Speed reducer performance comparison test method applied to humanoid robot

Abstract

The invention relates to the technical field of industrial robots, in particular to a speed reducer performance comparison test method applied to a humanoid robot. According to the method, motion information and task description information of an execution mechanism are obtained, and a multi-working-condition test sequence is constructed based on the motion information and the task description information; based on the multi-working-condition test sequence, the speed reducer to be tested is controlled to perform working condition replay, and multi-source operation data corresponding to the multi-working-condition test sequence is acquired in the working condition replay process; performing time reference alignment and measurement calibration processing on the multi-source operation data to obtain standardized test data; performing data preprocessing and feature calculation based on the standardized test data to obtain a unified index set; and performing normalization processing and statistical comparative analysis on the unified index set to obtain a comparative evaluation result of the speed reducer to be tested.

Description

Technical Field

[0001] This application relates to the field of industrial robot technology, and in particular to a method for comparative testing of speed reducer performance applied to humanoid robots. Background Technology

[0002] With the development of humanoid robot technology, the performance requirements for reducers in joint actuators are constantly increasing. Differences in reducer performance in terms of load-bearing capacity, transmission accuracy, and dynamic response directly affect the overall motion stability and control performance of the robot. Therefore, in the process of R&D selection, supplier comparison, and batch consistency verification, it is usually necessary to conduct performance comparison tests on reducers of different models or batches. In existing technologies, reducer performance testing often employs fixed test benches or integrated whole-machine debugging methods. Parameters such as speed, torque, efficiency, temperature rise, and vibration are measured under preset operating conditions, and test data is generated manually or through independent data acquisition software for analysis and comparison. However, this type of testing primarily focuses on single-performance verification, and the test conditions are usually relatively fixed, making it difficult to reflect the multi-state dynamic load characteristics exhibited by humanoid robot joints in actual operation. The lack of a unified basis for constructing operating conditions across different tests leads to weak comparability between test results. Furthermore, when comparing multiple reducers, judgments are often based on a single indicator or simple statistical results, lacking a unified indicator normalization and statistical comparison mechanism. This makes it difficult to comprehensively analyze multi-condition, multi-dimensional test results, resulting in poor accuracy. Therefore, improving the accuracy of reducer performance comparison testing for humanoid robots has become an urgent technical problem to be solved. Summary of the Invention

[0003] The main objective of this application is to provide a method for comparative testing of the performance of speed reducers applied to humanoid robots, aiming to solve the technical problem of how to improve the accuracy of comparative testing of speed reducer performance applied to humanoid robots.

[0004] To achieve the above objectives, this application provides a method for comparative testing of the performance of a reducer applied to a humanoid robot, the method comprising the following steps: Obtain motion information and task description information of the actuator, and construct a multi-condition test sequence based on the motion information and the task description information; Based on the multi-condition test sequence, the reducer under test is controlled to perform condition replay, and multi-source operating data corresponding to the multi-condition test sequence is acquired during the condition replay process. The multi-source operational data is time-referenced and calibrated to obtain standardized test data. Based on the standardized test data, data preprocessing and feature calculation are performed to obtain a unified set of indicators; The unified index set is normalized and statistically compared to obtain the comparative evaluation results of the speed reducer under test.

[0005] In one embodiment, the step of acquiring motion information and task description information of the actuator, and constructing a multi-condition test sequence based on the motion information and the task description information, includes: The motion trajectory data and load-related operation records of the actuator are acquired, and the motion trajectory data and load-related operation records are parsed to obtain task feature information; Based on the task feature information, the execution process is divided into working conditions, typical working condition segments corresponding to different operating states are extracted, and corresponding control mode parameters are configured for each typical working condition segment. The multi-condition test sequence is obtained by combining and arranging the typical operating condition segments and their corresponding control mode parameters.

[0006] In one embodiment, the step of performing time base alignment and measurement calibration on the multi-source operational data to obtain standardized test data includes: Based on a unified time reference, time identifier matching is performed on each data channel in the multi-source operational data, and data with different sampling frequencies are resampled to obtain time-synchronized data. Based on the time synchronization data, delay compensation and event location calibration are performed on each data channel to obtain aligned running data; The alignment running data is processed to correct measurement errors based on preset calibration parameters, thereby generating the standardized test data.

[0007] In one embodiment, the step of performing data preprocessing and feature calculation based on the standardized test data to obtain a unified indicator set includes: The standardized test data is subjected to anomaly identification and data smoothing, and the processing results are divided into intervals based on the working condition execution status to obtain a preprocessed dataset. Feature parameters are extracted based on the preprocessed dataset, and the feature parameters are processed according to the preset index definition strategy to obtain multidimensional index data. The multidimensional indicator data is mapped to an indicator structure and organized in a unified format to generate the unified indicator set.

[0008] In one embodiment, the step of identifying and smoothing abnormal data in the standardized test data, and dividing the processing results into intervals based on the operating condition execution status to obtain a preprocessed dataset includes: The standardized test data is subjected to data integrity detection and outlier identification processing. Data samples that deviate from the preset range of variation are identified and removed to obtain preliminary processed data. The preliminary data is smoothed to reduce the impact of fluctuations between adjacent data sampling points, thus obtaining continuous operating data. Based on the operating condition execution status information corresponding to the multi-condition test sequence, the continuous running data is divided into intervals and labeled with status to obtain the preprocessed dataset.

[0009] In one embodiment, the step of normalizing and statistically comparing the unified index set to obtain the comparative evaluation results of the speed reducer under test includes: Based on a preset index scaling strategy, the data of each index in the unified index set are subjected to dimension unification and interval mapping to obtain normalized index data. Based on the normalized index data, the working condition dimension is aggregated according to the preset index correlation relationship to obtain the index statistics data corresponding to different working conditions. Based on the statistical data of the aforementioned indicators, difference analysis and sorting are performed to generate comparative evaluation results for the speed reducer under test.

[0010] Furthermore, to achieve the above objectives, this application also proposes a performance comparison testing device for a reducer used in a humanoid robot, the reducer performance comparison testing device for a humanoid robot comprising: The multi-condition testing module is used to acquire motion information and task description information of the actuator, and to construct a multi-condition testing sequence based on the motion information and the task description information. The multi-source operation module is used to control the speed reducer under test to perform operation condition replay based on the multi-operation condition test sequence, and to acquire multi-source operation data corresponding to the multi-operation condition test sequence during the operation condition replay process. The standardized testing module is used to perform time base alignment and measurement calibration on the multi-source operating data to obtain standardized test data. The unified indicator module is used to perform data preprocessing and feature calculation based on the standardized test data to obtain a unified indicator set. The target module is used to normalize and statistically compare the unified index set to obtain the comparative evaluation results of the speed reducer under test.

