement detection device, and the output torque detection device are connected to the same sampling clock. A single sampling trigger signal controls the three types of detection devices to read the rotational position of the robot joint's input end, the actual rotational position of the output end, and the torque state transmitted to the output end via the joint transmission mechanism at the same sampling time. The input angular displacement, actual output angular displacement, and output torque corresponding to each sampling time are saved in the order of sampling trigger time, ensuring a unique temporal correspondence among the three types of detection data generated at the same sampling time. The acquisition range begins from the robot joint maintaining the forward motion state, continuously passing through the forward load release, motion direction conversion, and post-reversal load establishment processes, until the actual output angular displacement continuously changes along the backward motion direction and the output torque remains in the post-reversal transmission state. This yields the input angular displacement, actual output angular displacement, and output torque arranged according to a unified sampling time sequence.

[0022] The transmission ratio between the input and output ends is determined based on the actual transmission mechanism used in the robot joints. In this embodiment, the ratio of the input end angular displacement to the theoretical output end angular displacement is used as the transmission ratio. The theoretical output end angular displacement is obtained by removing this transmission ratio from the input end angular displacement at each sampling time, without the influence of transmission backlash or elastic deformation. The input and output ends use the assembly-completed reference position as the common positional basis, and the positive direction of the angular displacement is pre-defined. When the actual transmission mechanism causes the input and output ends to rotate in opposite directions, the theoretical output end angular displacement is recorded according to the pre-determined directional relationship. The conversion result is recorded according to the original sampling time, ensuring that each input end angular displacement corresponds to a unique theoretical output end angular displacement.

[0023] For each sampling moment in the unified sampling sequence, the theoretical and actual angular displacements of the output terminal are read. Under the conditions that both use the same angular displacement starting position, the same definition of the rotation direction, and the same sampling time base, the signed angular displacement difference obtained by subtracting the actual angular displacement from the theoretical angular displacement is taken as the transmission error displacement; where the positive or negative sign only indicates the direction of deviation relative to the unified rotation direction. The same processing method is applied to all sampling moments to obtain the transmission error displacements arranged continuously along the unified sampling sequence.

[0024] The transmission error displacement and the synchronously acquired output torque are matched one by one according to a unified sampling sequence. The sampling time is used as the basis for data association between the two, so that the transmission error displacement and the output torque with the same sampling time are included in the same data record. The original time order of each data record in the commutation process is maintained. Data from different sampling times are not cross-matched, time-shifted, or reordered. Each data record simultaneously represents the angular displacement transmission state and torque transmission state at the corresponding sampling time. The transmission error displacement and the output torque associated according to the sampling sequence are obtained from all continuous data records. This provides a data basis with consistent time relationship for setting continuous sampling windows and identifying the pre-commutation load area, commutation transition area, and post-commutation load area based on the continuous change state of the output torque and transmission error displacement.

[0025] The beneficial effects are as follows: by uniformly sampling the input end angular displacement, actual output end angular displacement, and output end transmitted torque, the position transmission state and torque transmission state during the robot joint reversal process have a consistent time reference. The input end angular displacement at each sampling moment is converted into the corresponding theoretical output end angular displacement through the transmission ratio. The transmission error displacement is determined by the theoretical output end angular displacement and the actual output end angular displacement at the same sampling moment. This allows the angular displacement changes of the input end motion that are not actually transmitted to the output end to be continuously reflected along the reversal process. Furthermore, the transmission error displacement is associated with the synchronous output end transmitted torque at each sampling moment and recorded. This allows the subsequent load state identification to be processed based on the torque and displacement changes determined by the time correspondence, thereby improving the accuracy and continuity of the data correspondence in the reversal process.

[0026] S2. Based on the continuous change of the output torque and transmission error displacement during the commutation process, identify the pre-commutation load area, commutation transition area, and post-commutation load area. In this embodiment of the invention, identifying the pre-commutation load-bearing area, the commutation transition area, and the post-commutation load-bearing area based on the continuous change of the output torque and transmission error displacement during the commutation process includes: Set continuous sampling windows along the sampling time sequence to obtain the output torque and transmission error displacement within each continuous sampling window; Using the transmission error displacement as the independent variable and the output torque as the dependent variable, the output torque and transmission error displacement within each continuous sampling window are linearly fitted to obtain the response slope and fitting residual of the continuous sampling window. Among them, when the range of transmission error displacement within the continuous sampling window does not exceed the angular displacement detection resolution, the continuous sampling window is not used for load state determination. The allowable deviation of the response slope for each consecutive sampling window is determined based on the fitting residual, and the difference in response slope between adjacent consecutive sampling windows is compared with the corresponding allowable deviation of the response slope. Based on the output torque sign that exceeds the output torque detection resolution and appears continuously along the sampling time sequence in the continuous sampling data before commutation, the direction of the transmission torque before commutation is determined, and the transmission direction opposite to the direction of the transmission torque before commutation is determined as the direction of the transmission torque after commutation; from the sampling data of the output torque along the direction of the transmission torque before commutation, the sampling segment that appears continuously along the sampling time sequence and whose response slope difference between adjacent continuous sampling windows is within the allowable deviation of the corresponding response slope is extracted to obtain the bearing area before commutation; From the sampled data of the torque transmitted from the output end along the direction of the torque transmitted after commutation, extract the sampling segment that appears continuously along the sampling time sequence and whose response slope difference between adjacent consecutive sampling windows is within the allowable deviation of the corresponding response slope, and obtain the bearing area after commutation; The sampling section between the end position of the pre-commutation bearing area and the beginning position of the post-commutation bearing area is identified as the commutation transition area.

[0027] The transmission error displacement and output torque corresponding to adjacent sampling times are continuously extracted along the established sampling time sequence. Continuous sampling windows are formed with the same number of sampling data. The transmission error displacement and output torque in each continuous sampling window maintain a one-to-one correspondence with the original sampling time. Adjacent continuous sampling windows move one sampling position backward along the sampling time sequence, so that the data in the previous continuous sampling window, except for the earliest sampling time, are included in the next continuous sampling window. This process is repeated continuously from the starting sampling position to the ending sampling position of the commutation process. This yields continuous sampling windows that cover the entire commutation process and are continuously arranged according to the sampling time sequence, as well as the corresponding output torque and transmission error displacement within each continuous sampling window.

