A torque compensation control method and system for a mechanical arm joint servo module

Abstract

The present invention discloses a torque compensation control method and system for a robot arm joint servo module, comprising S1, establishing a dynamic model of the robot arm joint servo module to obtain a desired torque, S2, comparing the desired torque with the actual torque monitored in real time to obtain a torque fluctuation value, and S3, automatically compensating the identified torque fluctuation value through a PID control algorithm to adjust the actual output torque of the joint to be close to the desired torque; the present invention realizes accurate prediction and compensation of the torque fluctuation of the robot arm joint servo module by establishing a dynamic model and a real-time monitoring system containing nonlinear factors, and uses empirical mode decomposition and Hilbert transform to systematically analyze the time-frequency characteristics of the torque fluctuation and identify key physical phenomena such as friction and backlash. Based on these analyses, the system dynamically generates adjustment control instructions to ensure that the robot arm joints accurately follow the predetermined trajectory, thereby improving the control accuracy, stability and reliability of the robot arm.

Description

Technical Field

[0001] The present invention relates to the technical field of robot control, and in particular to a torque compensation control method and system for a robot arm joint servo module. Background Art

[0002] In modern industrial automation and robotics, the control accuracy of a robot arm's joint servo module is directly related to the overall system performance and operational quality. However, due to factors such as nonlinear friction, backlash, and load variations in the mechanical structure, the joint servo module is prone to torque fluctuations when performing precision operations, leading to inaccurate positioning and unstable movements. This is particularly true for applications requiring high-precision force control, such as assembly, polishing, and handling lightweight and fragile objects. Traditional control methods struggle to achieve precise torque compensation and control, severely impacting operational efficiency and reliability. Summary of the Invention

[0003] The purpose of the present invention is to provide a torque compensation control method and system for a robotic arm joint servo module, establish an accurate dynamic model of the robotic arm joint servo module, and take into account influencing factors such as nonlinear friction, backlash effect, and load change, providing a theoretical basis for torque compensation; integrate high-precision torque sensors at the joints to monitor the difference between the actual output torque and the expected torque of the joints in real time, ensuring the high precision and real-time nature of the feedback information; and dynamically adjust the control instructions based on the joint dynamic model and the torque difference fed back in real time.

[0004] The purpose of the present invention can be achieved through the following technical solutions:

[0005] The present application provides a method for controlling torque compensation of a robot arm joint servo module, comprising:

[0006] S1. Collect the initial data of the robot arm joint servo module, combine the kinematic model and the dynamic characteristics of the robot arm, establish the dynamic model of the robot arm joint servo module, and obtain the expected torque by calculating the joint torque required under a specific motion trajectory;

[0007] S2. Integrate high-precision torque sensors at key locations on each joint to monitor and record the actual output torque in real time. Compare the expected torque with the actual torque monitored in real time, calculate the difference between the actual and expected torques, and obtain the torque fluctuation value.

[0008] Among them, the moment fluctuation value is processed by EMD and decomposed into multiple IMFs to identify and quantify the fluctuation characteristics on different time scales;

[0009] S3. Generate control instructions based on IMFS. According to the control adjustment instructions, the PID control algorithm automatically compensates for the identified torque fluctuation value, adjusts the actual output torque of the joint to be close to the desired torque, and sends the adjusted control instructions to the servo driver in real time to drive the robotic arm joint to move according to the desired trajectory.

[0010] Furthermore, the dynamic model of the robotic arm joint servo module includes:

[0011] Collect initial data of the servo module, including the motor's torque constant, resistance, inductance, gear ratio, connecting rod length, joint and connecting rod mass distribution, center of mass position, and moment of inertia;

[0012] Determine the joint structure and kinematic constraints of the robotic arm, establish a kinematic model using the DH parameter method, and define the position and orientation of each joint;

[0013] The specific torque generated by the motor is expressed as:

[0014] ;

[0015] Where E is the back electromotive force, which is proportional to the angular velocity ω of the motor, R is the resistance of the motor, and L is the inductance of the motor. It is the rate of change of current over time, indicating the dynamic change of current.