[0011] Furthermore, to achieve the above objectives, this application also proposes a speed reducer performance comparison testing device for humanoid robots. The device includes: a memory, a processor, and a speed reducer performance comparison testing program for humanoid robots stored in the memory and executable on the processor. The speed reducer performance comparison testing program for humanoid robots is configured to implement the steps of the speed reducer performance comparison testing method for humanoid robots as described above.

[0012] In addition, to achieve the above objectives, this application also proposes a storage medium storing a speed reducer performance comparison test program for humanoid robots. When the speed reducer performance comparison test program for humanoid robots is executed by a processor, it implements the steps of the speed reducer performance comparison test method for humanoid robots as described above.

[0013] In addition, to achieve the above objectives, this application also proposes a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the speed reducer performance comparison test method applied to humanoid robots as described above.

[0014] This application acquires the motion information and task description information of the actuator, and constructs a multi-condition test sequence based on the motion information and task description information; controls the reducer under test to perform condition replay based on the multi-condition test sequence, and acquires multi-source operating data corresponding to the multi-condition test sequence during the condition replay process; performs time reference alignment and measurement calibration processing on the multi-source operating data to obtain standardized test data; performs data preprocessing and feature calculation based on the standardized test data to obtain a unified index set; and performs normalization processing and statistical comparative analysis on the unified index set to obtain the comparative evaluation results of the reducer under test. This application acquires the motion information and task description information of the actuator, and constructs a multi-condition test sequence based on this information. This ensures that the test conditions originate from the actual operating information of the actuator, allowing subsequent testing to be conducted based on a unified test condition sequence. This reduces data deviations caused by differences in test conditions and improves the comparability of test results between different reducers. By controlling the reducer under test to replay its test conditions based on the multi-condition test sequence and acquiring corresponding multi-source operating data during the replay process, this application ensures that all operating data are generated synchronously under the same operating conditions. This guarantees a clear correspondence between test data and operating conditions, reduces measurement errors caused by inconsistencies in the testing process, and improves the reliability of the data acquisition phase. Furthermore, by performing time reference alignment and measurement calibration on the multi-source operating data, the data from different sources is accurately reflected in the test conditions. Operational data from the same data channel are correlated under a unified time reference, and measurement deviations are corrected to reduce the impact of time misalignment and measurement errors between multi-source data on the analysis results, thereby improving data consistency and usability. By performing data preprocessing and feature calculation based on standardized test data, a unified set of indicators is obtained, transforming data from different sources into evaluation indicators in a unified form. This allows subsequent analysis to be conducted based on a consistent data expression method, thereby reducing the interference of data noise and outlier data on the evaluation results. By normalizing and statistically comparing the unified set of indicators, a comprehensive comparison of each indicator is performed under a unified scale, enabling indicators of different dimensions to participate in the same evaluation process. This reduces the impact of differences in indicator dimensions on the comparison results, thereby improving the objectivity and accuracy of the performance comparison evaluation results of the reducer. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating the first embodiment of the speed reducer performance comparison test method applied to humanoid robots according to this application; Figure 2 This is a schematic diagram of a sub-process in the second embodiment of the speed reducer performance comparison test method applied to humanoid robots in this application; Figure 3 This is a schematic diagram of a sub-process in the third embodiment of the speed reducer performance comparison test method applied to humanoid robots in this application; Figure 4This is a schematic diagram of the module structure of a speed reducer performance comparison test device applied to a humanoid robot in one embodiment of this application; Figure 5 This is a schematic diagram of the equipment structure of the hardware operating environment involved in the performance comparison test method of the reducer applied to a humanoid robot in one embodiment of this application.

[0016] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0017] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of this application.

[0018] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0019] It should be noted that with the development of humanoid robot technology, the performance requirements of reducers for joint actuators are constantly increasing. Differences in reducer performance in terms of load-bearing capacity, transmission accuracy, and dynamic response will directly affect the overall motion stability and control performance of the robot. Therefore, in the process of R&D selection, supplier comparison, and batch consistency verification, it is usually necessary to conduct performance comparison tests on reducers of different models or batches. In existing technologies, reducer performance testing often employs fixed test benches or integrated whole-machine debugging methods. Parameters such as speed, torque, efficiency, temperature rise, and vibration are measured under preset operating conditions, and test data is generated manually or through independent data acquisition software for analysis and comparison. However, this type of testing primarily focuses on single-performance verification, and the test conditions are usually relatively fixed, making it difficult to reflect the multi-state dynamic load characteristics exhibited by humanoid robot joints in actual operation. The lack of a unified basis for constructing operating conditions across different tests leads to weak comparability between test results. Furthermore, when comparing multiple reducers, judgments are often based on a single indicator or simple statistical results, lacking a unified indicator normalization and statistical comparison mechanism. This makes it difficult to comprehensively analyze multi-condition, multi-dimensional test results, resulting in poor accuracy. Therefore, improving the accuracy of reducer performance comparison testing for humanoid robots has become an urgent technical problem to be solved.

[0020] The main solution of this application is as follows: Obtain the motion information and task description information of the actuator, and construct a multi-condition test sequence based on the motion information and task description information; control the reducer under test to perform condition replay based on the multi-condition test sequence, and acquire multi-source operating data corresponding to the multi-condition test sequence during the condition replay process; perform time reference alignment and measurement calibration processing on the multi-source operating data to obtain standardized test data; perform data preprocessing and feature calculation based on the standardized test data to obtain a unified index set; perform normalization processing and statistical comparative analysis on the unified index set to obtain the comparative evaluation results of the reducer under test.

[0021] This application acquires the motion information and task description information of the actuator, and constructs a multi-condition test sequence based on this information. This ensures that the test conditions originate from the actual operating information of the actuator, allowing subsequent testing to be conducted based on a unified test condition sequence. This reduces data deviations caused by differences in test conditions and improves the comparability of test results between different reducers. By controlling the reducer under test to replay its test conditions based on the multi-condition test sequence and acquiring corresponding multi-source operating data during the replay process, this application ensures that all operating data are generated synchronously under the same operating conditions. This guarantees a clear correspondence between test data and operating conditions, reduces measurement errors caused by inconsistencies in the testing process, and improves the reliability of the data acquisition phase. Furthermore, by performing time reference alignment and measurement calibration on the multi-source operating data, the data from different sources is accurately reflected in the test conditions. Operational data from the same data channel are correlated under a unified time reference, and measurement deviations are corrected to reduce the impact of time misalignment and measurement errors between multi-source data on the analysis results, thereby improving data consistency and usability. By performing data preprocessing and feature calculation based on standardized test data, a unified set of indicators is obtained, transforming data from different sources into evaluation indicators in a unified form. This allows subsequent analysis to be conducted based on a consistent data expression method, thereby reducing the interference of data noise and outlier data on the evaluation results. By normalizing and statistically comparing the unified set of indicators, a comprehensive comparison of each indicator is performed under a unified scale, enabling indicators of different dimensions to participate in the same evaluation process. This reduces the impact of differences in indicator dimensions on the comparison results, thereby improving the objectivity and accuracy of the performance comparison evaluation results of the reducer.

