Human arm impedance measuring method and device

Abstract

The invention discloses a human arm impedance measurement method and device, and the method comprises the steps: 1, designing a double-arm humanoid robot, constructing a kinetic model, and collecting the operation data of the interaction between a measured human arm and the two arms of the robot through the robot; step 2, carrying out preprocessing operations of denoising, numerical derivation, data synchronization and mean value calculation on the operation data; 3, establishing a human body impedance model through a linear regression method by using the preprocessed data, calculating impedance parameters based on mass, damping and rigidity under the condition of different arm angles and interaction forces, and minimizing a loss function to optimize model parameters; and step 4, according to the impedance parameters obtained by calculation, fitting and drawing a three-dimensional curved surface by using a B-spline curved surface equation, and displaying impedance characteristics under different arm angles and interaction force conditions. According to the invention, comprehensive analysis of the impedance characteristic of the arm of the human body in the anthropomorphic operation process can be realized, and cooperative work of robots and human beings and application in the fields of biomechanical research and the like are facilitated.

Description

Technical Field

[0001] The present invention relates to the technical field of impedance simulation measurement, and in particular to a method and device for measuring the impedance of a human arm. Background Art

[0002] At present, the method of measuring the impedance of the human arm mainly relies on a single sensor or a simple mechanical device, which has problems such as low measurement accuracy and complex operation. In the prior art, it is difficult to apply a specific disturbance force to the human arm in a real operating environment to stimulate the impedance characteristics of the human arm and measure the impedance characteristics of the human arm, especially it is difficult to obtain the nonlinear impedance characteristics of the impedance under different arm angles and interaction forces. Summary of the invention

[0003] In order to solve the above technical problems existing in the prior art, the present invention proposes a method and device for measuring the impedance of a human arm to improve the measurement accuracy and the convenience of operation, and to achieve impedance parameter fitting under different arm angles and interaction forces, and to measure the impedance characteristics of the human arm under different arm angles and interaction forces. The specific technical scheme is as follows:

[0004] A method for measuring human arm impedance, comprising:

[0005] Step 1: Design a dual-arm humanoid robot and build its dynamic model. The robot collects the operational data of the interaction between the human arm under test and the robot's dual arms, and marks and classifies the data under different arm angles and interaction forces of the human arm under test.

[0006] Step 2: preprocessing the operation data by denoising, numerical derivation, data synchronization and mean calculation;

[0007] Step 3: Use the preprocessed data to establish a human body impedance model through a linear regression method, calculate the impedance parameters based on mass, damping, and stiffness under different arm angles and interaction forces, and minimize the loss function to optimize the model parameters;

[0008] Step 4: Based on the calculated impedance parameters, the B-spline surface equation is used to fit and draw a three-dimensional surface to show the impedance characteristics under different arm angles and interaction forces.

[0009] Furthermore, the denoising uses bandpass filtering technology; the numerical derivation derives velocity and acceleration from displacement data through numerical difference calculation to obtain complete motion information; the data synchronization uses timestamp or signal synchronization technology to calibrate and align multi-sensor data to ensure that all sensor data are synchronized so that each time point corresponds to a data sample; the data mean calculation takes the average of multiple collected operation data of the same arm angle and interaction force to obtain a unique operation data sequence.

[0010] Furthermore, the specific expression for establishing the human body impedance model is as follows:

[0011]

[0012] Where: t∈(t0,t e ), t0 is the starting time of applying disturbance to the human arm, t e is the disturbance cancellation time, dx(t) and dF(t) are preprocessed data;

[0013] dx(t)=x(t)-x(t0), dF(t)=F(t)-F(t0), x(t) and F(t) are the time series of arm position and force;

[0014] are impedance parameters, representing mass, damping and stiffness respectively, represents the arm angle, f ext represents the interaction force;

[0015] By using the linear regression method, formula (1) is organized into a matrix format:

[0016] Y=XΘ+E,

[0017] in:

[0018] Y = dF(t);

[0019]

[0020] Furthermore, the loss function expression of the human body impedance model is:

[0021] By minimizing the loss function, each set of arm angles is calculated and the interaction force f ext The corresponding impedance parameters.