[0016] Furthermore, the dynamic model of the robot arm joint servo module includes a nonlinear friction model, a backlash model and a load change model.

[0017] The nonlinear friction model is used to describe the friction force in joint motion. It is related to the static friction coefficient, dynamic friction coefficient and Stribeck velocity. The friction torque is expressed as:

[0018] ;

[0019] in, and represent the static and dynamic friction coefficients, respectively, It is positive pressure. is the coefficient of the Stribeck effect, Joint velocity or angular velocity, is the Stribeck speed,

[0020] The backlash model is used to describe the gap effect of the joint during movement, indicating the nonlinear behavior at the start and stop of movement. The additional torque caused by the backlash is expressed as:

[0021] ;

[0022] in, is the backlash stiffness, which indicates the system's sensitivity to backlash, sat( ) is a saturation function that describes the effect of backlash when the joint velocity exceeds a certain threshold;

[0023] The load change model is used to describe the changes in dynamic behavior caused by load changes, including changes in load mass, center of mass position, and moment of inertia. The load torque is expressed as:

[0024] ;

[0025] in, is the moment of inertia of the load, is the load mass, g is the acceleration due to gravity, is the distance from the load center of mass to the joint, is the angular acceleration of the joint.

[0026] Furthermore, the torque generated by the motor and the dynamic model of the robot arm joint servo module are integrated, and the dynamic equation of the robot arm is established through the Newton-Raphson iteration method:

[0027] ;

[0028] Where M is the inertia matrix, C is the centrifugal force and Coriolis force terms, G is the gravity vector, u is the control input vector, and q is the joint variable vector;

[0029] The torque obtained by the dynamic model of the robot arm joint servo module and the torque generated by the motor are integrated into the control input vector u, which is specifically expressed as:

[0030] ;

[0031] in, is the friction torque, is the backlash torque, is the load torque, substitute the control input vector u into the dynamic equation and solve it to get the desired torque.

[0032] Further analysis of the decomposed IMFs includes:

[0033] S211, obtaining torque fluctuation values, preprocessing the torque fluctuation values, and finding all extreme value points of the torque fluctuation value sequence;

[0034] S212. Construct upper and lower limit functions using extreme points, use cubic spline interpolation, and calculate the local average of the upper and lower limit functions as the envelope of the signal;

[0035] S213, subtracting the envelope from the original signal to obtain the first IMF candidate, and checking whether the IMF conditions are met;

[0036] S214. When the IMF candidate meets the conditions, it is regarded as the first IMF; when the conditions are not met, steps S211-S213 are repeated until the IMF is extracted when the conditions are met;

[0037] Furthermore, the extracted IMF is subtracted from the original signal, and the extraction is repeated on the remaining signal until the remaining signal becomes monotonic or no longer meets the IMF conditions. The remaining signal is used as the residual term of the decomposition. Each IMF obtained by decomposition is analyzed to identify the physical meaning and time scale. Based on the analysis results of the IMFs, dynamically adjusted control instructions are generated to compensate for torque fluctuations.

[0038] Furthermore, dynamically adjusted control instructions are generated, including:

[0039] S221, perform Hilbert transform on each IMF to obtain instantaneous frequency and amplitude, and perform time-frequency analysis;

[0040] S222. Analyze the frequency characteristics and vibration intensity of the signal at different time points by obtaining the instantaneous frequency and amplitude of each IMF;

[0041] S223, calculating the energy of each IMF by squaring the magnitude of the IMF and multiplying it by the corresponding instantaneous frequency, and time-integrating the result;

[0042] S224. Match the features of the IMFs with the physical characteristics of the robotic arm, identify the physical phenomenon represented by each IMF, and generate control adjustment instructions based on the results of feature extraction and physical phenomenon identification.