[0022] It should be noted that the execution subject of the method in this embodiment can be a computing service device with data processing, network communication, and program execution functions, or it can be the aforementioned speed reducer performance comparison testing device applied to humanoid robots with the same or similar functions. This embodiment and the following embodiments will be described using a speed reducer performance comparison testing device applied to humanoid robots as an example.

[0023] Based on this, a first embodiment of the method for comparing the performance of reducers in humanoid robots, as described in this application, is proposed. Please refer to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the method for comparing the performance of a speed reducer applied to a humanoid robot, as described in this application.

[0024] In this embodiment, the method includes the following steps: S1: Obtain motion information and task description information of the actuator, and construct a multi-condition test sequence based on the motion information and the task description information; It should be noted that the actuator refers to the joint drive unit in a humanoid robot used to complete motion output. Motion information refers to the data set reflecting the motion state of the actuator. Task description information refers to information used to describe the actuator's task execution method or operating scenario, including task type, action mode, operation phase division, and load change characteristics. Multi-condition test sequence refers to a set of test conditions constructed based on motion information and task description information and arranged in a predetermined order.

[0025] Specifically, firstly, motion information generated by the actuator during actual or simulated operation is acquired, along with corresponding task description information. The motion information can originate from data recorded by the robot control system, prototype operation logs, or motion planning results. The task description information characterizes the phase division and operating mode of the actions performed by the actuator. Subsequently, the motion information is parsed and processed, breaking down the continuous motion process according to changes in motion state, and extracting data segments that characterize the features of different operating stages, thus establishing a correspondence between the motion information and the task description information.

[0026] Furthermore, based on this, the motion stages are categorized and organized according to the task description information. Motion segments with similar operational characteristics are abstracted into multiple test conditions, and each test condition is sequentially arranged and parameter-associated to form a continuous test process structure between different conditions. By combining and organizing the test conditions, a multi-condition test sequence that reflects the characteristics of the actuator's operation process is constructed for subsequent replay of the test conditions of the reducer under test.

[0027] This step constructs a multi-condition test sequence by combining the motion information and task description information of the actuator. This ensures that the test conditions originate from the actual operation of the actuator. Furthermore, the motion stages are uniformly divided and organized using task description information. This allows the tests of different reducers to be conducted on a consistent and operationally related basis, reducing the impact of artificially set differences in test conditions on the test results. Consequently, it improves the consistency and comparability of data sources in subsequent performance comparison tests, providing a foundation for improving the accuracy of reducer performance comparison tests.

[0028] S2: Based on the multi-condition test sequence, control the reducer under test to perform condition replay, and acquire multi-source operating data corresponding to the multi-condition test sequence during the condition replay process; S3: Perform time base alignment and measurement calibration on the multi-source operating data to obtain standardized test data; It should be noted that "operational condition replay" refers to using the control system to drive the test device according to the operating state parameters defined in the multi-operational condition test sequence, causing the reducer under test to operate according to a predetermined motion law and load change process. "Multi-source operating data" refers to the set of operating data obtained from different acquisition sources during the operational condition replay process. "Time reference alignment" refers to mapping operating data acquired from different data sources to the same time reference system, establishing a correspondence between the data in the time dimension. "Measurement calibration processing" refers to correcting systematic errors or measurement deviations in the acquired data according to preset calibration rules, ensuring that data from different acquisition channels have a consistent measurement reference.

[0029] Specifically, firstly, based on the established multi-condition test sequence, the test control system sends control commands to the drive unit and loading unit, causing the reducer under test to sequentially execute the corresponding motion state and load change process according to the multi-condition test sequence. During the execution of the conditions, each acquisition unit synchronously records the reducer's operating status information and control feedback information, acquires multi-source operating data reflecting the reducer's dynamic behavior from multiple data sources, and associates and identifies the acquired data with the currently executed condition stage to form a data record corresponding to the multi-condition test sequence.

[0030] Furthermore, the acquired multi-source operational data is then processed uniformly. First, based on a unified time reference, the data from different sampling channels are time-mapped and synchronized, establishing a correspondence between data from different sources on the same time axis. Then, combined with preset calibration information, each data channel is measured and calibrated to correct any offsets or inconsistencies that occur during the acquisition process, thereby forming a data set with consistent time and a unified measurement benchmark, resulting in standardized test data.

[0031] This step involves replaying the operating conditions of the reducer under test using a multi-condition test sequence, generating multi-source operating data under unified operating conditions. Furthermore, time reference alignment and measurement calibration processes eliminate time deviations and measurement differences between different data sources, thereby establishing a consistent time correlation and measurement benchmark among the various operating data. This reduces the impact of uncertainties in the data acquisition stage on subsequent analysis results, thereby improving the consistency and reliability of data used for performance comparison analysis and providing a data foundation for improving the accuracy of reducer performance comparison testing.

[0032] S4: Based on the standardized test data, perform data preprocessing and feature calculation to obtain a unified set of indicators; S5: Normalize and statistically compare the unified index set to obtain the comparative evaluation results of the speed reducer under test; It should be noted that data preprocessing refers to the process of organizing and optimizing data before index calculation. Feature calculation refers to extracting data features that characterize the operating status of the reducer based on the preprocessed data. A unified index set refers to a set of evaluation indicators constructed according to unified index definition rules. Normalization refers to converting indicators with different dimensions or numerical ranges to a unified scale, enabling comparison of indicators within the same evaluation system. Statistical comparative analysis refers to performing statistical calculations and correlation analysis on the results of multiple indicators to establish evaluation relationships between different reducers under test. Comparative evaluation results refer to the performance comparison conclusions of reducers obtained based on the unified index system, used to characterize the differences between different tested objects.

[0033] Specifically, firstly, a data preprocessing process is carried out based on standardized test data. This involves identifying and processing outlier samples, and then dividing the continuous data into intervals based on the operational stage information corresponding to the multi-condition test sequences, enabling the data to be organized according to different operating conditions. After data processing, data features reflecting changes in the reducer's operating state are extracted from each operating interval. These extracted features are then calculated and processed according to preset indicator definition rules, transforming the multi-source operating data into indicator data with a unified expression form, thus forming a unified indicator set.