[0022] Furthermore, the step 4 is specifically as follows: according to the impedance parameters obtained, using the B-spline surface equation, solve and collect each arm angle and the interaction force f ext The corresponding mass, damping and stiffness data form a three-dimensional data point set, and the arm angle Interaction force f ext The surface composed of impedance parameters describes different arm angles and the interaction force f ext Impedance characteristics.

[0023] A human arm impedance measuring device, comprising:

[0024] A two-arm humanoid robot, the robot comprising: a two-degree-of-freedom waist for adjusting posture;

[0025] Two left and right seven-degree-of-freedom robotic arms, used for interactive movement with human arms;

[0026] A two-degree-of-freedom head including a vision system for real-time detection of the arm angle of a human arm;

[0027] Seat and fixing belt: used to fix the test person; the person being tested sits on a seat with a backrest, the body trunk is fixed by the fixing belt, and the seat is connected to the robot base, so that the body trunk and the two-arm humanoid robot do not move relative to each other during the measurement process;

[0028] Wrist fixture: connected to the end of the seven-degree-of-freedom robotic arm, used to connect and fix the human arm;

[0029] End six-dimensional force sensor: It is set at the end of two seven-degree-of-freedom robotic arms and connected to the tester's arm through a wrist fixture to obtain the force between the human arm and the robotic arm.

[0030] Controller: Installed in the body of the dual-arm humanoid robot, it collects sensor data through the EtherCAT high-speed bus and is used to control the movement of the dual-arm humanoid robot, process the collected data, and calculate the impedance of the human arm.

[0031] Furthermore, the controller specifically includes:

[0032] A data acquisition module is used to collect the force, position and arm angle data at each time point during the operation and store the data;

[0033] Data preprocessing module, used for denoising, numerical derivation, data synchronization and mean calculation of collected data;

[0034] Impedance analysis and modeling module, which is used to establish a linear regression model based on mass, damping and stiffness, and calculate the impedance parameters under different arm angles and interaction forces;

[0035] The surface fitting module is used to perform three-dimensional surface fitting based on the impedance parameter data using the B-spline surface equation to display the characteristic surface of the arm angle and interaction force and impedance parameters of the human arm.

[0036] The present invention provides a high-precision and easy-to-operate human arm impedance measurement device and method, which can provide an evaluation of the human body impedance characteristics under the influence of arm angle and interaction force, and realize a comprehensive analysis of the human arm impedance characteristics during anthropomorphic operation, which is helpful for applications in the fields of collaborative work between robots and humans and biomechanics research. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a structural schematic diagram of a human arm impedance measuring device of the present invention;

[0038] Among them: 1- dual-arm humanoid robot; 2- visual system; 3- tester; 4- seat; 5- fixing belt; 6- wrist fixing device; 7- end six-dimensional force sensor;

[0039] Figure 2 This is a coordinate system relationship diagram of the dual-arm humanoid robot described in the present invention. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical scheme and technical effect of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0041] like Figure 1 As shown, a human arm impedance measuring device designed by the present invention comprises:

[0042] Dual-arm humanoid robot 1: used to measure the position of human arms, including the following components:

[0043] The two-degree-of-freedom waist enables the robot to flexibly adjust its posture;

[0044] A 2-DOF head with an advanced vision system 2 for real-time detection of the angle of a human arm;

[0045] The two high-precision seven-degree-of-freedom robotic arms on the left and right accurately capture the operating position and hand movement.

[0046] The seat 4 and the fixing belt 5 are used to fix the test person 3 to avoid relative movement of the human body. Specifically, the person being tested sits on the seat 4 with a backrest, the human torso is fixed by the fixing belt 5, and the seat 4 is connected to the robot base to ensure that there is no relative movement between the human torso and the two-arm humanoid robot 1 during the experiment.

[0047] Wrist fixing device 6: used to connect and fix the human arm to prevent relative displacement between the arm of the person being measured and the end six-dimensional force sensor 7 and the robot arm.