[0028] Using the transmission error displacement corresponding to the sampling time within each continuous sampling window as the independent variable and the output torque at the same sampling time as the dependent variable, a linear fit is performed on the transmission error displacement and output torque of each group within the window according to the original correspondence. During the fitting process, each transmission error displacement is kept constant, and the position and inclination of the straight line representing the relationship between the transmission error displacement and the output torque are adjusted. The minimum summation of the squared differences between the actual output torque within the window and the stress torque value of the straight line at the same transmission error displacement is taken as the fitting completion condition. The fitted straight line is then moved along with the transmission... The degree of change in the output torque corresponding to the change in error displacement is determined as the response slope. The difference between the actual output torque and the stress torque value of the fitted straight line is used to determine the fitting residual. At the same time, the maximum and minimum values ​​of the transmission error displacement within the window are read, and the difference between the two is determined as the transmission error displacement variation range of the continuous sampling window. When the variation range is equal to or lower than the angular displacement detection resolution calibrated by the angular displacement detection device, the change in transmission error displacement within the window cannot be distinguished from the minimum angular displacement change that the detection device can distinguish. Therefore, the continuous sampling window is not used for load condition determination.

[0029] For each continuous sampling window where the range of transmission error displacement exceeds the resolution of angular displacement detection, the allowable deviation of the response slope is determined based on the corresponding fitting residual. Specifically, first, the absolute value of the difference between the actual output torque and the stress torque value of the fitted straight line at the same transmission error displacement within the window is taken, and the maximum torque deviation is determined from this. Then, the torque values ​​of the fitted straight line at the minimum and maximum transmission error displacement positions within the window are read respectively. When constructing the first allowable boundary, the maximum torque deviation is adjusted by shifting the torque value at the minimum transmission error displacement position in the direction of decreasing torque, and by shifting the torque value at the maximum transmission error displacement position in the direction of increasing torque. The opposite adjustment direction is used when constructing the second allowable boundary. The corresponding boundary slope is obtained by connecting the two adjusted endpoints of each allowable boundary, and the difference with the larger absolute value among the differences between the two boundary slopes and the original response slope is determined as the allowable deviation of the response slope for that continuous sampling window. For two consecutive sampling windows adjacent in the sampling time sequence, the absolute value of the difference between their response slopes is compared with the sum of the allowable deviations of the response slopes of the two windows. If the former does not exceed the latter, the two windows are determined to maintain a continuous torque-displacement response relationship; if the former exceeds the latter, the response relationship is determined to have changed. In this specification, the "allowable deviation of corresponding response slope" for adjacent consecutive sampling windows refers to the sum of the allowable deviations of the response slopes of the two adjacent consecutive sampling windows.

[0030] Before determining the pre-commutation load zone, zero-point calibration of the output torque is performed using data from no-load or known zero-torque states. Torque data that does not exceed the output torque detection resolution are considered zero-torque data and are not included in the pre-commutation or post-commutation load zone. The pre-commutation torque direction is determined based on the signs of the output torques that exceed the output torque detection resolution and appear consecutively along the sampling time sequence in the continuous sampling data before commutation. The direction opposite to the pre-commutation torque direction is determined as the post-commutation torque direction. From the calibrated sampling data along the pre-commutation torque direction, consecutive sampling windows are checked sequentially along the sampling time sequence. Output torques used for load state determination within each window are considered to be along the pre-commutation torque direction as a pre-commutation torque transmission condition. The difference in response slope between the current and subsequent consecutive sampling windows is considered to be within the corresponding allowable range as a response continuity condition. When adjacent windows continuously meet two conditions, their covered data are classified into the same candidate segment; when multiple candidate segments appear, the continuous segment that is adjacent to the switching process and is the last one formed along the sampling time sequence is selected as the pre-switching carrying area; when the response slope difference of the adjacent continuous sampling windows at the end of the segment first changes from the corresponding allowable range to exceed the corresponding allowable range, the starting sampling position of the next continuous sampling window is taken as the ending position of the pre-switching carrying area.

[0031] From the calibrated sampling data along the direction of the transmitted torque after commutation, the continuous sampling windows are checked sequentially along the sampling time sequence. The condition that the transmitted torque at the output end used for load-bearing status determination within the window is along the direction of the transmitted torque after commutation is taken as the condition for torque transmission after commutation. The condition that the difference in response slope between the current continuous sampling window and the next continuous sampling window is within the corresponding allowable range is taken as the condition for response continuity. When adjacent windows continuously meet both conditions, their covered data are grouped into the same candidate segment. When multiple candidate segments appear, the continuous segment adjacent to the commutation process and formed earliest along the sampling time sequence is selected as the load-bearing area after commutation. When the difference in response slope between adjacent continuous sampling windows at the beginning of this segment first changes from exceeding the corresponding allowable range to being within the corresponding allowable range, the starting sampling position of the next continuous sampling window is taken as the starting position of the load-bearing area after commutation.

[0032] The end position of the pre-commutation load-bearing area and the start position of the post-commutation load-bearing area are determined along a unified sampling time sequence. All continuous sampling data between the end position of the pre-commutation load-bearing area and the start position of the post-commutation load-bearing area are retained according to the original sampling time sequence, and this continuous sampling segment is identified as the commutation transition area, thus forming a sequentially connected pre-commutation load-bearing area, commutation transition area, and post-commutation load-bearing area. The direction of the transmitted torque before commutation can correspond to either the positive or negative direction of the calibrated torque data, while the direction of the transmitted torque after commutation is always opposite. Therefore, this method is applicable to both the conversion of the transmitted torque at the output end from positive to negative and from negative to positive. If the pre-commutation load-bearing area or the post-commutation load-bearing area fails to form, the timing relationship between the two load-bearing areas does not conform to the above order, or the effective continuous sampling window is insufficient to determine the boundary, the current commutation test is marked as invalid, and data is re-acquired; subsequent boundary calculations and test result generation are not performed.

[0033] The beneficial effects are as follows: by using continuous sampling windows to maintain the local correspondence between the output torque and the transmission error displacement during the commutation process, and by using the resolution of angular displacement detection to exclude continuous sampling windows that cannot determine the effective transmission error displacement change, the load state determination is established within the data range that the detection device can resolve. The response slope and fitting residual are obtained through linear fitting, and the allowable deviation of the response slope of each continuous sampling window is determined by using the actual torque deviation range corresponding to the fitting residual. This ensures that the continuity of the response relationship between adjacent continuous sampling windows has a judgment criterion that matches the discrete state of the corresponding sampling data. Then, by combining the directional state of the output torque relative to the direction of the transmission torque before and after commutation, the load zone before commutation and the load zone after commutation are determined respectively. The commutation transition zone is determined by using the continuous sampling data between the two load zones. Thus, the load state division result is obtained with continuous sampling time sequence and determined segment transition position, providing a clear data segment basis for the subsequent determination of the first and second load boundaries.