[0034] Furthermore, the unified set of indicators is then normalized to unify the differences in numerical range and dimensions of different indicators, ensuring that all indicators are on a comparable, unified scale. Based on this, statistical comparative analysis is performed on the normalized indicators. By comprehensively calculating and sorting data from different operating conditions and indicator dimensions, comparative evaluation results are generated to characterize the performance differences of the speed reducer under test, thus completing the performance comparison analysis process.

[0035] This step first converts standardized test data into a unified set of indicators, allowing data from different sources and of different types to be expressed in a consistent indicator format. Then, normalization is used to eliminate differences in indicator dimensions and numerical ranges. Finally, statistical comparative analysis is used to form a unified evaluation basis, thereby avoiding deviations in evaluation results caused by single indicators or differences in raw data. This ensures that the performance comparison of reducers is based on a unified indicator system and unified calculation rules, thereby improving the consistency and objectivity of performance comparison results and enhancing the accuracy of performance comparison tests of reducers applied to humanoid robots.

[0036] This embodiment acquires the motion information and task description information of the actuator, and constructs a multi-condition test sequence based on the motion information and task description information; it controls the reducer under test to perform condition replay based on the multi-condition test sequence, and acquires multi-source operating data corresponding to the multi-condition test sequence during the condition replay process; it performs time reference alignment and measurement calibration processing on the multi-source operating data to obtain standardized test data; it performs data preprocessing and feature calculation based on the standardized test data to obtain a unified index set; it performs normalization processing and statistical comparative analysis on the unified index set to obtain the comparative evaluation results of the reducer under test. This embodiment acquires the motion information and task description information of the actuator, and constructs a multi-condition test sequence based on the motion information and task description information. This ensures that the test conditions originate from the actual operating information of the actuator, allowing subsequent testing to be conducted based on a unified test condition sequence. This reduces data deviations caused by differences in test conditions and improves the comparability of test results between different reducers. By controlling the reducer under test to replay the test conditions based on the multi-condition test sequence and acquiring corresponding multi-source operating data during the replay process, all operating data are generated synchronously under the same operating conditions. This ensures a clear correspondence between test data and operating condition status, reduces measurement errors caused by inconsistencies in the testing process, and improves the reliability of the data acquisition stage. Furthermore, by performing time reference alignment and measurement calibration on the multi-source operating data, the data from different sources is accurately reflected in the test conditions. Operational data from the same data channel are correlated under a unified time reference, and measurement deviations are corrected to reduce the impact of time misalignment and measurement errors between multi-source data on the analysis results, thereby improving data consistency and usability. By performing data preprocessing and feature calculation based on standardized test data, a unified set of indicators is obtained, transforming data from different sources into evaluation indicators in a unified form. This allows subsequent analysis to be conducted based on a consistent data expression method, thereby reducing the interference of data noise and outlier data on the evaluation results. By normalizing and statistically comparing the unified set of indicators, a comprehensive comparison of each indicator is performed under a unified scale, enabling indicators of different dimensions to participate in the same evaluation process. This reduces the impact of differences in indicator dimensions on the comparison results, thereby improving the objectivity and accuracy of the performance comparison evaluation results of the reducer.

[0037] Based on the first embodiment described above, a second embodiment of the method for comparing the performance of a reducer in a humanoid robot, as applied in this application, is proposed. Please refer to... Figure 2 , Figure 2 This is a schematic diagram of a sub-process in the second embodiment of the speed reducer performance comparison test method applied to humanoid robots in this application.

[0038] like Figure 2 As shown, in this embodiment, step S1 includes: S11: Obtain the motion trajectory data and load-related operation records of the actuator, and parse the motion trajectory data and load-related operation records to obtain task feature information; S12: Based on the task feature information, the execution process is divided into working conditions, typical working condition segments corresponding to different operating states are extracted, and corresponding control mode parameters are configured for each typical working condition segment. S13: Based on the typical working condition segments and the corresponding control mode parameters, the multi-working condition test sequence is obtained by combining and arranging them.

[0039] It should be noted that motion trajectory data refers to motion information data describing the changes of the actuator over time. Load-related operation records refer to data records related to the force or working state during the operation of the actuator. Task characteristic information refers to the operational characteristic description information obtained by parsing the motion trajectory data and load-related operation records. Working condition division refers to dividing the continuous execution process into multiple stages with different operating characteristics based on task characteristic information. Typical working condition segments refer to representative operating stage data extracted from the working condition division results. Control mode parameters refer to the set of parameters used to describe the control methods and operating control conditions during the test.

[0040] Specifically, firstly, motion trajectory data and corresponding load-related operation records generated by the actuator during actual or simulated operation are acquired, and the two types of data are correlated and analyzed. By performing state identification and feature extraction on the continuous operation data, the actuator's operation process is converted into task feature information that reflects the characteristics of different task stages, establishing a correspondence between motion behavior and load changes, thus forming a structured description of the execution process. Based on this, the execution process is divided into working conditions according to the task feature information. Stages with similar operating characteristics are extracted into multiple typical working condition segments, and corresponding control mode parameters are configured for different typical working condition segments to limit the execution mode of the corresponding working condition during the test. Subsequently, each typical working condition segment and its control mode parameters are combined and arranged according to preset logic, so that different operating states form a continuous test process in a determined order, thereby constructing a multi-working-condition test sequence for subsequent test execution.

[0041] This step analyzes the motion trajectory data and load-related operating records of the actuator to extract task feature information and completes the division of working conditions and construction of typical working condition segments accordingly. This ensures that the test working conditions originate from the actual operation process of the actuator and are combined and arranged through unified control mode parameters. As a result, the multi-working-condition test sequence can systematically reflect the changes in the operating status of the actuator, reduce the differences caused by manually setting test working conditions, and enable different reducers under test to be tested under consistent working conditions. This improves the comparability and accuracy of performance comparison test results.

[0042] Based on the first embodiment described above, in this embodiment, step S3 includes: S31: Based on a unified time reference, time identifier matching is performed on each data channel in the multi-source running data, and data with different sampling frequencies are resampled to obtain time synchronization data; S32: Based on the time synchronization data, perform delay compensation and event location calibration on each data channel to obtain aligned running data; S33: Based on preset calibration parameters, the alignment running data is subjected to measurement error correction processing to generate the standardized test data.

[0043] It should be noted that: A unified time reference refers to a time standard used to establish a common time base for different data channels, enabling data from each channel to correspond on the same time scale. Time label matching refers to establishing a correspondence between the time labels of data from different channels based on the unified time reference, ensuring consistency in the time dimension for data from different sources. Resampling processing refers to a data processing method that reorganizes data sampling points for data with different sampling frequencies to meet the requirements of a unified time interval. Time-synchronized data refers to a data set with a consistent time scale formed after time label matching and resampling processing. Delay compensation refers to correcting time lags generated during data acquisition or transmission, ensuring consistency across data channels at the actual time of event occurrence. Event location calibration refers to marking key state changes during operation, enabling different data channels to correspond to the same operational event. Preset calibration parameters are pre-set parameter information used to correct measurement deviations.