[0043] Furthermore, the compensation torque is calculated and the control instruction is generated according to the current torque fluctuation value, the error signal is determined, and the output is controlled by the PID control algorithm, which is specifically expressed as:

[0044] ;

[0045] in, is the proportional gain, is the integral gain, is the differential gain;

[0046] Through PID control output, the compensation torque is calculated and added to the actual torque to reduce the error, which is expressed as

[0047] , the final torque command is the sum of the actual measured torque and the compensation torque, expressed as ;The final torque command is sent to the servo driver, and the driver adjusts the motor output according to the command to compensate for torque fluctuations so that the actual output torque is close to the expected torque.

[0048] A torque compensation control system for a robotic arm joint servo module includes a data acquisition and dynamic modeling module, a torque monitoring and fluctuation analysis module, and a control instruction generation and optimization module.

[0049] The data acquisition and dynamic modeling module is used to collect the initial data of the servo module, establish the kinematic model of the robot arm using the DH parameter method, and establish the dynamic model of the robot arm in combination with the dynamic characteristics;

[0050] The torque monitoring and fluctuation analysis module monitors and records the actual output torque, compares the expected torque with the actual torque monitored in real time, calculates the torque fluctuation value, performs empirical mode decomposition on the torque fluctuation value to decompose it into multiple intrinsic mode functions, and performs Hilbert transform to obtain instantaneous frequency and amplitude for time-frequency analysis;

[0051] The control instruction generation and optimization module extracts diagnostic features from IMFs, identifies the physical phenomenon represented by each IMF, generates control adjustment instructions based on the results of feature extraction and physical phenomenon identification, applies the PID control algorithm, dynamically adjusts control parameters, calculates compensation torque, generates final control instructions, and sends the control instructions to the servo drive in real time to execute compensation actions and adjust the actual output torque of the robot arm joint.

[0052] The beneficial effects of the present invention are:

[0053] By collecting the initial data of the servo module and establishing a dynamic model, including motor characteristics, nonlinear friction model, backlash model and load change model, the expected torque under specific operating conditions can be accurately predicted. The actual output torque is then monitored in real time using a high-precision torque sensor and compared with the expected torque to calculate the torque fluctuation value, providing a basis for compensation.

[0054] Empirical mode decomposition (EMD) is then used to decompose the torque fluctuation signal, identifying and quantifying the fluctuation characteristics at different time scales. Time-frequency analysis is then performed using the Hilbert transform to gain a deeper understanding of the inherent mechanism of torque fluctuations. Diagnostic features are extracted from the decomposed intrinsic mode functions (IMFs), and the physical phenomena represented by each IMF, such as friction, backlash, and load variation, are identified.

[0055] Based on the results of feature extraction and physical phenomenon identification, combined with the prediction model and PID control algorithm, control instructions are dynamically generated to achieve accurate compensation for torque fluctuations, and the adjusted control instructions are sent to the servo drive in real time. This method can effectively solve the torque fluctuation problem faced by the robotic arm joint servo module in precision operations, significantly improve the control accuracy and stability, and is particularly suitable for scenarios that require high-precision force control, such as assembly, polishing, and handling lightweight and fragile objects. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] For better understanding and implementation, the technical solution of the present application is described in detail below with reference to the accompanying drawings.

[0057] Figure 1 A schematic flow chart of a torque compensation control method for a robotic arm joint servo module provided in Example 1 of the present application;

[0058] Figure 2 A schematic flow chart of IMFs obtained by analysis and decomposition in a torque compensation control method for a robot arm joint servo module provided in Example 1 of the present application;

[0059] Figure 3 A schematic diagram of a flow chart of generating dynamically adjusted control instructions for a torque compensation control method for a robot arm joint servo module provided in Example 1 of the present application;

[0060] Figure 4 This is a structural diagram of a torque compensation control system of a robotic arm joint servo module provided in Example 2 of the present application. DETAILED DESCRIPTION

[0061] To further illustrate the technical means and effectiveness of the present invention in achieving its intended purpose, exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all possible embodiments consistent with the present application. Rather, they are merely examples of methods and systems consistent with certain aspects of the present application, as detailed in the appended claims.