[0034] S3. Determine the first load-bearing boundary at the position where the pre-commutation load-bearing area transitions to the commutation transition area, and determine the second load-bearing boundary at the position where the commutation transition area transitions to the post-commutation load-bearing area. In this embodiment of the invention, determining the first bearing boundary at the position where the pre-reversing bearing area transitions to the reversing transition area, and determining the second bearing boundary at the position where the reversing transition area transitions to the post-reversing bearing area, includes: Read the continuous sampling windows adjacent to the pre-commutation load area and the commutation transition area, and take the position where the response slope difference between the adjacent continuous sampling windows changes from being within the corresponding response slope allowable deviation to exceeding the corresponding response slope allowable deviation as the first load boundary; Read the continuous sampling windows adjacent to the commutation transition zone and the load-bearing zone after commutation, and use the position where the difference in response slope between adjacent continuous sampling windows changes from exceeding the allowable deviation of the corresponding response slope to being within the allowable deviation of the corresponding response slope as the second load-bearing boundary.

[0035] Read the continuous sampling windows corresponding to the end of the pre-commutation bearing area and the beginning of the adjacent commutation transition area. Then, according to the original sampling sequence, check the relationship between the response slope difference of adjacent continuous sampling windows and the corresponding allowable deviation of the response slope window, window by window, from the pre-commutation bearing area to the commutation transition area. If the response slope difference is equal to or lower than the corresponding allowable deviation, it is considered to be within the allowable deviation; if the response slope difference is higher than the allowable deviation, it is considered to exceed the allowable deviation. When the response slope difference of the previous group of adjacent continuous sampling windows is still within the allowable deviation, the process continues. If the response slope difference of the next set of adjacent consecutive sampling windows exceeds the corresponding response slope allowable deviation for the first time within the allowable deviation range, it is determined that the original pre-conversion bearing response relationship between the output torque and the transmission error displacement ends at this position. The starting sampling position of the next consecutive sampling window after the first exceedance is taken as the position for the transition from the pre-conversion bearing area to the conversion transition area. This sampling position is determined as the first bearing boundary. At the same time, the transmission error displacement and output torque at the sampling time corresponding to the first bearing boundary are read and associated with the first bearing boundary to provide boundary data for the subsequent conversion of the transmission error displacement of the first bearing boundary to a unified transmission torque state.

[0036] Read the continuous sampling windows corresponding to the end of the commutation transition zone and the beginning of the commutation load-bearing zone adjacent to that end. Check the relationship between the response slope difference of adjacent continuous sampling windows and the corresponding allowable deviation of the response slope from the commutation transition zone to the commutation load-bearing zone according to the original sampling sequence. When the response slope difference of the previous group of adjacent continuous sampling windows still exceeds the corresponding allowable deviation of the response slope, and the response slope difference of the next group of adjacent continuous sampling windows is equal to or lower than the corresponding allowable deviation of the response slope for the first time, it is determined that the response relationship conversion in the commutation process has ended and the continuous commutation load-bearing response relationship is established. The starting sampling position of the next continuous sampling window when the first entry into the corresponding allowable deviation of the response slope occurs is taken as the position of the conversion from the commutation transition zone to the commutation load-bearing zone. This sampling position is determined as the second load-bearing boundary. At the same time, read the transmission error displacement and output torque at the sampling time corresponding to the second load-bearing boundary and record them in association with the second load-bearing boundary. Thus, the first load-bearing boundary and the second load-bearing boundary are obtained by determining along the same sampling sequence and using the same boundary positioning rule.

[0037] The beneficial effect is that by checking the state transition between the response slope difference and the corresponding allowable deviation of the response slope of adjacent consecutive sampling windows along the sampling time sequence, the position where the pre-commutation load response relationship ends is determined as the first load boundary, and the position where the post-commutation load response relationship begins to be continuously established is determined as the second load boundary. This ensures that both load boundaries are determined by the actual transition position of the response relationship between the output torque and the transmission error displacement. At the same time, using the starting sampling position of the next consecutive sampling window as a unified boundary positioning reference can avoid inconsistencies in the selection of boundary sampling positions caused by the overlap of consecutive sampling windows. Furthermore, both the first and second load boundaries have corresponding transmission error displacements and output torques, providing a clear positional basis and a definite data correspondence for subsequent boundary displacement conversion based on the pre-commutation elastic reference relationship and the post-commutation elastic reference relationship.

[0038] S4. Extract the elastic response intervals before and after the commutation from the pre-commutation load area and the post-commutation load area respectively, and establish the pre-commutation elastic reference relationship and the post-commutation elastic reference relationship based on the pre-commutation elastic response interval and the post-commutation elastic response interval. In this embodiment of the invention, the step of extracting the pre-commutation elastic response interval and the post-commutation elastic response interval from the pre-commutation load-bearing region and the post-commutation load-bearing region, respectively, includes: A continuous sampling window is set along the sampling timing of the pre-commutation and post-commutation load-bearing areas to obtain the adjacent sampling changes of the output torque and transmission error displacement within each continuous sampling window; According to the sampling order, the output torque change and transmission error displacement change within the same continuous sampling window are used as vector components to construct torque change vector and displacement change vector respectively. The direction consistency coefficient of each continuous sampling window is determined based on the cosine of the angle between the torque change vector and the displacement change vector. The limit value of the direction consistency coefficient is determined based on the detection resolution of the output torque and transmission error displacement, as well as the magnitude of the torque change vector and displacement change vector. Continuous sampling segments with a direction consistency coefficient not lower than the corresponding limit value are extracted along the sampling time sequence to obtain the elastic response interval before commutation and the elastic response interval after commutation.

[0039] The determination of the limiting value of the direction consistency coefficient based on the detection resolution of the output torque and transmission error displacement, as well as the magnitudes of the torque change vector and displacement change vector, includes: Based on the relationship between the detection resolution of the torque transmitted at the output end and the magnitude of the torque change vector, the allowable deviation of the direction corresponding to the torque change vector is determined; Based on the relationship between the detection resolution of the transmission error displacement and the magnitude of the displacement change vector, the allowable deviation of the direction corresponding to the displacement change vector is determined. The allowable included angle is determined based on the permissible deviation of the directions corresponding to the torque change vector and the displacement change vector, and the cosine value of the included angle corresponding to the allowable included angle is used as the limiting value of the direction consistency coefficient.

[0040] The establishment of the pre-commutation elastic reference relationship and the post-commutation elastic reference relationship includes: Obtain the changes in output transmitted torque and transmission error displacement at adjacent sampling times within the elastic response interval before and after commutation. Based on the ratio between the change in transmission error displacement and the change in the corresponding output torque, the unit torque displacement change in the elastic response range before and after commutation is determined respectively. By selecting the sampling time in the middle of the pre-commutation elastic response interval and the post-commutation elastic response interval along the sampling time sequence, the corresponding output end transmission torque and transmission error displacement are obtained. Based on the unit torque displacement change, the corresponding change of transmission error displacement when the output end transmission torque changes is determined, thus obtaining the pre-commutation elastic reference relationship and the post-commutation elastic reference relationship.