[0044] Specifically, firstly, after acquiring multi-source operational data, a unified time reference is used as the benchmark to match the time stamps in each data channel, enabling data from different acquisition systems to be mapped to the same time axis. Since the sampling frequencies of different data channels may differ, resampling is performed on data with inconsistent time intervals, arranging the data from each channel according to a unified time step, thus forming a time-consistent data set and obtaining time-synchronized data.

[0045] Furthermore, based on the time-synchronized data, delay compensation is applied to each data channel to correct time offsets caused by differences in acquisition response or data transmission processes. Simultaneously, key state changes during operation are calibrated to establish a correspondence between data from different channels under the same operational event. After achieving time-level unification, preset calibration parameters are used to correct measurement errors in each data channel, uniformly adjusting deviations generated during acquisition, ultimately generating standardized test data with consistent time and a unified measurement benchmark.

[0046] This step achieves time signature matching and resampling of multi-source operating data through a unified time reference, establishing a consistent relationship between different data channels in the time dimension. Furthermore, it eliminates time offsets during data acquisition through delay compensation and event location calibration, and corrects measurement errors by combining preset calibration parameters. This simultaneously unifies the time and measurement benchmarks of the data, reducing the impact of inconsistencies between multi-source data on subsequent analysis. It ensures that subsequent index calculations are based on consistent data, thereby improving the accuracy and reliability of the speed reducer performance comparison test results.

[0047] This embodiment acquires the motion information and task description information of the actuator, and constructs a multi-condition test sequence based on the motion information and task description information; it controls the reducer under test to perform condition replay based on the multi-condition test sequence, and acquires multi-source operating data corresponding to the multi-condition test sequence during the condition replay process; it performs time reference alignment and measurement calibration processing on the multi-source operating data to obtain standardized test data; it performs data preprocessing and feature calculation based on the standardized test data to obtain a unified index set; it performs normalization processing and statistical comparative analysis on the unified index set to obtain the comparative evaluation results of the reducer under test. This embodiment acquires the motion information and task description information of the actuator, and constructs a multi-condition test sequence based on the motion information and task description information. This ensures that the test conditions originate from the actual operating information of the actuator, allowing subsequent testing to be conducted based on a unified test condition sequence. This reduces data deviations caused by differences in test conditions and improves the comparability of test results between different reducers. By controlling the reducer under test to replay the test conditions based on the multi-condition test sequence and acquiring corresponding multi-source operating data during the replay process, all operating data are generated synchronously under the same operating conditions. This ensures a clear correspondence between test data and operating condition status, reduces measurement errors caused by inconsistencies in the testing process, and improves the reliability of the data acquisition stage. Furthermore, by performing time reference alignment and measurement calibration on the multi-source operating data, the data from different sources is accurately reflected in the test conditions. Operational data from the same data channel are correlated under a unified time reference, and measurement deviations are corrected to reduce the impact of time misalignment and measurement errors between multi-source data on the analysis results, thereby improving data consistency and usability. By performing data preprocessing and feature calculation based on standardized test data, a unified set of indicators is obtained, transforming data from different sources into evaluation indicators in a unified form. This allows subsequent analysis to be conducted based on a consistent data expression method, thereby reducing the interference of data noise and outlier data on the evaluation results. By normalizing and statistically comparing the unified set of indicators, a comprehensive comparison of each indicator is performed under a unified scale, enabling indicators of different dimensions to participate in the same evaluation process. This reduces the impact of differences in indicator dimensions on the comparison results, thereby improving the objectivity and accuracy of the performance comparison evaluation results of the reducer.

[0048] Based on the second embodiment described above, a third embodiment of the method for comparing the performance of reducers in humanoid robots, as applied in this application, is proposed. Please refer to... Figure 3 , Figure 3 This is a schematic diagram of a sub-process in the third embodiment of the speed reducer performance comparison test method applied to humanoid robots in this application.

[0049] In this embodiment, step S4 includes: S41: Perform abnormal data identification and data smoothing on the standardized test data, and divide the processing results into intervals based on the working condition execution status to obtain a preprocessed dataset; S42: Extract feature parameters based on the preprocessed dataset, and process the feature parameters according to the preset index definition strategy to obtain multidimensional index data; S43: Map the multidimensional indicator data to an indicator structure and organize it in a unified format to generate the unified indicator set.

[0050] It should be noted that abnormal data identification refers to the process of detecting and marking data samples in standardized test data that deviate from normal trends or do not conform to operating conditions. Data smoothing refers to the stabilization process performed on continuously sampled data. Feature parameters refer to data feature results extracted from the preprocessed dataset that can characterize the operating behavior or state changes of the reducer. Pre-defined index definition strategy refers to the pre-set standards used to specify the calculation methods and organizational rules of the indicators, which are used to convert feature parameters into unified evaluation indicators. Multidimensional index data refers to a dataset composed of multiple different types of indicators. Index structure mapping refers to the process of corresponding and classifying indicators from different sources or of different types according to a unified index system.

[0051] Specifically, firstly, abnormal data identification and processing are performed on the standardized test data. By judging the consistency between data change trends and operational status, abnormal sampling points are identified and marked or processed to reduce the impact of abnormal data on subsequent analysis. After anomaly handling, data smoothing is performed to ensure that continuously sampled data maintains stable changes over time. Subsequently, based on the operational status information corresponding to the multi-condition test sequence, the processed data is divided into intervals, organizing the data according to different test conditions to form a preprocessed dataset corresponding to each operational condition stage.

[0052] Furthermore, based on this, feature parameters reflecting the operating status of the reducer are extracted from the preprocessed dataset. These extracted feature parameters are then uniformly processed according to a predefined index definition strategy, converting data from different sources into index data with a consistent expression, resulting in multidimensional index data. Further, index structure mapping and unified format organization are performed on the multidimensional index data, classifying and arranging each index according to a unified index system to ensure consistency in structure and expression, thereby generating a unified index set.

[0053] This step first improves data stability through anomaly identification and data smoothing, and divides the data into intervals based on the operating conditions to establish a clear correspondence between the data and specific test conditions. Then, through feature parameter extraction and preset index definition strategies, the original operating data is transformed into multi-dimensional index data in a unified expression form. Furthermore, a unified index set is formed through index structure mapping, so that data from different sources and under different operating conditions can be expressed and compared in a consistent index system. This reduces the impact of data noise, expression differences, and structural inconsistencies on the analysis results, thereby improving the objectivity and accuracy of the speed reducer performance comparison test results.