[0048] The end six-dimensional force sensors 7 are respectively arranged at the ends of the mechanical arms of the dual-arm humanoid robot 1 and connected to the arms of the tester 3, and are used to measure the force between the human arm and the dual arms of the humanoid robot.

[0049] Controller: Installed in the body of the dual-arm humanoid robot, it collects sensor data through the EtherCAT high-speed bus and is used to control the movement of the dual-arm humanoid robot, process the collected data, and calculate the impedance of the human arm.

[0050] Specifically, the controller includes: a data acquisition module for collecting force, position and arm angle data at each time point during the operation and storing the data;

[0051] Data preprocessing module, used for denoising, numerical derivation, data synchronization and mean calculation of collected data;

[0052] Impedance analysis and modeling module, which is used to establish a linear regression model based on mass, damping and stiffness, and calculate the impedance parameters under different arm angles and interaction forces;

[0053] The surface fitting module is used to perform three-dimensional surface fitting based on the impedance parameter data using the B-spline surface equation to display the characteristic surface of the arm angle and interaction force and impedance parameters of the human arm.

[0054] The present invention also proposes a method for measuring human arm impedance, which is specifically:

[0055] Construct the dynamic model of the dual-arm humanoid robot 1:

[0056] like Figure 2 As shown, the two-arm humanoid robot 1 is divided into three parts: left arm, right arm, and waist. B is the base coordinate system of the left and right arms and the output coordinate system of the waist; O is the base coordinate system of the waist, B r is the right arm joint coordinate system; B l is the left arm joint coordinate system; E r Output coordinate system for the right arm; E r Output coordinate system for the left arm;

[0057] Using the modified DH parameter method, all joints are revolute joints, and the dynamics of each part are obtained according to the Newton-Euler method:

[0058] z0: [0 0 1] T ;

[0059] n: number of joints;

[0060] g: gravitational acceleration vector described by waist base coordinate system O;

[0061] The link coordinate system i is the angular velocity at the joint i; for the robot arm, when i = 0, it is the angular velocity at the base coordinate system B, which is generated by the waist movement; for the waist, when i = 0, it is the angular velocity at the base coordinate system O and For the waist, when i=n+1, it is the angular velocity of the waist output coordinate system B;

[0062] The angular acceleration at the link coordinate system i is the angular acceleration at the base coordinate system B when i=0 for the robot arm, which is generated by the waist movement; for the waist, when i=0, it is the angular acceleration at the base coordinate system O and For the waist, when i=n+1, it is the angular acceleration of the waist output coordinate system B;

[0063] The linear velocity at the connecting rod coordinate system i; for the robot arm, when i = 0, it is the linear velocity at the base coordinate system B, which is generated by the waist movement; for the waist, when i = 0, it is the linear velocity at the base coordinate system O and For the waist, when i=n+1, it is the linear velocity of the waist output coordinate system B;

[0064] The linear acceleration at the connecting rod coordinate system i; for the robot arm, when i = 0, it is the linear acceleration at the base coordinate system B, which is generated by the waist movement; for the waist, when i = 0, it is the linear acceleration at the base coordinate system O and For the waist, when i=n+1, it is the linear acceleration of the waist output coordinate system B;

[0065] The linear acceleration described by the mass center coordinate system in the connecting rod coordinate system i;

[0066] The rotation matrix of coordinate system i-1 relative to i;

[0067] The position description in the i-1 coordinate system under the description of coordinate system i;

[0068] The center of mass position described by coordinate system i. The origin of the center of mass coordinate system ic is located at the center of mass of the connecting rod, and the orientation of each coordinate axis is the same as that of the connecting rod coordinate system i.

