Dynamic parameter testing method and platform for joint module structure of humanoid robot

Abstract

The invention discloses a humanoid robot joint module structure dynamic parameter testing method and platform, and belongs to the technical field of robot testing. The method comprises the following steps: constructing a test platform, fixing a joint module stator, connecting an output end with a connecting rod with a load, and installing a high-resolution encoder; performing rigidity test, recording torque and angle change by horizontally placing the connecting rod and increasing and decreasing weights, and fitting to obtain a rigidity curve; carrying out a damping test, vertically placing the on-load connecting rod, applying instantaneous excitation, collecting vibration attenuation data, and calculating a damping coefficient. The test platform is simple in structure and mainly comprises a basic platform, a mounting support, a connecting rod, a weight and an encoder. The system comprises the platform and a data acquisition and processing unit. According to the method, the problem that the rigidity and damping parameters of the joint module cannot be conveniently obtained in the prior art is solved, the method is economical and efficient, the obtained parameters can be directly used for robot dynamics modeling and motion control optimization, and the motion precision and the vibration reduction performance are improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of robot testing, and relates to a method for testing structural dynamics parameters of a humanoid robot joint module, a testing platform and system for implementing the method. BACKGROUND

[0002] The rapid development of humanoid robot technology puts high demands on the performance of integrated joint modules, which are the core motion execution units. The performance of the joint module directly determines the motion accuracy, dynamic response capability, load capacity, and overall reliability of the robot. In order to optimize the motion control algorithm of the robot, especially in terms of vibration reduction and trajectory tracking during high-speed and high-precision motion, it is not enough to only know the static performance parameters of the joint module (such as maximum torque, position accuracy), and it is necessary to master its key structural dynamics parameters, especially the stiffness (reflecting the ability to resist deformation) and damping (reflecting the ability to dissipate vibration energy) characteristics.

[0003] Currently, the industry focuses more on static or quasi-static performance in testing robot joints or modules. For example, patent application CN119223608A discloses a humanoid robot joint module performance testing system, which fixes the joint module through positioning claws and positioning blocks, and uses a loading hydraulic cylinder to apply pressure. The main test is the compression resistance, stability, and durability of the joint module under static or low-speed conditions. Although this system can obtain data such as pressure and deformation, it does not involve dynamic excitation, collection, and analysis of structural dynamics parameters, and cannot provide stiffness and damping information for model-based control algorithms.

[0004] Another patent application CN120588284A focuses on the flexible fixation and damage prevention design of the testing equipment, and fixes the joint module through flexible elements such as arc-shaped pads, and tests its swing under different loads. Although this scheme considers dynamic load testing, its testing purpose is more focused on evaluating the working performance of the joint module under load and the influence of temperature rise, and the test results (such as swing frequency) are macroscopic manifestations of comprehensive performance, and the structural stiffness and damping coefficient of the joint module itself are not decoupled through specific excitation and data analysis methods.

[0005] In addition, patent application CN120363258A shows an automatic joint performance testing device that can automatically replace different weights of counterweights for load testing, and even simulate impact environments. This device focuses on the automation and efficiency of the testing process, as well as environmental adaptability testing such as impact resistance, waterproofness, and dustproofness. Its testing target is still torque output capability, anti-interference capability, etc., and it does not design a special testing process and algorithm for extracting internal structural dynamics parameters.

[0006] There are also some technologies focusing on the sensing and efficiency testing of the joint itself, such as patent application CN118700207A discloses a linear joint integrated with a tensile-torsional composite force sensor and its test platform, which can calculate the transmission efficiency in real time. This belongs to the category of joint body design optimization, rather than external, non-invasive dynamic parameter calibration of the existing joint module product.