[0054] Based on the second embodiment described above, in this embodiment, step S41 includes: S411: Perform data integrity detection and outlier identification on the standardized test data, identify data samples that deviate from the preset range of variation, and remove the identification results to obtain preliminary processed data; S412: Perform data smoothing on the preliminary processed data to reduce the impact of fluctuations between adjacent data sampling points and obtain continuous running data; S413: Based on the operating condition execution status information corresponding to the multi-condition test sequence, the continuous running data is divided into intervals and labeled with status to obtain the preprocessed dataset.

[0055] It should be noted that data integrity detection refers to checking the continuity and validity of data records to determine whether there is missing data, abnormal interruptions, or invalid sampling. The preset variation range refers to a pre-defined interval of data variation based on test conditions or historical operating data. Interval division and status labeling refers to the process of dividing continuous data into different operating stages and attaching corresponding status labels based on the operating condition execution status information. Operating condition execution status information refers to status identification information derived from multi-condition test sequences, used to indicate the test condition stage corresponding to the current data.

[0056] Specifically, firstly, the standardized test data undergoes a data integrity check. By examining the continuity of data records and the validity of sampling, potential data gaps, anomalous jumps, or data samples that do not conform to the operating state are identified. Subsequently, anomaly detection processing is performed on each data channel according to a preset range of variation. Data that significantly deviates from the normal operating trend is identified and removed, resulting in preliminary processed data containing only valid sampling information, thereby reducing the interference of anomalous data on subsequent analysis.

[0057] Furthermore, after obtaining the preliminary processed data, smoothing processing is performed on the data to make the changes between adjacent sampling points more continuous and stable, reducing the fluctuations caused by sampling noise or instantaneous disturbances, thus obtaining continuous operating data. Subsequently, combined with the operating condition execution status information recorded in the multi-condition test sequence, the continuous operating data is divided into intervals, and the data is organized according to different test condition stages. Corresponding status labels are added to each data interval, thereby forming a preprocessed dataset corresponding to the test conditions.

[0058] This step eliminates data samples that do not conform to the operating rules through data integrity detection and outlier identification, ensuring the validity and consistency of the input data. It also reduces the impact of sampling fluctuations on data trends through data smoothing. Furthermore, it divides the data into intervals and labels the status based on the operating conditions, establishing a clear correspondence between the data and the specific test conditions. This allows subsequent feature calculations to be based on stable data with operating condition semantics, reducing the deviation of analysis results caused by abnormal data and random fluctuations, thereby improving the accuracy and reliability of the speed reducer performance comparison test.

[0059] In this embodiment, step S5 includes: S51: Based on a preset index scaling strategy, the data of each index in the unified index set are subjected to dimension unification and interval mapping to obtain normalized index data. S52: Based on the normalized index data, perform working condition dimension aggregation processing according to the preset index correlation relationship to obtain index statistics data corresponding to different working conditions. S53: Based on the statistical data of the indicators, perform difference analysis and sorting to generate a comparative evaluation result of the speed reducer under test.

[0060] It should be noted that: Preset indicator scaling strategy refers to a pre-defined set of rules used to standardize the expression range and comparison methods of different indicator data. Dimensional unification refers to the process of converting indicators with different physical meanings or numerical units into comparable data expressions. Interval mapping refers to the process of converting indicator data to a unified numerical interval according to preset rules. Normalized indicator data refers to the set of indicator data after dimensional unification and interval mapping. Preset indicator correlation refers to the pre-defined organizational relationship between different indicators in the evaluation system. Operating condition dimension aggregation refers to classifying, summarizing, and organizing indicator data according to test operating conditions, enabling the indicators to reflect the overall performance under different operating conditions. Indicator statistical data refers to the statistical results data formed after operating condition dimension aggregation. Comparative evaluation results refer to the performance comparison conclusions of the reducer based on indicator difference analysis and ranking.

[0061] Specifically, firstly, for each indicator data in the unified indicator set, dimensional unification processing is performed according to a preset indicator scaling strategy, converting indicators with different units or numerical ranges into a unified expression form. Then, interval mapping is applied to the processed indicator data, allowing each indicator to be represented within a unified numerical range, thereby forming normalized indicator data and making different indicators comparable under the same evaluation scale.

[0062] Furthermore, after obtaining the normalized index data, based on the preset index correlation relationships, the indicators are aggregated according to the working condition dimensions corresponding to the multi-working-condition test sequence. The index data under the same working condition is summarized and organized to obtain the index statistics for different working conditions. Subsequently, difference analysis is performed based on the index statistics to compare the index performance of different speed reducers under test under various working-condition dimensions. The comprehensive evaluation order is formed through sorting, and finally, the comparative evaluation results of the speed reducers under test are generated.

[0063] This step first uses dimensional unification and interval mapping to express different types of indicators on a unified scale, eliminating the impact of differences in indicator units and numerical ranges on the comparison results. Furthermore, it establishes a correspondence between indicators and test conditions through working condition-based aggregation, enabling the evaluation results to reflect the comprehensive performance under different operating conditions. Subsequently, it uses difference analysis and ranking to form a unified evaluation basis, thus avoiding evaluation biases caused by differences in single indicators or scales. This ensures that the performance comparison of reducers is based on a consistent indicator system and unified analysis rules, thereby improving the objectivity and accuracy of the performance comparison test results of reducers applied to humanoid robots.

[0064] This embodiment acquires the motion information and task description information of the actuator, and constructs a multi-condition test sequence based on the motion information and task description information; it controls the reducer under test to perform condition replay based on the multi-condition test sequence, and acquires multi-source operating data corresponding to the multi-condition test sequence during the condition replay process; it performs time reference alignment and measurement calibration processing on the multi-source operating data to obtain standardized test data; it performs data preprocessing and feature calculation based on the standardized test data to obtain a unified index set; it performs normalization processing and statistical comparative analysis on the unified index set to obtain the comparative evaluation results of the reducer under test. This embodiment acquires the motion information and task description information of the actuator, and constructs a multi-condition test sequence based on the motion information and task description information. This ensures that the test conditions originate from the actual operating information of the actuator, allowing subsequent testing to be conducted based on a unified test condition sequence. This reduces data deviations caused by differences in test conditions and improves the comparability of test results between different reducers. By controlling the reducer under test to replay the test conditions based on the multi-condition test sequence and acquiring corresponding multi-source operating data during the replay process, all operating data are generated synchronously under the same operating conditions. This ensures a clear correspondence between test data and operating condition status, reduces measurement errors caused by inconsistencies in the testing process, and improves the reliability of the data acquisition stage. Furthermore, by performing time reference alignment and measurement calibration on the multi-source operating data, the data from different sources is accurately reflected in the test conditions. Operational data from the same data channel are correlated under a unified time reference, and measurement deviations are corrected to reduce the impact of time misalignment and measurement errors between multi-source data on the analysis results, thereby improving data consistency and usability. By performing data preprocessing and feature calculation based on standardized test data, a unified set of indicators is obtained, transforming data from different sources into evaluation indicators in a unified form. This allows subsequent analysis to be conducted based on a consistent data expression method, thereby reducing the interference of data noise and outlier data on the evaluation results. By normalizing and statistically comparing the unified set of indicators, a comprehensive comparison of each indicator is performed under a unified scale, enabling indicators of different dimensions to participate in the same evaluation process. This reduces the impact of differences in indicator dimensions on the comparison results, thereby improving the objectivity and accuracy of the performance comparison evaluation results of the reducer.