[0069] m i : Connecting rod i is the mass of the connecting rod where the coordinate system ci is located;

[0070] The inertia tensor of the link i relative to the center of mass coordinate system ci;

[0071] J i : moment of inertia of joint i;

[0072] f i i : Force acting on coordinate system i; for the robot arm, when i=0, it is the force acting on the base coordinate system B, and when i=n+1, it is the external force acting on the end coordinate system n+1; for the waist, when i=0, it is the force acting on the base coordinate system O, and when i=m+1, it is the sum of the robot arm forces acting on the coordinate system m+1, i.e., the coordinate system B;

[0073] The torque acting on coordinate system i; for the manipulator, when i=0, it is the torque acting on the base coordinate system B, and when i=n+1, it is the external torque acting on the end coordinate system n+1; for the waist, when i=0, it is the torque acting on the base coordinate system O, and when i=m+1, it is the sum of the manipulator torques acting on the coordinate system m+1;

[0074] Friction torque of joint i;

[0075] f ext : External forces and moments;

[0076] τ i : joint i torque;

[0077] q i : joint i position;

[0078] Joint i velocity;

[0079] Joint i acceleration;

[0080] Use subscripts l, r, and w to distinguish the left arm, right arm, and waist.

[0081] Dynamics of each part:

[0082] Velocity and acceleration extrapolation: i = 1, 2, ..., n k ;

[0083]

[0084] Force and moment internalization: i = n k ,n k -1,…,2,1,k=l,r,w;

[0085]

[0086] The state equation is expressed as:

[0087]

[0088] in:

[0089]

[0090] J k is the Jacobian matrix.

[0091] The left arm, right arm and waist act together on coordinate system B. From the above recursive equations of force and moment, it can be seen that the sum of the force and moment at the base coordinate system B of the left and right arms is the external force on the waist, that is:

[0092]

[0093] From the above recursive equations of velocity and acceleration, it can be seen that the velocity and acceleration of the waist output coordinate system B are used as the recursive input of the velocity and acceleration of the left and right arms, that is:

[0094]

[0095] The waist force / torque push-in can be divided into two parts:

[0096]

[0097] Obviously there are:

[0098]

[0099] in:

[0100] The dynamic equations are deduced by combining with the robot arm respectively:

[0101]

[0102]

[0103] in:

[0104]

[0105] They are transformed into:

[0106]

[0107] Add together the whole body dynamics model:

[0108]

[0109] Among them: Among them: They are the acceleration, velocity and position of the humanoid joints of both arms respectively; is the joint torque of the two-arm humanoid;

[0110] is the inertia matrix;

[0111] is the Coriolis force and centripetal force matrix;

[0112] is the gravity vector; is the Jacobian matrix; It is the interaction force between the human body and the human figure with two arms; Friction of joints for humanoid robots.

[0113] Zero-force drag function realization:

[0114] The dual-arm humanoid robot works in torque mode and uses the above dynamic model to calculate the driving force to drive the robot to move freely. The external force f in the dynamic model ext It is measured by the end six-dimensional force sensor 7.

[0115] Apply external perturbation force:

[0116] Design the arm angle according to the task requirements. and interaction force f ext The external force perturbation function The disturbance has a short action time and acts randomly multiple times during the operation task to stimulate the impedance characteristics of the human arm. Due to the short disturbance time, the arm angle during the disturbance and interaction force f ext Approximately a constant value.

[0117] Data collection and processing, including:

[0118] Preset experimental parameters: arm angle and the interaction force f ext Perform group setting, arm angle Set to 0, 10, 20, ..., 90 degrees, the interaction force f ext Set to 0, 10, 20, …, 90 Newtons.

[0119] Two phases of experiments were conducted: non-interference operation and interference operation. The operation tasks in both phases were the same, and the acquisition tasks covered different arm angles. and the interaction force f ext , as follows:

[0120] The first stage: the human arm follows the normal two-arm operation task process and drags the humanoid robot arm to complete the operation task without disturbance. During the operation, the end six-dimensional force sensor 7 is used to collect the operation force, the two arms of the humanoid two-arm robot are used to measure the operation position, and the head vision system 2 of the humanoid two-arm robot is used to measure the arm angle of the human arm.