[0041] Continuous sampling windows are set up according to the original sampling sequence along the determined pre-commutation and post-commutation load areas. Each continuous sampling window uses the same number of sampling data and moves to the sampling position one by one along the sampling sequence. For each continuous sampling window, the output end transmission torque and transmission error displacement at two adjacent sampling times are read in sequence. The change in output end transmission torque at the later sampling time relative to the previous sampling time is recorded as the corresponding change in output end transmission torque, and the change in transmission error displacement at the later sampling time relative to the previous sampling time is recorded as the corresponding change in transmission error displacement. This ensures that the same number of output end transmission torque changes and transmission error displacement changes with one-to-one position are obtained in the sampling sequence within a continuous sampling window.

[0042] Following the original sequence of adjacent sampling positions within the continuous sampling window, the torque changes at each output end are arranged sequentially and used as components of the torque change vector at the corresponding sampling positions, thus constructing the torque change vector corresponding to the continuous sampling window. Simultaneously, the transmission error displacement changes at the same adjacent sampling positions as the aforementioned output torque changes are arranged in the same order and used as corresponding components of the displacement change vector, thus constructing the displacement change vector corresponding to the continuous sampling window. This ensures that the torque change vector and the displacement change vector have the same number of components, the same component arrangement order, and the same corresponding sampling positions. The two vectors respectively represent the distribution of the output torque change direction and the transmission error displacement change direction within the same continuous sampling window throughout the entire sampling segment.

[0043] The degree of directional similarity between the torque change vector and displacement change vector corresponding to the same continuous sampling window is determined. Specifically, the components with the same sampling position in the two vectors are combined accordingly. The angle between the overall directions of the two vectors is determined by the consistency of the positive and negative directions and the degree of change of the corresponding components. The cosine value of the angle is read as the directional consistency coefficient of the continuous sampling window. When the output torque and transmission error displacement change synchronously in similar directions within the window, the angle between the two vectors decreases and the directional consistency coefficient increases accordingly. When the two vectors are inconsistent in direction at different sampling positions or the corresponding relationship of change changes, the angle between the two vectors increases and the directional consistency coefficient decreases accordingly. Thus, the directional consistency coefficient is used to characterize the elastic response direction of whether the output torque change and the transmission error displacement change are consistent within the same continuous sampling window.

[0044] When determining the permissible deviation of the direction corresponding to the torque change vector based on the detection resolution of the output torque, the torque change vector formed by the continuous sampling window is used as a reference. According to the minimum torque change that the output torque detection device can distinguish, permissible disturbance boundaries are set for each component in the increasing and decreasing directions. In this specification, the effective components of the torque change vector and displacement change vector refer to vector components whose absolute value of change exceeds the corresponding detection resolution and whose positive or negative direction can be determined. Components that do not exceed the corresponding detection resolution are still retained in the vector, but their positive or negative direction is not used as a constraint to maintain the original direction of change. Without changing the original direction of change of the effective components, the permissible disturbance boundary values ​​of each component are selected and combined to determine the combination with the largest deflection relative to the original torque change vector direction. This maximum deflection is taken as the permissible deviation of the direction corresponding to the torque change vector. When the magnitude of the torque change vector is zero, the changes of each component cannot be distinguished from the detection resolution, or the permissible disturbance makes it impossible to determine the direction of change of the effective components, the permissible deviation of the direction is not calculated for this window, and the window is marked as invalid.

[0045] When determining the permissible directional deviation corresponding to the displacement change vector based on the detection resolution of the transmission error displacement, the displacement change vector within the same continuous sampling window is used as the reference. Based on the minimum angular displacement change that the angular displacement detection device can resolve, permissible disturbance boundaries are set for each component in both the increasing and decreasing directions. Without changing the original direction of change of the effective component, the permissible disturbance boundary values ​​for each component are selected and combined. The combination with the largest deflection relative to the original displacement change vector direction is determined, and this maximum deflection is taken as the permissible directional deviation corresponding to the displacement change vector. When the magnitude of the displacement change vector is zero, the changes of each component cannot be distinguished from the detection resolution, or the permissible disturbance makes it impossible to determine the direction of change of the effective component, the permissible directional deviation is not calculated for that window, and the window is marked as invalid.

[0046] For the same continuous sampling window where both the torque change vector and displacement change vector are valid, the permissible deviations of their directions are added together to obtain the permissible angle within which the two types of detected quantities in the window can still be considered to maintain a unidirectional elastic response relationship. The permissible angle should be within a valid angle range that can uniquely determine the cosine value of the angle. When either vector is invalid, or when the sum of the permissible deviations of the two directions cannot form a distinguishable directional determination range, the continuous sampling window is marked as invalid and not included in the elastic response interval extraction. For valid windows, the cosine value of the angle corresponding to the permissible angle is determined as the limiting value of the directional consistency coefficient.

[0047] The directional consistency coefficient of each continuous sampling window is compared with the corresponding limit value along the sampling time sequence of the pre-commutation load area and the post-commutation load area respectively. The continuous sampling window with the directional consistency coefficient equal to or higher than the corresponding limit value is identified as the window in which the change of the output torque and the change of the transmission error displacement maintain an elastic directional consistency relationship. The data covered by the continuous sampling windows that continuously meet the condition along the sampling time sequence are merged into the same continuous sampling segment. When the directional consistency coefficient of the next continuous sampling window is lower than the corresponding limit value, the extraction of the current continuous sampling segment ends. The continuous sampling segment obtained according to the condition in the pre-commutation load area is determined as the pre-commutation elastic response interval, and the continuous sampling segment obtained according to the same condition in the post-commutation load area is determined as the post-commutation elastic response interval. In this way, both the pre-commutation elastic response interval and the post-commutation elastic response interval correspond to the load data in which the change of the output torque and the change of the transmission error displacement continuously maintain a directional consistency relationship.

[0048] The sampled data, arranged in the original sampling sequence, are read from both the pre-commutation and post-commutation elastic response intervals. For every two adjacent sampling times, the change in output torque and the change in transmission error displacement are determined sequentially, and their corresponding sampling positions are retained. When determining the unit torque displacement change, only adjacent sampled data whose absolute value of the output torque change exceeds the output torque detection resolution are used. The positive and negative directions of both the output torque change and the transmission error displacement change are retained for subsequent processing. The data in the pre-commutation and post-commutation elastic response intervals are independent. If no valid adjacent sampled data exists in any interval, that interval is marked as invalid, and data is re-acquired; the corresponding unit torque displacement change is not calculated further.