[0065] In one embodiment, after constructing a multi-condition test sequence, instead of directly assigning equal weights to each condition in subsequent analysis, a condition contribution evaluation mechanism is introduced. Specifically, the frequency and duration of each condition are statistically analyzed from the historical operating data of the actuator, and the corresponding operating percentage parameter for each condition is calculated based on the statistical results. Subsequently, according to the operating percentage parameter, corresponding condition weight identifiers are assigned to different conditions in the multi-condition test sequence, and the condition weight identifiers are associated with and stored with the condition execution status information. In the subsequent indicator statistical process, the data generated by each condition participates in the statistical calculation according to its corresponding weight, so that the test results can reflect the condition distribution characteristics of the actuator in actual operation.

[0066] In one embodiment, after acquiring multi-source operational data during the replay of operating conditions, dynamic consistency verification is further performed on the time-synchronized data. Specifically, after time base alignment is completed, correlation detection is performed on the changing trends between different data channels. When inconsistencies are detected among multiple channels during the same operating condition phase, the corresponding data segments are marked for consistency, and anomaly correlation tags are recorded. In the subsequent data preprocessing stage, data segments with anomaly correlation tags are either downweighted or independently identified and stored to avoid local acquisition anomalies affecting the overall analysis. This mechanism improves data reliability through cross-channel correlation judgment.

[0067] In one embodiment, when extracting feature parameters based on a preprocessed dataset, not only are data features calculated within a single interval, but also contextual information of the operating phase is introduced. Specifically, during feature calculation, the preceding and subsequent operating conditions of the current operating interval are used as auxiliary reference information, and phase transition identifiers are added to the feature parameters, enabling the indicators to reflect the continuous characteristics during the change of operating state. For example, a transition state identifier parameter is generated during the operating condition switching phase to distinguish the data features of the stable operating phase from the state change phase. This processing method enables the indicators to describe the dynamic operating process rather than just reflecting the static state.

[0068] In one embodiment, after generating the comparative evaluation results of the reducer under test, a stability verification process is further performed. Specifically, the order of the multi-condition test sequences is rearranged by perturbation, statistical comparative analysis is repeatedly performed, and the consistency of the ranking among the multiple analysis results is compared. When the ranking change exceeds a preset range, stability identification information is added to the evaluation results. The final output comparative evaluation results include performance ranking information and the corresponding stability identification. This method is used to assist in judging the sensitivity of the evaluation results to changes in the combination of operating conditions.

[0069] This application also provides a device for comparing the performance of a speed reducer used in humanoid robots. Please refer to... Figure 4 , Figure 4This is a schematic diagram of the module structure of a speed reducer performance comparison testing device applied to a humanoid robot according to an embodiment of this application. The speed reducer performance comparison testing device applied to a humanoid robot includes: The multi-condition test module 401 is used to acquire motion information and task description information of the actuator, and construct a multi-condition test sequence based on the motion information and the task description information; The multi-source operation module 402 is used to control the speed reducer under test to perform operation condition replay based on the multi-operation condition test sequence, and to acquire multi-source operation data corresponding to the multi-operation condition test sequence during the operation condition replay process. The standardized test module 403 is used to perform time reference alignment and measurement calibration on the multi-source operating data to obtain standardized test data. The unified indicator module 404 is used to perform data preprocessing and feature calculation based on the standardized test data to obtain a unified indicator set. The target module 405 is used to perform normalization processing and statistical comparative analysis on the unified index set to obtain the comparative evaluation results of the speed reducer under test.

[0070] The performance comparison testing device for reducers applied to humanoid robots provided in this application embodiment employs the performance comparison testing method for reducers applied to humanoid robots described in the above embodiments, and can solve the technical problem of how to improve the accuracy of performance comparison testing of reducers applied to humanoid robots. Compared with the prior art, the beneficial effects of the performance comparison testing device for reducers applied to humanoid robots provided in this application embodiment are the same as the beneficial effects of the performance comparison testing method for reducers applied to humanoid robots provided in the above embodiments, and other technical features in the performance comparison testing device for reducers applied to humanoid robots are the same as the features disclosed in the methods of the above embodiments, and will not be repeated here.

[0071] This application provides a speed reducer performance comparison testing device for humanoid robots. The speed reducer performance comparison testing device for humanoid robots includes: at least one processor; and a memory communicatively connected to at least one processor; wherein the memory stores instructions executable by at least one processor, and the instructions are executed by at least one processor to enable at least one processor to perform the speed reducer performance comparison testing method for humanoid robots in the above embodiments.

[0072] The following is for reference. Figure 5 , Figure 5 This is a schematic diagram of the hardware operating environment of a speed reducer performance comparison test method applied to a humanoid robot in one embodiment of this application. It shows a schematic diagram of the structure of a speed reducer performance comparison test device suitable for implementing the embodiments of this application. Figure 5The speed reducer performance comparison test equipment shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0073] like Figure 5 As shown, the performance comparison testing equipment for reducers used in humanoid robots may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to programs stored in read-only memory (ROM) 1002 or programs loaded from storage device 1003 into random access memory (RAM) 1004. RAM 1004 also stores various programs and data required for the operation of the performance comparison testing equipment for reducers used in humanoid robots. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via bus 1005. Input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the gearbox performance comparison testing equipment used in humanoid robots to exchange data wirelessly or via wired communication with other devices. Although the figure shows a gearbox performance comparison testing equipment for humanoid robots with various systems, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems can be implemented alternatively.

[0074] In particular, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. When the computer program is executed by the processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0075] The speed reducer performance comparison testing equipment for humanoid robots provided in this application, employing the speed reducer performance comparison testing method for humanoid robots described in the above embodiments, can solve the technical problem of how to improve the accuracy of speed reducer performance comparison testing for humanoid robots. Compared with the prior art, the beneficial effects of the speed reducer performance comparison testing equipment for humanoid robots provided in this application are the same as the beneficial effects of the speed reducer performance comparison testing method for humanoid robots provided in the above embodiments, and other technical features in this speed reducer performance comparison testing equipment for humanoid robots are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0076] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0077] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0078] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, which are used to execute the speed reducer performance comparison test method applied to a humanoid robot in the above embodiments.