[0007] In summary, the prior art lacks a method and device specifically for humanoid robot integrated joint module, which can conveniently, economically and effectively test its key structural dynamic parameters (stiffness curve and damping coefficient). The existing test scheme is either too complex and expensive (such as integrated complex automatic loading and environment simulation system), or the test target is irrelevant to the extraction of structural dynamic parameters. Therefore, developing a test scheme to fill this gap is of great significance to improve the model accuracy and performance of humanoid robot motion control algorithm. SUMMARY

[0008] The present application aims to overcome the shortcomings of the prior art and provide a humanoid robot joint module structural dynamic parameter testing method, platform and system. This method can conveniently obtain the stiffness curve and damping coefficient of the joint module through simple test settings and processes, providing key structural dynamic model parameters for high-precision motion control of humanoid robots, especially vibration suppression. To achieve the above purpose, the present application adopts the following technical solutions.

[0009] In a first aspect, the present application provides a method for testing the structural dynamic parameters of a humanoid robot joint module, comprising the following steps: S1: Constructing a test platform, fixing the stator of the joint module to be tested on the mounting support of the platform, connecting the adapter link and the load to the output end of the joint module, and installing a high-resolution encoder on the output end; S2: Performing stiffness curve testing, including: S2-1: Adjusting the adapter link without load to a horizontal position and performing inclination calibration; S2-2: Increasing and decreasing the weights on the adapter link in sequence to apply different torques, and simultaneously recording the output end angle change value detected by the high-resolution encoder; S2-3: Obtaining a series of angle-torque data points according to the recorded torque values and corresponding output end angle change values; S2-4: Program fitting the angle-torque data points to obtain the stiffness curve of the joint module; S3: Performing damping coefficient testing, including: S3-1: Adjusting the adapter link with the load to a vertical position; S3-2: Apply instantaneous excitation in a specific direction perpendicular to the adapter link, and collect the output end angle vibration data detected by the high-resolution encoder; S3-3: Based on the collected angular vibration data, the damping coefficient of the joint module is calculated by the program.

[0010] This invention provides a complete and streamlined testing method that can systematically acquire structural dynamic parameters (stiffness and damping) crucial for the motion control of humanoid robots, solving the problem that existing technologies can only test static performance (such as position and pressure) and cannot evaluate dynamic characteristics. By combining simple platform construction, specific test postures (horizontal and vertical), and data acquisition and fitting processes, efficient and economical testing of the intrinsic dynamic characteristics of joint modules is achieved.

[0011] Furthermore, in step S2-1, an inclinometer is used to calibrate the horizontal position of the transition link. By introducing an inclinometer for precise calibration, the accuracy of the initial state (zero torque point) during stiffness testing is ensured, thereby improving the reference accuracy of the angle-torque data points and ultimately enhancing the fitting accuracy of the stiffness curve.

[0012] Furthermore, in step S2-2, the weights are increased or decreased at intervals of equal or varying mass to cover the torque range of the joint module within the expected working range. Through a controllable weight increase / decrease strategy, deformation data of the joint module at different torque levels within the working range can be collected in a targeted manner, making the resulting stiffness curve more reflective of its nonlinear stiffness characteristics under actual working conditions.

[0013] Furthermore, in steps S2-4, the program fitting employs the least squares method or a polynomial fitting algorithm.

[0014] Furthermore, in step S3-2, the instantaneous excitation is applied by striking with a hammer in the transverse direction along the connecting rod. Using hammer striking as the excitation source is simple, low-cost, and can generate damped vibration signals that are easy to analyze and identify, making it particularly suitable for obtaining the damping characteristics of the system.

[0015] Furthermore, in step S3-2, the sampling rate for acquiring the angular vibration data is set to 1000Hz or higher. Setting a higher sampling rate can capture the high-frequency components and details of the vibration signal, ensuring the integrity of the attenuation curve data and providing a high-quality data foundation for the subsequent accurate calculation of the damping coefficient.

[0016] Furthermore, in step S3-3, the damping coefficient is obtained by calculating the logarithmic decay rate of the angular vibration data or by using frequency domain analysis. Through mature data processing methods, the calculation results of the damping coefficient are scientific and accurate, enhancing the practicality and reliability of the method.

[0017] Secondly, the present invention provides a testing platform for the structural dynamic parameters of a humanoid robot joint module for implementing the above-mentioned testing method, comprising: platform; Mounting bracket, fixed on the platform, is used to mount and fix the stator of the joint module to be tested; The adapter link has one end for connecting to the output end of the joint module; The load is configured on the adapter link; A high-resolution encoder is used to connect to the output of the joint module to detect changes in its angle.