[0049] For each set of valid adjacent sampled data within the pre-commutation elastic response interval, the corresponding transmission error displacement change is converted into the displacement change corresponding to a unit output torque change based on the output torque change. The arithmetic mean of the results obtained from all valid adjacent sampled data within the pre-commutation elastic response interval is then taken as the unit torque displacement change within the pre-commutation elastic response interval. The same processing method is used for the post-commutation elastic response interval. The arithmetic mean of the unit output torque displacement change results corresponding to each valid adjacent sampled data is taken as the unit torque displacement change within the post-commutation elastic response interval. Thus, the unit torque displacement change is obtained to characterize the degree of change in transmission error displacement corresponding to a unit change in output torque under the pre-commutation and post-commutation load conditions.

[0050] The starting and ending sampling positions of the pre-commutation elastic response interval and the post-commutation elastic response interval are determined according to the sampling sequence. The sampling time in the middle of the sequence from the starting to the ending sampling position is selected as the reference sampling time for the corresponding elastic response interval. When the number of sampling positions is even, the earlier sampling time of the two middle sampling positions is selected as the unified selection reference. The output torque and transmission error displacement corresponding to the reference sampling time of the pre-commutation elastic response interval are read, and this set of data is used as the reference point for the pre-commutation elastic response relationship. The change in unit torque displacement in the pre-commutation elastic response interval is used as the corresponding scale between the change in output torque and the change in transmission error displacement. When any detectable change occurs in the output torque relative to this reference point, the change in transmission error displacement relative to the reference point is determined according to the change in unit torque displacement, thus obtaining the pre-commutation elastic reference relationship.

[0051] The output torque and transmission error displacement corresponding to the reference sampling time in the middle of the elastic response interval after reversal are read, and this set of data is used as the reference point for the elastic response relationship after reversal. The unit torque displacement change in the elastic response interval after reversal is used as the corresponding scale between the output torque change and the transmission error displacement change under the load state after reversal. When the output torque changes with respect to the reference point, the change in transmission error displacement is determined according to the unit torque displacement change in the elastic response interval after reversal. Thus, the elastic reference relationship after reversal is obtained, so that the elastic reference relationship before reversal and the elastic reference relationship after reversal correspond to the elastic displacement response characteristics of the robot joint under the two load directions, respectively, and provide a corresponding relationship for the subsequent conversion of the transmission error displacement of the first load boundary and the second load boundary to the same transmission torque state.

[0052] The beneficial effect is that by constructing the output torque change and transmission error displacement change within the continuous sampling window into torque change vectors and displacement change vectors with the same sampling order, and using the degree of directional consistency between the two vectors to filter the pre-commutation elastic response interval and the post-commutation elastic response interval, the data involved in establishing the elastic reference relationship continuously maintains the corresponding direction of torque change and displacement change. At the same time, based on the detection resolution of the two types of detection data and the actual degree of change of the corresponding change vectors, the limit values ​​of the allowable deviation of direction and the directional consistency coefficient are determined, so that the judgment condition of the elastic response interval corresponds to the detectability of the current data. Then, based on the pre-commutation elastic response interval and the post-commutation elastic response interval, the unit torque displacement change is determined and the pre-commutation elastic reference relationship and the post-commutation elastic reference relationship are established, thereby obtaining the basis for the independent elastic displacement response of the two bearing directions, providing a basis for subsequently eliminating the elastic displacement difference caused by the influence of different transmission torque states on the first bearing boundary and the second bearing boundary.

[0053] S5. Based on the elastic reference relationship before and after the switching, the transmission error displacement of the first bearing boundary and the second bearing boundary is converted to the zero transmission torque state to obtain the displacement of the first reference boundary and the displacement of the second reference boundary. In this embodiment of the invention, the step of converting the transmission error displacement of the first bearing boundary and the second bearing boundary to a zero transmission torque state based on the pre-commutation elastic reference relationship and the post-commutation elastic reference relationship, to obtain the first reference boundary displacement and the second reference boundary displacement, includes: Obtain the output torque and transmission error displacement of the first and second load-bearing boundaries; Based on the pre-transformation elastic reference relationship and the output end transmitted torque and transmission error displacement of the first bearing boundary, the transmission error displacement corresponding to the zero transmitted torque state is determined, and the first reference boundary displacement is obtained. Based on the elastic reference relationship after reversal and the output end transmitted torque and transmission error displacement of the second bearing boundary, the transmission error displacement corresponding to the zero transmitted torque state is determined, and the displacement of the second reference boundary is obtained.

[0054] Read the sampling positions corresponding to the determined first and second bearing boundaries. Based on the unique correspondence of each sampling position in the original sampling sequence, extract the output torque and transmission error displacement corresponding to the first bearing boundary and the output torque and transmission error displacement corresponding to the second bearing boundary, respectively, while keeping the two sets of boundary data consistent with the original bearing direction. The first bearing boundary corresponds to the position where the bearing area before the changeover transitions to the changeover transition area, and the second bearing boundary corresponds to the position where the changeover transition area turns to the bearing area after the changeover. Thus, obtain the output torque and transmission error displacement of the first bearing boundary and the output torque and transmission error displacement of the second bearing boundary used for zero transmission torque state conversion.

[0055] Using the pre-commutation elastic reference relationship as the basis for displacement conversion of the first bearing boundary, the output end transmission torque and transmission error displacement corresponding to the first bearing boundary are read, and the direction and amount of torque change corresponding to the output end transmission torque returning to the zero transmission torque state are determined. Based on the degree of transmission error displacement change corresponding to the unit output end transmission torque change in the pre-commutation elastic reference relationship, the elastic displacement change corresponding to the above torque change is determined. Then, along the displacement change direction determined by the pre-commutation elastic reference relationship, the elastic displacement component caused by the current pre-commutation transmission torque is subtracted from the transmission error displacement of the first bearing boundary. The transmission error displacement corresponding to the zero transmission torque state after this direction correction is determined as the first reference boundary displacement.

[0056] Using the post-commutation elastic reference relationship as the basis for displacement conversion of the second bearing boundary, the output end transmitted torque and transmission error displacement corresponding to the second bearing boundary are read, and the direction and amount of torque change corresponding to the return of the output end transmitted torque to the zero transmitted torque state are determined. Based on the degree of transmission error displacement change corresponding to the unit output end transmitted torque change in the post-commutation elastic reference relationship, the amount of elastic displacement change corresponding to the above torque change is determined. Then, along the displacement change direction determined by the post-commutation elastic reference relationship, the elastic displacement component caused by the current post-commutation transmitted torque is subtracted from the transmission error displacement of the second bearing boundary. The transmission error displacement corresponding to the zero transmitted torque state after this direction correction is determined as the second reference boundary displacement.