[0062] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. As used in this application and the appended claims, the singular forms "a," "an," "the," and "the" are intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0063] The following describes in detail the specific implementation methods, features and effects of the present invention in conjunction with the accompanying drawings and preferred embodiments.

[0064] Example 1

[0065] See also Figure 1-Figure 3 This embodiment provides a torque compensation control method and system for a robotic arm joint servo module, establishes a dynamic model of the robotic arm joint servo module, and takes into account the influencing factors of nonlinear friction, backlash effect, and load change, providing a theoretical basis for torque compensation; integrates a high-precision torque sensor at the joint to monitor the difference between the actual output torque and the expected torque of the joint in real time, ensuring the high precision and real-time nature of the feedback information; dynamically adjusts the control instructions according to the joint dynamic model and the torque difference fed back in real time.

[0066] The present invention provides a method for controlling torque compensation of a robot arm joint servo module, comprising the following steps:

[0067] S1. Collect initial data of the robot arm joint servo module and establish a dynamic model of the robot arm joint servo module by combining the kinematic model and the dynamic characteristics of the robot arm. This model can describe the behavior of the joint under different operating conditions and obtain the expected torque by calculating the joint torque required under a specific motion trajectory.

[0068] The specific dynamic model of the robotic arm joint servo module includes:

[0069] Collect initial data of the servo module, including the motor's torque constant, resistance, inductance, gear ratio, connecting rod length, mass distribution of the joints and connecting rods, center of mass position, and moment of inertia; determine the joint structure and kinematic constraints of the robotic arm, use the DH parameter method to establish a kinematic model, define the position and orientation of each joint, achieve precise control of the robotic arm joints, and ensure that the robotic arm moves along the predetermined trajectory.

[0070] Among them, the motor parameters include the motor's torque constant, resistance, inductance, etc., which have a direct impact on the motor's response speed and output torque; the gear ratio is the speed ratio or torque ratio of the gears in the transmission mechanism, which affects the output speed and torque of the joint; the connecting rod length is the size of each connecting rod of the robotic arm, which affects the working range and accuracy of the robotic arm; the mass distribution includes mass, center of mass position and moment of inertia, etc., which affects the dynamic characteristics and load-bearing capacity of the robotic arm.

[0071] The torque generated by the motor is expressed as:

[0072] ;

[0073] Where E is the back electromotive force, which is proportional to the angular velocity ω of the motor, R is the resistance of the motor, and L is the inductance of the motor. It is the rate of change of current over time, indicating the dynamic change of current.

[0074] Furthermore, the dynamic model of the robot arm joint servo module includes a nonlinear friction model, a backlash model and a load change model.

[0075] The nonlinear friction model is used to describe the friction force in joint motion, which usually includes static friction and dynamic friction. Static friction is a constant force independent of speed until a certain threshold is reached. Dynamic friction varies with speed and is related to the static friction coefficient, dynamic friction coefficient and Stribeck speed. The friction torque is expressed as:

[0076] ;

[0077] in, and represent the static and dynamic friction coefficients, respectively, It is positive pressure. is the coefficient of the Stribeck effect, Joint velocity or angular velocity, is the Stribeck speed,

[0078] The backlash model is used to describe the gap effect of the joint during movement, indicating the nonlinear behavior at the start and stop of movement. The additional torque caused by the backlash is expressed as:

[0079] ;

[0080] in, is the backlash stiffness, which indicates the system's sensitivity to backlash, sat( ) is a saturation function that describes the effect of backlash when the joint velocity exceeds a certain threshold.