[0121] The second stage: The human arm follows the normal two-arm operation task process. In the process of the human arm dragging the humanoid robot arm to complete the task, according to the external force disturbance function Apply an interfering external force, and during the operation process, use the end six-dimensional force sensor 7 to collect the operating force, use the two arms of the humanoid dual-arm robot to measure the operating position, and use the head vision system 2 of the humanoid dual-arm robot to measure the arm angle of the operating human body.

[0122] Training period: The first stage is repeated many times to help operators become proficient in the operating procedures and adapt to the measurement structure. The data used is used to analyze the operating proficiency and consistency of operations.

[0123] Interference data collection period: After multiple disturbance-free operations and good operation consistency, multiple second-stage data collections are performed. The arm angle measured by the visual system According to the principle of proximity, the interaction force f measured by force sensing is approximately a value among 0, 10, 20, ..., 90 degrees. ext According to the principle of proximity, the approximate value is 0, 10, 20, ..., 90 Newtons, and finally different arm angles are obtained. and the interaction force f ext The time series of position x(t) and force F(t) under.

[0124] Preprocessing of data collected under multiple interference conditions, including:

[0125] De-noising: Use bandpass filtering technology to filter and reduce high-frequency noise and low-frequency operation noise in the signal, and retain the effective intermediate frequency signal;

[0126] Numerical Derivation: Derivation of velocity and acceleration from displacement data through numerical difference calculation to obtain complete motion information;

[0127] Data synchronization: Use timestamp or signal synchronization technology to calibrate and align multi-sensor data to ensure that all sensor data are synchronized so that each time point corresponds to a data sample;

[0128] Data mean calculation: for the same arm angle and the interaction force f ext The multiple acquisition sequence data are averaged to obtain a unique operation data sequence to obtain accurate motion information.

[0129] The human body impedance model is established and the expression is as follows:

[0130]

[0131] Where: t∈(t0,t e ), t0 is the starting time of the disturbance, t e is the disturbance cancellation moment;

[0132] dx(t)=x(t)-x(t0), dF(t)=F(t)-F(t0);

[0133] All are about q, F is the unit diagonal matrix function, representing mass, damping and stiffness respectively.

[0134] Fitting human body impedance parameters:

[0135] A linear regression model was established and the impedance parameters were obtained by data fitting. dx(t) and dF(t) construct the linear regression equation and organize the equation into a matrix format:

[0136] Y=XΘ+E,

[0137] in:

[0138] Y = dF(t);

[0139]

[0140] Define the loss function of the model:

[0141]

[0142] Minimize the loss function and calculate each set of arm angles and the interaction force f ext The corresponding impedance parameters

[0143]

[0144] Fitting human body impedance parameter surface:

[0145] Using B-spline surface equation Solve the fitting control points of mass, damping and stiffness respectively, obtain the three-dimensional surface fitting equation, and obtain the arm angle Interaction force f ext The surface composed of the impedance parameters describes the different arm angles and the interaction force f ext Impedance characteristics.

[0146] In combination with the above method principle, the experimental process of using the device of the present invention to measure the impedance of a human arm includes six steps: experimental preparation, equipment debugging, data acquisition, data preprocessing, impedance model establishment and human body impedance parameter surface fitting.

[0147] In the experimental preparation steps, include:

[0148] Preparation of tester 3: Select healthy testees without upper limb diseases to ensure the accuracy of the data; the testees need to wear light clothing to avoid affecting the operation; the testees are fixed on a chair with a backrest and a fixing belt to ensure the stability of their torso and prevent relative movement.

[0149] Equipment installation and debugging: The end six-dimensional force sensor 7 is installed at the end of the mechanical arm of the dual-arm humanoid robot 1 and calibrated to ensure measurement accuracy.

[0150] Experimental environment setup: Ensure that the laboratory environment has moderate temperature and humidity, maintain sufficient light, and reduce external interference; check the stability of the laboratory bench and peripheral equipment to ensure that they will not be affected by external factors during operation.

[0151] The equipment debugging steps include:

[0152] Initialization of the dual-arm humanoid robot 1: Initialize the dual-arm humanoid robot 1, including two seven-degree-of-freedom mechanical arms, a two-degree-of-freedom waist and a head, and put them in an initial state; communicate with the robot through a control panel or a computer connection to set dynamic parameters and calibrate sensors.