[0057] The first and second reference boundary displacements are both recorded under the same zero-transmission torque state. This ensures that the two reference boundary displacements no longer use the original output torque of their respective bearing boundaries as the displacement comparison condition, while maintaining the relationship between the first reference boundary displacement and the first bearing boundary, and the second reference boundary displacement and the second bearing boundary. This yields the first and second reference boundary displacements under the same transmission torque reference condition, providing a unified displacement comparison basis for subsequently determining the robot joint transmission backlash using the positional difference between the two reference boundary displacements. To clarify the above conversion relationship, under the condition that the aforementioned angular displacement starting position and rotation direction definitions remain consistent, and all data retain positive and negative directions, the first reference boundary displacement can be determined by subtracting the product of the pre-commutation unit torque displacement change and the output torque of the first bearing boundary from the transmission error displacement of the first bearing boundary; the second reference boundary displacement can be determined by subtracting the product of the post-commutation unit torque displacement change and the output torque of the second bearing boundary from the transmission error displacement of the second bearing boundary. The aforementioned unit torque displacement change is determined by the elastic reference relationship before and after the changeover, respectively. When the corresponding elastic response interval is marked as invalid according to the aforementioned rules, the zero transmission torque state conversion of the corresponding boundary is not performed, and the data is re-acquired.

[0058] The beneficial effect is that by using the elastic reference relationship before and after the commutation to perform zero transmission torque state conversion on the transmission error displacements corresponding to the first and second bearing boundaries, the elastic displacement components caused by the transmission torque at their respective output ends in the two bearing boundaries are corrected accordingly. This transforms the first and second bearing boundaries to the same transmission torque reference conditions, resulting in the first and second reference boundary displacements that can be directly compared in position. This avoids the elastic deformation from being mixed into the boundary displacement differences due to the different bearing directions and transmission torque states of the two bearing boundaries, providing a unified and clearly corresponding boundary displacement basis for subsequent determination of transmission backlash.

[0059] S6. Based on the first and second reference boundary displacements, determine the transmission backlash of the robot joint, and based on the elastic reference relationship before and after the reversal, determine the transmission stiffness before and after the reversal, and generate the robot joint transmission test results.

[0060] In this embodiment of the invention, the process of determining the transmission backlash of the robot joint based on the first and second reference boundary displacements, and determining the transmission stiffness before and after the commutation based on the pre-commutation and post-commutation elastic reference relationships, and generating robot joint transmission test results, includes: Determine the displacement interval between the first reference boundary displacement and the second reference boundary displacement, and use the displacement interval as the transmission backlash of the robot joint; Based on the proportional relationship between the change in output torque and the change in transmission error displacement in the elastic reference relationship before and after commutation, the change in output torque corresponding to a unit change in transmission error displacement is determined, and the transmission stiffness before and after commutation is obtained. The transmission backlash, transmission stiffness before reversal, and transmission stiffness after reversal are correlated and recorded to obtain the test results of the robot joint transmission.

[0061] The first and second reference boundary displacements, already converted to a zero-transmission torque state, are read, ensuring they are defined under the same transmission torque reference, angular displacement starting position, and rotation direction. The difference between the two reference boundary displacements is calculated, and the absolute value of this angular displacement difference is taken. This absolute angular displacement interval is defined as the displacement range during which the robot joint transitions from one load-bearing contact state to another load-bearing contact state before effective commutation occurs at the output end. This displacement range is then defined as the transmission backlash of the robot joint. This process no longer includes the elastic displacements corresponding to the original transmission torques of the first and second load-bearing boundaries in the transmission backlash.

[0062] The correspondence between the output torque change and the transmission error displacement change in the pre-commutation elastic reference relationship is read. Using the unit transmission error displacement change as a unified displacement change reference, the output torque change corresponding to this unit displacement change is determined, and the absolute value of this torque change is taken as the pre-commutation transmission stiffness. The post-commutation transmission stiffness is determined in the same way by the post-commutation elastic reference relationship. When the transmission error displacement is recorded in radians, the transmission stiffness is expressed in Newton-meters per radian; when recorded in angles, the transmission stiffness is expressed in Newton-meters per degree, maintaining the same unit throughout. When the correspondence cannot produce a valid result for the unit displacement torque change, the stiffness test in that direction is marked as invalid, and data is re-acquired.

[0063] The determined transmission backlash, pre-commutation transmission stiffness, and post-commutation transmission stiffness are associated and recorded according to the same robot joint reversal test. At the same time, the first and second reference boundary displacements corresponding to the transmission backlash, as well as the bearing directions corresponding to the pre-commutation and post-commutation transmission stiffness, are retained. This allows the same test record to simultaneously characterize the clearance transmission characteristics of the robot joint and the elastic bearing characteristics under two bearing directions. The associated transmission backlash, pre-commutation transmission stiffness, and post-commutation transmission stiffness are then determined as the robot joint transmission test results.

[0064] The beneficial effects are as follows: by comparing the first and second reference boundary displacements under zero transmission torque conditions, the transmission backlash can be determined based on the boundary displacement interval under a unified bearing reference. The change in output transmission torque corresponding to the unit transmission error displacement change is determined by using the elastic reference relationship before and after reversal. The transmission stiffness before and after reversal corresponds to the actual elastic response characteristics of the robot joint in the two bearing directions, respectively. The transmission backlash, transmission stiffness before and after reversal are then linked and recorded. Thus, in the same reversal test, robot joint transmission test results that can simultaneously reflect transmission backlash characteristics and bidirectional elastic bearing characteristics are obtained, improving the completeness of robot joint transmission performance characterization and the consistency between results.

[0065] like Figure 2 The diagram shown is a functional block diagram of a robot joint transmission backlash stiffness joint testing system provided in an embodiment of the present invention.

[0066] The robot joint transmission backlash stiffness joint testing system 100 described in this invention can be installed in an electronic device. Depending on the functions implemented, the robot joint transmission backlash stiffness joint testing system 100 may include a data acquisition module 101, a load identification module 102, a boundary determination module 103, a benchmark construction module 104, a boundary conversion module 105, and a test output module 106. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and are stored in the memory of the electronic device.