[0081] The load change model is used to describe the changes in dynamic behavior caused by load changes, including changes in load mass, center of mass position, and moment of inertia. The load torque is expressed as:

[0082] ;

[0083] in, is the moment of inertia of the load, is the load mass, g is the acceleration due to gravity, is the distance from the load center of mass to the joint, is the angular acceleration of the joint.

[0084] Furthermore, the torque generated by the motor and the dynamic model of the robot arm joint servo module are integrated, and factors such as nonlinear friction, backlash, and load variation are considered. Then, the dynamic equation of the robot arm is established using the Newton-Raphson iteration method: ;

[0085] Where M is the inertia matrix, C is the centrifugal and Coriolis force terms, G is the gravity vector, u is the control input vector, and q is the joint variable vector.

[0086] The torque obtained by the dynamic model of the robot arm joint servo module and the torque generated by the motor are integrated into the control input vector u, which is specifically expressed as:

[0087] ;

[0088] in, is the friction torque, is the backlash torque, is the load torque, substitute the control input vector u into the dynamic equation and solve it to get the desired torque.

[0089] Specifically, the expected torque calculated using the dynamic model of the robot arm's joint servo module integrates multiple factors, including motor characteristics, friction model, backlash effects, and load variations. During the calculation process, dynamic parameters such as the inertia matrix, centrifugal and Coriolis force terms, and gravity vector are combined with the control input vector based on the robot's dynamic equations. The expected torque is then calculated by integrating the torque generated by the motor model with the theoretical torque required by the robot arm's joints under specific operating conditions. This expected torque not only accounts for the effects of nonlinear friction (including static and dynamic friction) and backlash, but also considers changes in load mass, center of mass position, and moment of inertia, making the final expected torque more accurate.

[0090] S2. Integrate high-precision torque sensors at key locations on each joint to monitor and record the actual output torque in real time. Compare the expected torque with the actual torque monitored in real time, calculate the difference between the actual and expected torques, and obtain the torque fluctuation value.

[0091] Furthermore, the torque fluctuation value is processed by EMD and decomposed into multiple IMFs. The IMFs are intrinsic mode functions to identify and quantify the fluctuation characteristics on different time scales.

[0092] Analyze the decomposed IMFs, extract features that are helpful for diagnosis, and generate dynamic adjustment control instructions based on the EMD analysis results. Specifically:

[0093] S211, obtaining torque fluctuation values, preprocessing the torque fluctuation values, and finding all extreme value points of the torque fluctuation value sequence;

[0094] S212. Use extreme points to construct upper and lower limit functions, use cubic spline interpolation, and calculate the local average of the upper and lower limit functions as the envelope of the signal; the main function of the envelope is to help identify and separate the intrinsic mode function in the signal.

[0095] S213. Subtract the envelope from the original signal to obtain the first IMF candidate. Check whether the IMF conditions are met (the number of local maxima and minima in the entire data set is equal, and the average value of the envelope is zero). The original signal refers to the sequence of torque fluctuation values, that is, the sequence of torque data measured during the operation of the robot arm joint servo module over time.

[0096] S214. If the IMF candidate meets the conditions, it is considered as the first IMF; if not, repeat steps S211-S213 until the conditions are met;

[0097] Furthermore, the extracted IMF is subtracted from the original signal, and the above process is repeated for the remaining signal. The next IMF is extracted and the above process is repeated until the remaining signal becomes monotonic or no longer meets the IMF conditions. At this time, the remaining signal serves as the residual term of the decomposition; the residual term is the remaining part after subtracting all the extracted IMFs from the original torque fluctuation value, representing the monotonic trend or constant component in the signal that cannot be further decomposed. Each IMF obtained by decomposition is analyzed to identify their physical meaning and time scale. Based on the analysis results of the IMFs, dynamically adjusted control instructions are generated. These instructions will be used to compensate for the torque fluctuation.