[0153] Zero-force mode debugging: Ensure that the dual-arm robot operates in zero-force mode so that the person being tested can easily drag the robot arm and follow its movement; perform multiple operation tests to ensure that the zero-force control function of the end six-dimensional force sensor is normal, and record preliminary data to verify the accuracy of the system.

[0154] Safety check: Check the firmness of all mechanical connections and fixtures of the device to ensure that they will not loosen or fall off during the entire experiment; set up safety boundaries and emergency stop buttons so that the experiment can be stopped in time in case of unexpected situations to ensure the safety of personnel.

[0155] The data collection step includes:

[0156] The first stage of non-disturbance operation data collection: the person under test drags the two-arm humanoid robot 1 to complete the normal operation task according to the predetermined task operation process; the end six-dimensional force sensor 7 is used to collect the force at each time point during the operation, while the two-arm humanoid robot 1 measures the operation position, and the head vision system 2 measures the arm angle of the operating human arm; the operation is repeated many times, and different arm angles are recorded and the interaction force f ext Operation data under certain conditions to ensure that the data samples are sufficient.

[0157] The second stage of disturbed operation data collection: setting the external force disturbance function During the operation, the robot applies random transient external force disturbances to simulate emergencies in actual operations; the person being tested continues to follow the predetermined task operation process and drags the dual-arm robot to complete the task; under the influence of interference, sensors and measuring equipment are still used to collect force, position and arm angle data to ensure the integrity of data sampling under interference.

[0158] Data labeling and classification: All collected data are classified according to different arm angles. and the interaction force f ext Marking and classification are carried out for subsequent processing and analysis to ensure the accuracy and completeness of data records and provide a basis for subsequent data processing.

[0159] In the data preprocessing step, include:

[0160] Denoising,perform denoising on each data track to improve the accuracy and reliability of the data.

[0161] Numerical derivation ensures the smoothness and continuity of data during the numerical derivation process, avoiding the impact of mutation points on subsequent modeling.

[0162] Data synchronization ensures that data from multiple sensors are completely synchronized on the time axis to avoid data inconsistency caused by time errors.

[0163] Data mean calculation, for multiple operation data under the same experimental conditions, calculate their mean and obtain a unique operation data sequence, ensuring that the mean calculation process is strict and reasonable and eliminating accidental errors and random noise.

[0164] The impedance model building step includes:

[0165] Linear regression model construction: Based on the preprocessed data, the impedance model equation is constructed using the linear regression method, and the equation is expressed in matrix form for easy calculation and processing.

[0166] Define the loss function and use gradient descent or other optimization algorithms to minimize the loss function and obtain each set of arm angles and the interaction force f ext The corresponding impedance parameters.

[0167] Parameter optimization: Based on the collected data, the model parameters are gradually adjusted and optimized through an iterative optimization process to ensure the accuracy and robustness of the model; the stability and generalization ability of the model parameters are verified using methods such as cross-validation.

[0168] The human body impedance parameter surface fitting step includes:

[0169] According to the impedance parameters obtained, the B-spline surface equation is used to collect the impedance of each arm angle. and the interaction force fext The corresponding mass, damping and stiffness data form a three-dimensional data point set.

[0170] Surface drawing and verification: Use 3D drawing tools or programming languages ​​to visualize the fitted surface and display the mechanical property surface under different conditions. Verify the fitted surface and check its prediction accuracy on unknown data points to ensure the accuracy of the fitting model.