[0067] In this embodiment, the functions of each module / unit are as follows: The data acquisition module 101 is used to convert the input end angular displacement during the robot joint reversal process into the theoretical output end angular displacement, compare it with the actual output end angular displacement, obtain the transmission error displacement, and obtain the output end transmitted torque. The load identification module 102 is used to identify the pre-commutation load area, the commutation transition area, and the post-commutation load area based on the continuous change of the output torque and transmission error displacement during the commutation process. The boundary determination module 103 is used to determine the first bearing boundary at the position where the bearing area before the changeover transition area turns into the changeover transition area, and to determine the second bearing boundary at the position where the changeover transition area turns into the bearing area after the changeover. The benchmark construction module 104 is used to extract the elastic response interval before switching and the elastic response interval after switching from the pre-switching load area and the post-switching load area respectively, and to establish the elastic benchmark relationship before switching and the elastic benchmark relationship after switching based on the elastic response interval before switching and the elastic response interval after switching. The boundary conversion module 105 is used to convert the transmission error displacement of the first bearing boundary and the second bearing boundary to a zero transmission torque state based on the elastic reference relationship before and after the conversion, so as to obtain the first reference boundary displacement and the second reference boundary displacement. The test output module 106 is used to determine the transmission backlash of the robot joint based on the first reference boundary displacement and the second reference boundary displacement, and to determine the transmission stiffness before and after the switching based on the elastic reference relationship before and after the switching, and to generate the robot joint transmission test results. The above modules are associated according to the data processing sequence of the same sampling time: the data acquisition module 101 provides the transmission error displacement and the output torque to the load identification module 102; the load identification module 102 outputs the pre-commutation load area, commutation transition area and post-commutation load area to the boundary determination module 103 to determine the first load boundary and the second load boundary, and to the reference construction module 104 to extract the pre-commutation elastic response interval and the post-commutation elastic response interval and establish the corresponding elastic reference relationship; the first load boundary, the second load boundary and their corresponding data output by the boundary determination module 103 and the pre-commutation elastic reference relationship and the post-commutation elastic reference relationship output by the reference construction module 104 are jointly provided to the boundary conversion module 105 to obtain the first reference boundary displacement and the second reference boundary displacement; the test output module 106 determines the transmission backlash based on the first reference boundary displacement and the second reference boundary displacement, and determines the pre-commutation transmission stiffness and the post-commutation transmission stiffness based on the pre-commutation elastic reference relationship and the post-commutation elastic reference relationship.

[0068] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0069] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0070] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0071] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0072] The data processing objects in this embodiment are the input end angular displacement, actual output end angular displacement, output end transmitted torque, and intermediate results determined by the above data, obtained according to a unified sampling time sequence. Each processing step is executed by the processor of the electronic device in the order described in this application. The resulting transmission backlash, transmission stiffness before reversal, and transmission stiffness after reversal are used as the output results of the same robot joint reversal test.

[0073] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for jointly testing the backlash stiffness of robot joint transmissions, characterized in that, The method includes: S1. Convert the input end angular displacement during the robot joint reversal process into the theoretical output end angular displacement, compare it with the actual output end angular displacement, obtain the transmission error displacement, and obtain the output end transmitted torque. S2. Based on the continuous change of the output torque and transmission error displacement during the commutation process, identify the pre-commutation load area, commutation transition area, and post-commutation load area. S3. Determine the first load-bearing boundary at the position where the pre-commutation load-bearing area transitions to the commutation transition area, and determine the second load-bearing boundary at the position where the commutation transition area transitions to the post-commutation load-bearing area. S4. Extract the elastic response intervals before and after the commutation from the pre-commutation load area and the post-commutation load area respectively, and establish the pre-commutation elastic reference relationship and the post-commutation elastic reference relationship based on the pre-commutation elastic response interval and the post-commutation elastic response interval. S5. Based on the elastic reference relationship before and after the switching, the transmission error displacement of the first bearing boundary and the second bearing boundary is converted to the zero transmission torque state to obtain the displacement of the first reference boundary and the displacement of the second reference boundary. S6. Based on the first and second reference boundary displacements, determine the transmission backlash of the robot joint, and based on the elastic reference relationship before and after the reversal, determine the transmission stiffness before and after the reversal, and generate the robot joint transmission test results.

2. The method for joint testing of backlash stiffness in robot joint transmission as described in claim 1, characterized in that, The process of converting the input angular displacement during the robot joint reversal process into a theoretical output angular displacement, comparing it with the actual output angular displacement to obtain the transmission error displacement, and acquiring the output transmitted torque includes: During the process of the robot joint switching from the forward motion direction to the opposite backward motion direction, the input end angular displacement, the actual output end angular displacement, and the output end transmitted torque are obtained according to the unified sampling timing sequence. The theoretical angular displacement of the output end is determined based on the transmission ratio between the input and output ends of the robot joint and the angular displacement of the input end at each sampling time. The transmission error displacement is obtained by processing the difference between the theoretical angular displacement and the actual angular displacement of the output end at the same sampling time. According to the sampling sequence, the transmission error displacement and the output torque are recorded together.

3. The method for joint testing of backlash stiffness in robot joint transmission as described in claim 1, characterized in that, The method of identifying the pre-commutation load-bearing zone, commutation transition zone, and post-commutation load-bearing zone based on the continuous change of the output torque and transmission error displacement during the commutation process includes: Set continuous sampling windows along the sampling time sequence to obtain the output torque and transmission error displacement within each continuous sampling window; Using the transmission error displacement as the independent variable and the output torque as the dependent variable, the output torque and transmission error displacement within each continuous sampling window are linearly fitted to obtain the response slope and fitting residual of the continuous sampling window. Among them, when the range of transmission error displacement within the continuous sampling window does not exceed the angular displacement detection resolution, the continuous sampling window is not used for load state determination. The allowable deviation of the response slope for each consecutive sampling window is determined based on the fitting residual, and the difference in response slope between adjacent consecutive sampling windows is compared with the corresponding allowable deviation of the response slope. Based on the output torque sign that exceeds the output torque detection resolution and appears continuously along the sampling time sequence in the continuous sampling data before commutation, the direction of the transmission torque before commutation is determined, and the transmission direction opposite to the direction of the transmission torque before commutation is determined as the direction of the transmission torque after commutation; from the sampling data of the output torque along the direction of the transmission torque before commutation, the sampling segment that appears continuously along the sampling time sequence and whose response slope difference between adjacent continuous sampling windows is within the allowable deviation of the corresponding response slope is extracted to obtain the bearing area before commutation; From the sampled data of the torque transmitted from the output end along the direction of the torque transmitted after commutation, extract the sampling segment that appears continuously along the sampling time sequence and whose response slope difference between adjacent consecutive sampling windows is within the allowable deviation of the corresponding response slope, and obtain the bearing area after commutation; The sampling section between the end position of the pre-commutation bearing area and the beginning position of the post-commutation bearing area is identified as the commutation transition area.