[0018] The test platform provided by this invention has a simple structure, mainly consisting of a basic platform, mounting components, loading linkages, and encoders. It has low manufacturing costs and is easy to assemble and operate. Its core feature is that it supports the execution of the method of this invention through a specific configuration.

[0019] Furthermore, the load is a standard weight that can be increased or decreased. Using a standard weight as the load ensures that the applied torque value is accurate, known, and traceable, fundamentally guaranteeing the accuracy of the input quantity (torque) in stiffness testing, simplifying the load system, and avoiding the use of complex electric loading mechanisms.

[0020] Thirdly, the present invention provides a system for testing the structural dynamic parameters of a humanoid robot joint module, comprising: The aforementioned testing platform; The data acquisition unit is communicatively connected to the high-resolution encoder and is used to record angle data; The data processing unit, connected to the data acquisition unit, is configured as follows: Based on the angle-torque data recorded by the data acquisition unit during the stiffness test, a stiffness curve is fitted and generated; The damping coefficient is calculated based on the angular vibration data recorded by the data acquisition unit during the damping test.

[0021] Compared with the prior art, the present invention has the following beneficial technical effects: Addressing specific technological gaps: For the first time, a complete testing scheme specifically designed for obtaining the structural dynamic parameters (stiffness, damping) of humanoid robot joint modules is proposed, overcoming the limitation of existing technologies that can only test static or macroscopic performance and cannot provide key model parameters for model-based advanced motion control.

[0022] The method is simple, cost-effective, and efficient: the testing method has a clear logic and well-defined steps. The required testing platform structure is extremely simple, mainly consisting of a base, mounting components, standard weights, and a high-precision encoder. This avoids complex and expensive automatic loading, multi-axis motion, or environmental simulation systems, significantly reducing testing costs and technical barriers, and facilitating its promotion in R&D and production.

[0023] Accurate and practical data: The accuracy of the zero point of the stiffness test reference is ensured through horizontal position calibration; the free decay vibration signal after excitation is collected through a high sampling rate encoder to ensure data quality. Mature mathematical methods (such as curve fitting and logarithmic decay rate calculation) are used to process the data, and the resulting stiffness curves and damping coefficients are reliable and can be directly used for robot dynamics modeling and control algorithm optimization to improve vibration reduction performance during motion.

[0024] High system integration: It integrates a simple hardware platform with a dedicated data processing logic (algorithm) to form a complete testing system, which clarifies the entire process from data acquisition to parameter generation, and achieves the specificity of the testing purpose and the directness of the result output. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the test platform structure provided in an embodiment of the present invention.

[0026] Figure 2 A flowchart of the testing method provided in an embodiment of the present invention.

[0027] Figure 3 This is a schematic diagram of the angle-torque data points and fitting curves in stiffness testing.

[0028] Figure 4 This is a schematic diagram of the angular vibration decay curve collected during the damping test.

[0029] Reference numerals: 1-Cast iron platform; 2-L-shaped mounting bracket; 3-Joint module stator; 4-Joint module output end; 5-Adapter link; 6-Load; 7-High-resolution encoder; 8-Inclinometer for level calibration; 9-Force hammer. Detailed Implementation

[0030] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the protection scope of the present invention.

[0031] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance, quantity, or position.

[0032] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0033] Example 1: Test Method like Figure 2 As shown in the figure, this embodiment describes in detail a method for testing the structural dynamic parameters of a humanoid robot joint module.

[0034] Step S100: Platform construction and initialization.

[0035] First, build such Figure 1 The simplified test platform is shown. A sturdy and stable cast iron platform 1 is placed on a level surface. The L-shaped mounting bracket 2 is securely fixed to the cast iron platform 1. The stator 3 (i.e., the fixed part) of the integrated joint module of the humanoid robot to be tested is rigidly mounted on the vertical surface of the L-shaped mounting bracket 2 using bolts or other clamps, ensuring that it does not move throughout the test.