[0098] Further generate dynamic adjustment control instructions, including:

[0099] S221, perform Hilbert transform on each IMF to obtain instantaneous frequency and amplitude, and then perform time-frequency analysis.

[0100] Specifically, each IMF is subjected to Hilbert transform to obtain an analytical signal, which is expressed as:

[0101] ;

[0102] Where H[x(t)] is the Hilbert transform of x(t), j is the imaginary unit, and then the instantaneous amplitude and instantaneous frequency of the analytical signal are calculated. ,

[0103] yes where a(t) represents the instantaneous amplitude and w(t) represents the instantaneous frequency.

[0104] Through time-frequency analysis, fluctuations in the torque signal caused by different sources, such as friction, backlash, and load changes, can be distinguished and identified. This precise identification can help determine the specific impact of each factor on torque fluctuations, providing a detailed basis for optimizing the control strategy of the robot arm joint servo module, thereby improving the system's precise control and compensation capabilities.

[0105] S222. By obtaining the instantaneous frequency and amplitude of each IMF, we can analyze the signal's frequency characteristics and vibration intensity at different time points, which helps us understand the signal's time-varying characteristics and dynamic behavior. In the torque signal analysis of the robot arm joint servo module, we can reveal the inherent mechanisms of torque fluctuations, such as friction, backlash, and load changes, providing a basis for precise control and compensation.

[0106] S223. Calculate the energy of each IMF by squaring the magnitude of the IMF and multiplying it by the corresponding instantaneous frequency, and then time-integrating the result. The energy distribution helps identify which IMFs have a significant impact on the torque fluctuation.

[0107] S224. Match the features of the IMFs with the physical characteristics of the robotic arm, and identify the physical phenomenon represented by each IMF, such as friction, backlash, external impact, or load change; and generate control adjustment instructions based on the results of feature extraction and physical phenomenon identification.

[0108] It should be explained that empirical mode decomposition (EMD) is a method used for nonlinear and non-stationary signal analysis. It is suitable for processing complex signals, such as the torque signal of a robotic arm joint servo module. By performing EMD processing on the torque signal, it can be decomposed into a series of modal functions (IMFs). These IMFs represent fluctuations on different time scales, which helps to analyze torque fluctuations caused by nonlinear friction, backlash effects, or load changes.

[0109] By decomposing the complex torque fluctuation signal into a series of intrinsic mode functions (IMFs), we can not only meticulously analyze the frequency components and energy distribution of the torque fluctuations, but also accurately identify periodic and aperiodic variations in the signal, providing critical input information for the control algorithm. Ultimately, dynamically adjusting the control algorithm based on this input information enables precise compensation of torque fluctuations, significantly improving the control accuracy and stability of the robotic arm, ensuring stable and efficient motion along the intended trajectory.

[0110] Among them, the adaptability of the EMD method does not require preset basis functions, but directly extracts intrinsic mode functions (IMFs) from the signal itself, which enables it to process various complex, nonlinear and non-stationary signals, while also allowing the study of torque fluctuation characteristics at different time scales.

[0111] S3. Based on the control adjustment instructions, the control algorithm automatically compensates for the identified torque fluctuation value, adjusts the actual output torque of the joint to be close to the desired torque, and compensates for the torque fluctuation caused by factors such as nonlinear friction, backlash effect, and load change. The adjusted control instructions are sent to the servo driver in real time to drive the robotic arm joint to move according to the desired trajectory.

[0112] Furthermore, the compensation torque is calculated and the control instruction is generated according to the current torque fluctuation value, and the error signal, that is, the error signal value, is determined. The output is controlled by the PID control algorithm, which is specifically expressed as:

[0113] ;

[0114] in, is the proportional gain, is the integral gain, is the differential gain;

[0115] Through PID control output, the compensation torque is calculated and added to the actual torque to reduce the error, which is expressed as

[0116] , the final torque command is the sum of the actual measured torque and the compensation torque, which is ;The final torque command is sent to the servo drive, which adjusts the motor output accordingly to compensate for torque fluctuations.