4. The method for joint testing of backlash and stiffness of robot joint transmission as described in claim 3, characterized in that, Determining the first load-bearing boundary at the position where the pre-commutation load-bearing area transitions to the commutation transition area, and determining the second load-bearing boundary at the position where the commutation transition area transitions to the post-commutation load-bearing area, includes: Read the continuous sampling windows adjacent to the pre-commutation load area and the commutation transition area, and take the position where the response slope difference between the adjacent continuous sampling windows changes from being within the corresponding response slope allowable deviation to exceeding the corresponding response slope allowable deviation as the first load boundary; Read the continuous sampling windows adjacent to the commutation transition zone and the load-bearing zone after commutation, and use the position where the difference in response slope between adjacent continuous sampling windows changes from exceeding the allowable deviation of the corresponding response slope to being within the allowable deviation of the corresponding response slope as the second load-bearing boundary.

5. The method for joint testing of backlash stiffness in robot joint transmission as described in claim 1, characterized in that, The extraction of the pre-commutation elastic response interval and the post-commutation elastic response interval from the pre-commutation load-bearing region and the post-commutation load-bearing region, respectively, includes: A continuous sampling window is set along the sampling timing of the pre-commutation and post-commutation load-bearing areas to obtain the adjacent sampling changes of the output torque and transmission error displacement within each continuous sampling window; According to the sampling order, the output torque change and transmission error displacement change within the same continuous sampling window are used as vector components to construct torque change vector and displacement change vector respectively. The direction consistency coefficient of each continuous sampling window is determined based on the cosine of the angle between the torque change vector and the displacement change vector. The limit value of the direction consistency coefficient is determined based on the detection resolution of the output torque and transmission error displacement, as well as the magnitude of the torque change vector and displacement change vector. Continuous sampling segments with a direction consistency coefficient not lower than the corresponding limit value are extracted along the sampling time sequence to obtain the elastic response interval before commutation and the elastic response interval after commutation.

6. The method for joint testing of backlash stiffness in robot joint transmission as described in claim 5, characterized in that, The determination of the limiting value of the direction consistency coefficient based on the detection resolution of the output torque and transmission error displacement, as well as the magnitudes of the torque change vector and displacement change vector, includes: Based on the relationship between the detection resolution of the torque transmitted at the output end and the magnitude of the torque change vector, the allowable deviation of the direction corresponding to the torque change vector is determined; Based on the relationship between the detection resolution of the transmission error displacement and the magnitude of the displacement change vector, the allowable deviation of the direction corresponding to the displacement change vector is determined. The allowable included angle is determined based on the permissible deviation of the directions corresponding to the torque change vector and the displacement change vector, and the cosine value of the included angle corresponding to the allowable included angle is used as the limiting value of the direction consistency coefficient.

7. The method for joint testing of backlash and stiffness of robot joint transmission as described in claim 1, characterized in that, The establishment of the pre-commutation elastic reference relationship and the post-commutation elastic reference relationship includes: Obtain the changes in output transmitted torque and transmission error displacement at adjacent sampling times within the elastic response interval before and after commutation. Based on the ratio between the change in transmission error displacement and the change in the corresponding output torque, the unit torque displacement change in the elastic response range before and after commutation is determined respectively. By selecting the sampling time in the middle of the pre-commutation elastic response interval and the post-commutation elastic response interval along the sampling time sequence, the corresponding output end transmission torque and transmission error displacement are obtained. Based on the unit torque displacement change, the corresponding change of transmission error displacement when the output end transmission torque changes is determined, thus obtaining the pre-commutation elastic reference relationship and the post-commutation elastic reference relationship.

8. The method for joint testing of backlash stiffness in robot joint transmission as described in claim 1, characterized in that, The method, based on the pre-commutation elastic reference relationship and the post-commutation elastic reference relationship, converts the transmission error displacement of the first bearing boundary and the second bearing boundary to a zero transmission torque state, obtaining the first reference boundary displacement and the second reference boundary displacement, including: Obtain the output torque and transmission error displacement of the first and second load-bearing boundaries; Based on the pre-transformation elastic reference relationship and the output end transmitted torque and transmission error displacement of the first bearing boundary, the transmission error displacement corresponding to the zero transmitted torque state is determined, and the first reference boundary displacement is obtained. Based on the elastic reference relationship after reversal and the output end transmitted torque and transmission error displacement of the second bearing boundary, the transmission error displacement corresponding to the zero transmitted torque state is determined, and the displacement of the second reference boundary is obtained.

9. The method for joint testing of backlash stiffness in robot joint transmission as described in claim 1, characterized in that, The transmission backlash of the robot joint is determined based on the first and second reference boundary displacements, and the transmission stiffness before and after the commutation is determined based on the pre-commutation and post-commutation elastic reference relationships, generating robot joint transmission test results, including: Determine the displacement interval between the first reference boundary displacement and the second reference boundary displacement, and use the displacement interval as the transmission backlash of the robot joint; Based on the proportional relationship between the change in output torque and the change in transmission error displacement in the elastic reference relationship before and after commutation, the change in output torque corresponding to a unit change in transmission error displacement is determined, and the transmission stiffness before and after commutation is obtained. The transmission backlash, transmission stiffness before reversal, and transmission stiffness after reversal are correlated and recorded to obtain the test results of the robot joint transmission.

10. A joint testing system for backlash stiffness of robot joint transmission, characterized in that, The system for implementing the joint testing method for backlash stiffness of robot joint transmission as described in claim 1 includes: The data acquisition module is used to convert the input end angular displacement during the robot joint reversal process into the theoretical output end angular displacement, compare it with the actual output end angular displacement, obtain the transmission error displacement, and obtain the output end transmitted torque. The load identification module is used to identify the pre-commutation load area, commutation transition area, and post-commutation load area based on the continuous change of the output torque and transmission error displacement during the commutation process. The boundary determination module is used to determine the first bearing boundary at the position where the bearing area before the changeover transitions to the bearing area after the changeover, and to determine the second bearing boundary at the position where the bearing area after the changeover transitions to the bearing area after the changeover. The benchmark construction module is used to extract the pre-commutation elastic response interval and the post-commutation elastic response interval from the pre-commutation load area and the post-commutation load area, respectively, and to establish the pre-commutation elastic benchmark relationship and the post-commutation elastic benchmark relationship based on the pre-commutation elastic response interval and the post-commutation elastic response interval. The boundary conversion module is used to convert the transmission error displacement of the first bearing boundary and the second bearing boundary to a zero transmission torque state based on the elastic reference relationship before and after the conversion, so as to obtain the first reference boundary displacement and the second reference boundary displacement. The test output module is used to determine the transmission backlash of the robot joint based on the first reference boundary displacement and the second reference boundary displacement, and to determine the transmission stiffness before and after the commutation based on the elastic reference relationship before and after the commutation, and to generate the robot joint transmission test results.