[0036] One end of the adapter link 5 is rigidly connected coaxially to the output end 4 (i.e., the rotating shaft) of the joint module. The other end of the adapter link 5 is used to suspend or mount the load. A high-resolution encoder 7 (e.g., an absolute or incremental photoelectric encoder) is installed on the output end 4 of the joint module. Its resolution should be high enough to capture minute angular changes (typically at least at the arcsecond level). The signal line of the high-resolution encoder 7 is connected to subsequent data acquisition equipment (such as a data acquisition card or an industrial control computer with acquisition capabilities).

[0037] Step S200: Stiffness curve test.

[0038] The goal of this step is to obtain the relationship between the output torque of the joint module and the elastic torsional angle at the output end, i.e., the stiffness curve. Stiffness is usually not a constant, but a curve that varies with torque, reflecting the nonlinear characteristics of the system.

[0039] S210: Preparation and Calibration. Remove all loads 6 from the adapter link 5. Using a high-precision inclinometer 8, carefully adjust the position of the adapter link 5 to ensure it is perfectly horizontal. This step is crucial as it defines the reference zero point for the joint output angle under zero torque loading. Record the reading of the high-resolution encoder 7 at this point as the initial angle θ0.

[0040] S220: Data Acquisition. Weights are used as the standard load. Weights of known mass are sequentially added to the connecting rod 5 at a position of a specific lever arm length L from the center of rotation (e.g., on the hook at the end of the connecting rod). Each weight has a mass of m, and the torque it generates is T = m × g × L (where g is the acceleration due to gravity). After each weight is added or removed, once the system stabilizes (i.e., the encoder reading no longer changes), the currently applied torque value T is synchronously recorded. i and the angle value θ read from the high-resolution encoder 7 i The torque value was started from zero and gradually increased to a certain percentage (e.g., 80%) of the rated torque of the joint module, and then gradually unloaded to examine the hysteresis effect. A set of data points was finally obtained {(T)}. i , Δθ i )}, where Δθ i = θ i -θ0.

[0041] S230: Curve Fitting. Import the angle-torque data points obtained in step S220 into data processing software (such as MATLAB, Python, etc.). Due to the gaps and elastic deformations in gears, bearings, etc., inside the joint module, the data points may exhibit nonlinearity. A numerical fitting algorithm, such as the least squares method, is used for polynomial fitting to obtain a curve K(T) = dT / d(Δθ) that best represents the trend of the data points. This curve is the stiffness curve of the joint module. Figure 3 As shown, the horizontal axis represents the angle change Δθ, and the vertical axis represents the torque T. The slope of the fitted curve at a certain point is the instantaneous stiffness under that torque. The stiffness curve can clearly show the stiffness performance of the joint module under different loads and is an important basis for controller design.

[0042] Step S300: Damping coefficient test.

[0043] The goal of this step is to obtain the damping ratio of the joint module system, which describes the system's ability to dissipate vibrational energy.

[0044] S310: Test configuration setup. Install a suitable fixed load 6 (e.g., 10-20% of the rated load) on the adapter link 5 to give it sufficient rotational inertia. Adjust the loaded adapter link 5 to a vertically downward position (i.e., the equilibrium position). At this point, the torque generated by gravity is balanced by the internal torque of the joint module. Record the angle of the encoder 7 at this point as a reference.

[0045] S320: Excitation and Data Acquisition. Using a modal hammer 9, apply a momentary impact excitation to the side of the connecting rod 5 (e.g., along the Y direction, i.e., horizontal and perpendicular to the rod). The impact should be clean and crisp, causing the system to generate a free-dampening vibration around the vertical equilibrium position. Simultaneously, activate the high-speed data acquisition system, continuously recording the angle change data output by the high-resolution encoder 7 at a sampling rate of at least 1000Hz (a higher sampling rate is preferred to ensure capture of the decay process) for several seconds until the vibration essentially subsides. The acquired signal should be as follows: Figure 4 As shown, it is a sine wave whose amplitude decays exponentially with time.