[0079] The aforementioned computer-readable storage medium carries one or more programs. When these programs are executed by a performance comparison testing device for a humanoid robot's reducer, the device performs the following: acquires motion information and task description information of the actuator, and constructs a multi-condition test sequence based on the motion and task description information; controls the reducer under test to replay the conditions based on the multi-condition test sequence, and acquires multi-source operating data corresponding to the multi-condition test sequence during the replay process; performs time reference alignment and measurement calibration on the multi-source operating data to obtain standardized test data; performs data preprocessing and feature calculation based on the standardized test data to obtain a unified index set; and performs normalization processing and statistical comparative analysis on the unified index set to obtain a comparative evaluation result of the reducer under test. Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof. These programming languages ​​include object-oriented programming languages—such as Java, Smalltalk, and C++—and conventional procedural programming languages—such as the "C" language or similar programming languages. The program code can execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or it can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0080] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0081] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0082] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described method for comparing the performance of reducers applied to humanoid robots. This solves the technical problem of how to improve the accuracy of performance comparison tests of reducers applied to humanoid robots. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the performance comparison test method for reducers applied to humanoid robots provided in the above embodiments, and will not be repeated here.

[0083] This application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method for comparing the performance of a speed reducer applied to a humanoid robot.

[0084] The computer program product provided in this application solves the technical problem of how to improve the accuracy of performance comparison tests of reducers applied to humanoid robots. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as the beneficial effects of the performance comparison test method for reducers applied to humanoid robots provided in the above embodiments, and will not be repeated here.

[0085] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent scope of this application.

Claims

1. A method for comparative testing of reducer performance applied to humanoid robots, characterized in that, The method includes: Obtain motion information and task description information of the actuator, and construct a multi-condition test sequence based on the motion information and the task description information; Based on the multi-condition test sequence, the reducer under test is controlled to perform condition replay, and multi-source operating data corresponding to the multi-condition test sequence is acquired during the condition replay process. The multi-source operational data is time-referenced and calibrated to obtain standardized test data. Based on the standardized test data, data preprocessing and feature calculation are performed to obtain a unified set of indicators; The unified index set is normalized and statistically compared to obtain the comparative evaluation results of the speed reducer under test.

2. The method as described in claim 1, characterized in that, The step of acquiring motion information and task description information of the actuator, and constructing a multi-condition test sequence based on the motion information and task description information, includes: The motion trajectory data and load-related operation records of the actuator are acquired, and the motion trajectory data and load-related operation records are parsed to obtain task feature information; Based on the task feature information, the execution process is divided into working conditions, typical working condition segments corresponding to different operating states are extracted, and corresponding control mode parameters are configured for each typical working condition segment. The multi-condition test sequence is obtained by combining and arranging the typical operating condition segments and their corresponding control mode parameters.

3. The method as described in claim 1, characterized in that, The step of performing time base alignment and measurement calibration on the multi-source operational data to obtain standardized test data includes: Based on a unified time reference, time identifier matching is performed on each data channel in the multi-source operational data, and data with different sampling frequencies are resampled to obtain time-synchronized data. Based on the time synchronization data, delay compensation and event location calibration are performed on each data channel to obtain aligned running data; The alignment running data is processed to correct measurement errors based on preset calibration parameters, thereby generating the standardized test data.

4. The method as described in claim 1, characterized in that, The step of performing data preprocessing and feature calculation based on the standardized test data to obtain a unified set of indicators includes: The standardized test data is subjected to anomaly identification and data smoothing, and the processing results are divided into intervals based on the working condition execution status to obtain a preprocessed dataset. Feature parameters are extracted based on the preprocessed dataset, and the feature parameters are processed according to the preset index definition strategy to obtain multidimensional index data. The multidimensional indicator data is mapped to an indicator structure and organized in a unified format to generate the unified indicator set.

5. The method as described in claim 4, characterized in that, The steps of identifying and smoothing abnormal data in the standardized test data, and dividing the processing results into intervals based on the operating condition to obtain a preprocessed dataset include: The standardized test data is subjected to data integrity detection and outlier identification processing. Data samples that deviate from the preset range of variation are identified and removed to obtain preliminary processed data. The preliminary data is smoothed to reduce the impact of fluctuations between adjacent data sampling points, thus obtaining continuous operating data. Based on the operating condition execution status information corresponding to the multi-condition test sequence, the continuous running data is divided into intervals and labeled with status to obtain the preprocessed dataset.

6. The method as described in claim 1, characterized in that, The step of normalizing and statistically comparing the unified index set to obtain the comparative evaluation results of the speed reducer under test includes: Based on a preset index scaling strategy, the data of each index in the unified index set are subjected to dimension unification and interval mapping to obtain normalized index data. Based on the normalized index data, the working condition dimension is aggregated according to the preset index correlation relationship to obtain the index statistics data corresponding to different working conditions. Based on the statistical data of the aforementioned indicators, difference analysis and sorting are performed to generate comparative evaluation results for the speed reducer under test.

7. A performance comparison and testing device for a speed reducer applied to a humanoid robot, characterized in that, The device includes: The multi-condition testing module is used to acquire motion information and task description information of the actuator, and to construct a multi-condition testing sequence based on the motion information and the task description information. The multi-source operation module is used to control the speed reducer under test to perform operation condition replay based on the multi-operation condition test sequence, and to acquire multi-source operation data corresponding to the multi-operation condition test sequence during the operation condition replay process. The standardized testing module is used to perform time base alignment and measurement calibration on the multi-source operating data to obtain standardized test data. The unified indicator module is used to perform data preprocessing and feature calculation based on the standardized test data to obtain a unified indicator set. The target module is used to normalize and statistically compare the unified index set to obtain the comparative evaluation results of the speed reducer under test.

8. A computer device, characterized in that, The device includes: a memory, a processor, and a speed reducer performance comparison test program for humanoid robots stored in the memory and executable on the processor, wherein the speed reducer performance comparison test program for humanoid robots is configured to implement the steps of the speed reducer performance comparison test method for humanoid robots as described in any one of claims 1 to 6.

9. A storage medium, characterized in that, The storage medium stores a speed reducer performance comparison test program for humanoid robots. When the speed reducer performance comparison test program for humanoid robots is executed by the processor, it implements the steps of the speed reducer performance comparison test method for humanoid robots as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the steps of the speed reducer performance comparison test method applied to a humanoid robot as described in any one of claims 1 to 6.

Images