[0171] The above is only a preferred implementation case of the present invention and does not limit the present invention in any form. Although the implementation process of the present invention is described in detail above, for those familiar with the art, they can still modify the technical solutions recorded in the above examples, or replace some of the technical features therein with equivalents. All modifications, equivalent replacements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for measuring human arm impedance, characterized in that: include: Step 1: Design a dual-arm humanoid robot and build its dynamic model. The robot collects the operational data of the interaction between the human arm under test and the robot's dual arms, and marks and classifies the data under different arm angles and interaction forces of the human arm under test. Step 2: preprocessing the operation data by denoising, numerical derivation, data synchronization and mean calculation; Step 3: Use the preprocessed data to establish a human body impedance model through a linear regression method, calculate the impedance parameters based on mass, damping, and stiffness under different arm angles and interaction forces, and minimize the loss function to optimize the model parameters; Step 4: Based on the calculated impedance parameters, the B-spline surface equation is used to fit and draw a three-dimensional surface to show the impedance characteristics under different arm angles and interaction forces.

2. The human arm impedance measurement method according to claim 1, characterized in that: The denoising uses bandpass filtering technology; the numerical derivation derives velocity and acceleration from displacement data through numerical difference calculation to obtain complete motion information; the data synchronization uses timestamp or signal synchronization technology to calibrate and align multi-sensor data to ensure that all sensor data are synchronized so that each time point corresponds to a data sample; the data mean calculation takes the average of multiple collected operation data of the same arm angle and interaction force to obtain a unique operation data sequence.

3. The human arm impedance measurement method according to claim 1, characterized in that: The specific expression for establishing the human body impedance model is as follows: Where: t∈(t0, t e ), t0 is the starting time of applying disturbance to the human arm, t e is the disturbance cancellation time, dx(t) and dF(t) are preprocessed data; dx(t)=x(t)-x(t0), dF(t)=F(t)-F(t0), x(t) and F(t) are the time series of arm position and force; are impedance parameters, representing mass, damping and stiffness respectively, represents the arm angle, f ext represents the interaction force; By using the linear regression method, formula (1) is organized into a matrix format: Y=XΘ+E, in: Y = dF(t); 4. The method for measuring human arm impedance according to claim 3, characterized in that: The loss function expression of the human body impedance model is: By minimizing the loss function, each set of arm angles is calculated and the interaction force f ext The corresponding impedance parameters.

5. The method for measuring human arm impedance according to claim 4, characterized in that: The step 4 is specifically as follows: according to the impedance parameters obtained, the B-spline surface equation is used to solve and collect each arm angle and the interaction force f ext The corresponding mass, damping and stiffness data form a three-dimensional data point set, and the arm angle Interaction force f ext The surface composed of impedance parameters describes different arm angles and the interaction force f ext Impedance characteristics.

6. A device using the human arm impedance measurement method according to any one of claims 1 to 5, characterized in that: include: A two-arm humanoid robot, the robot comprising: a two-degree-of-freedom waist for adjusting posture; Two left and right seven-degree-of-freedom robotic arms, used for interactive movement with human arms; A two-degree-of-freedom head including a vision system for real-time detection of the arm angle of a human arm; Seat and fixing belt: used to fix the test person; the person being tested sits on a seat with a backrest, the body trunk is fixed by the fixing belt, and the seat is connected to the robot base, so that the body trunk and the two-arm humanoid robot do not move relative to each other during the measurement process; Wrist fixture: connected to the end of the seven-degree-of-freedom robotic arm, used to connect and fix the human arm; End six-dimensional force sensor: It is set at the end of two seven-degree-of-freedom robotic arms and connected to the tester's arm through a wrist fixture to obtain the force between the human arm and the robotic arm. Controller: Installed in the body of the dual-arm humanoid robot, it collects sensor data through the EtherCAT high-speed bus and is used to control the movement of the dual-arm humanoid robot, process the collected data, and calculate the impedance of the human arm.

7. The device according to claim 6, characterized in that The controller specifically includes: A data acquisition module is used to collect the force, position and arm angle data at each time point during the operation and store the data; Data preprocessing module, used for denoising, numerical derivation, data synchronization and mean calculation of collected data; Impedance analysis and modeling module, which is used to establish a linear regression model based on mass, damping and stiffness, and calculate the impedance parameters under different arm angles and interaction forces; The surface fitting module is used to perform three-dimensional surface fitting based on the impedance parameter data using the B-spline surface equation to display the characteristic surface of the arm angle and interaction force and impedance parameters of the human arm.

Images