[0046] S330: Parameter Calculation. Analysis of the acquired angle-time series (vibration decay curve). The most commonly used method is the logarithmic decay rate method. From... Figure 4 On the decay curve, select the amplitude A of two adjacent peaks (or valleys) in the same direction. n and A n+1 The formula for calculating the logarithmic decay rate δ is: δ = ln(A n / A n+1 The damping ratio ζ of the system can be expressed by the formula ζ = δ / sqrt(4π 2 + δ 2 Approximate calculation. To improve accuracy, the peak values ​​of multiple cycles can be selected for calculation and then averaged. The final calculated ζ is the equivalent damping coefficient of the joint module in this test configuration. Alternatively, a Fast Fourier Transform (FFT) can be performed on the vibration signal, and the damping ratio can be calculated in the frequency domain using the half-power bandwidth method, as a cross-validation.

[0047] Through the two relatively independent but complementary test procedures S200 and S300, two core parameters describing the dynamic characteristics of the joint module structure can be obtained: the stiffness curve K(T) and the damping ratio ζ. These parameters can be input into the dynamic model of the humanoid robot for designing feedforward compensation, impedance control, or predictive controllers, significantly improving motion accuracy and suppressing end-effector vibration.

[0048] Example 2: Test Platform like Figure 1 As shown, this embodiment provides a dedicated testing platform for implementing the method described in Embodiment 1.

[0049] The platform is based on a high-rigidity cast iron platform 1, ensuring a stable foundation during testing and preventing the introduction of additional vibrations. L-shaped mounting brackets 2 are welded to or secured to the cast iron platform 1 with high-strength bolts, and their vertical surfaces provide mounting interfaces for the joint module stator 3. The mounting interfaces can be adapted to fit the flanges of different joint module models.

[0050] The adapter link 5 is a rigid member, with a connecting flange or sleeve at one end that matches the output end 4 of a specific joint module, and a hook or mounting plate at the other end for suspending the standard weights 6. The weights 6 are a series of metal blocks of known mass, which are the source of precise torque excitation.

[0051] The high-resolution encoder 7 is the core measurement component of this platform. It is directly coaxially connected to the output end 4 of the joint module via a coupling, ensuring accurate measurement of the true rotation angle of the output shaft. The encoder selection should meet the test accuracy requirements; for example, for high-precision joints, an absolute encoder with 23 bits or more may be required.

[0052] Optionally, the platform can also be equipped with a small tool rack for storing the inclinometer 8 and the force hammer 9. The entire platform has a simple structure without complex motion mechanisms or electronic control systems. Its core lies in ensuring the feasibility of test actions (horizontal loading, vertical suspension excitation) through mechanical configuration and acquiring data through high-precision sensors.

[0053] Example 3: Testing System This embodiment provides an integrated testing system, which includes a hardware platform and software algorithms.

[0054] The system includes: Hardware subsystem: namely, the test platform described in Example 2 (including cast iron platform 1, L-shaped support 2, adapter rod 5, weights 6, and high-resolution encoder 7).

[0055] Data acquisition unit: This includes a data acquisition card (DAQ Card) compatible with the high-resolution encoder 7 interface or an industrial computer (industrial control computer) with integrated acquisition capabilities. This unit is responsible for reading and storing the encoder's angle values ​​in real time during testing, according to a preset sampling rate (lower for stiffness testing, and higher than 1000Hz for damping testing). For stiffness testing, the current weight mass must be manually input or automatically read through an electronic scale interface, converted into a torque value, and then bound and stored with the angle value.

[0056] Data processing and output unit: This can be a computer with dedicated analysis software installed. The following processing modules are pre-installed or configured by the user within this unit: Stiffness analysis module: Reads the torque-angle data pairs stored in the data acquisition unit. It calls the built-in curve fitting algorithm library (such as least squares polynomial fitting) to automatically perform curve fitting. Multiple fitting orders are available for selection, and the original data points and the fitted curve can be graphically displayed (e.g., ...). Figure 3 It also outputs the function expression of the fitted curve or a list of key point stiffness values.

[0057] Damping Analysis Module: Reads the "time-angle" vibration decay data stored in the data acquisition unit. It automatically identifies the peak points of the decay waveform and calculates the damping ratio using the logarithmic decay rate formula. This module can also integrate FFT functionality, providing calculation results from the frequency domain half-power bandwidth method as a reference. Finally, it outputs the damping ratio ζ and can graphically display the decay curve and peak markers (e.g., ...). Figure 4 ).

[0058] Report generation module: Automatically integrates the final results of stiffness curves and damping coefficients, along with test conditions (such as joint module model, test date, load arm length, etc.) to generate a standardized test report.

[0059] This system combines simple dedicated hardware with targeted intelligent analysis algorithms to form a closed-loop solution. Users only need to operate the platform and execute the test steps according to the procedures, and the system can automatically complete the entire process from data acquisition and processing to parameter extraction and report generation, which greatly improves testing efficiency and result consistency, making structural dynamics parameter testing standardized and easy to use.

[0060] 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 the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

[0061] Although embodiments of the present invention have been shown and described above, it is understood that these embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions, and alterations to the above embodiments within the scope of the present invention without departing from its principles and spirit. The scope of protection of the present invention is defined by the claims and their equivalents.

Claims

1. A method for testing the structure dynamics parameters of a humanoid robot joint module, characterized in that, The method comprises the following steps: S1: constructing a test platform, fixing a stator of a joint module to be tested on a mounting support of the platform, connecting an adapter link and a load to an output end of the joint module, and installing a high-resolution encoder on the output end; S2: performing a stiffness curve test, comprising: S2-1: adjusting the adapter link without the load to a horizontal position and performing inclination calibration; S2-2: sequentially increasing and decreasing the weights on the adapter link to apply different torques, and synchronously recording the output end angle change values detected by the high-resolution encoder; S2-3: obtaining a series of angle-torque data points according to the recorded torque values and corresponding output end angle change values; S2-4: performing program fitting on the angle-torque data points to obtain a stiffness curve of the joint module; S3: performing a damping coefficient test, comprising: S3-1: adjusting the adapter link with the load to a vertical position; S3-2: applying a transient excitation in a specific direction perpendicular to the adapter link, and collecting output end angle vibration data detected by the high-resolution encoder; S3-3: calculating the damping coefficient of the joint module based on the collected angle vibration data.

2. The test method of claim 1, wherein, In step S2-1, an inclination instrument is used to calibrate the horizontal position of the adapter link.

3. The test method of claim 1, wherein, In step S2-2, the weights are increased and decreased at equal or variable mass intervals to cover the torque interval of the joint module in the expected working range.

4. The test method of claim 1, wherein, In step S2-4, the program fitting adopts a least squares method or a polynomial fitting algorithm.

5. The test method of claim 1, wherein, In step S3-2, the transient excitation is applied by a force hammer knock, and the knock direction is transverse to the adapter link.

6. The test method of claim 5, wherein, In step S3-2, the sampling rate for collecting the angle vibration data is set to 1000 Hz or above.

7. The test method of claim 1, wherein, In step S3-3, the damping coefficient is obtained by calculating the logarithmic decay rate of the angle vibration data or using a frequency domain analysis method.

8. A humanoid robot joint module structure dynamics parameter test platform for implementing the test method of any one of claims 1-7, characterized in that, The test platform comprises: a platform; a mounting support fixed to the platform for mounting and fixing a stator of a joint module to be tested; an adapter link having one end for connecting to an output end of the joint module; a load arranged on the adapter link; a high-resolution encoder for connecting to the output end of the joint module to detect the angle change thereof.

9. The test platform of claim 8, wherein, The load is a standard weight that can be increased or decreased.

10. A humanoid robot joint module structure dynamics parameter testing system, characterized in that, The test platform comprises: the test platform of claim 8 or 9; a data acquisition unit in communication connection with the high-resolution encoder for recording angle data; a data processing unit connected with the data acquisition unit and configured to: fit and generate a stiffness curve according to the angle-torque data recorded by the data acquisition unit in the stiffness test; and calculate and generate a damping coefficient according to the angle vibration data recorded by the data acquisition unit in the damping test.

Images