[0117] This embodiment collects initial data from the servo module, including motor parameters, gear ratios, connecting rod dimensions, and mass distribution, to establish a dynamic model that comprehensively considers both kinematic and dynamic characteristics. This model can accurately predict the joint torque required under a specific motion trajectory, i.e., the expected torque. Furthermore, by considering factors such as motor characteristics, nonlinear friction, backlash effects, and load variations, the calculation of the expected torque is more accurate, providing a theoretical basis for precise control of the robotic arm.

[0118] In practical applications, high-precision torque sensors are integrated at key positions of each joint to monitor and record the actual output torque in real time and compare it with the expected torque to obtain the torque fluctuation value. The empirical mode decomposition (EMD) technology is used to decompose the torque fluctuation value into multiple intrinsic mode functions (IMFs). The IMFs represent the fluctuation characteristics at different time scales. The time-frequency analysis of the IMFs is performed through Hilbert transform to gain a deeper understanding of the intrinsic mechanisms of torque fluctuation, such as friction, backlash, and load changes.

[0119] By combining feature extraction and physical phenomenon recognition, the control system can dynamically generate adjustment control instructions. These instructions will be used to compensate for torque fluctuations. Through the PID control algorithm, the control parameters are dynamically adjusted according to the error signal, the compensation torque is calculated, and added to the actual torque to reduce the error. Finally, the adjusted control instructions are sent to the servo drive in real time to drive the robot arm joints to move according to the desired trajectory.

[0120] This method not only improves the control accuracy of the robotic arm, ensuring its stable and efficient movement along the predetermined trajectory, but also enhances the adaptability of the system, enabling it to adapt to various operating conditions and external disturbances. Through precise torque compensation, the stability and reliability of the robotic arm are significantly improved, thus playing a key role in automated manufacturing, precision assembly and other fields requiring high-precision operations.

[0121] Example 2

[0122] This embodiment is applied to a torque compensation control method of a robot arm joint servo module in embodiment 1, such as Figure 4 As shown, it is specifically a torque compensation control system of a robotic arm joint servo module, including:

[0123] The data acquisition and dynamic modeling module is responsible for collecting the initial data of the servo module, including motor parameters, gear ratio, connecting rod size, mass distribution, etc., and using the DH parameter method to establish the kinematic model of the robotic arm. It also combines dynamic characteristics such as inertia matrix, centrifugal force and Coriolis force terms, and gravity vector to establish the dynamic model of the robotic arm.

[0124] The torque monitoring and fluctuation analysis module monitors and records the actual output torque, compares the expected torque with the actual torque monitored in real time, calculates the torque fluctuation value, performs empirical mode decomposition on the torque fluctuation value, decomposes it into multiple intrinsic mode functions (IMFs), and performs Hilbert transform to obtain the instantaneous frequency and amplitude for time-frequency analysis.

[0125] The control instruction generation and optimization module extracts diagnostic features from IMFs and identifies the physical phenomena represented by each IMF, such as friction, backlash, and load changes. Based on the results of feature extraction and physical phenomenon identification, it generates control adjustment instructions, applies the PID control algorithm, dynamically adjusts control parameters, calculates compensation torque, generates final control instructions, and sends the control instructions to the servo drive in real time to execute compensation actions and adjust the actual output torque of the robot arm joint.

[0126] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as above in terms of a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art can, without departing from the scope of the technical solution of the present invention, make some changes or modifications to equivalent embodiments using the technical contents disclosed above. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A method for torque compensation control of a robot arm joint servo module, characterized in that: The steps include: Establishing a dynamic model of a robot arm joint servo module, wherein the dynamic model of the robot arm joint servo module includes a nonlinear friction model, a backlash model, and a load change model, and obtaining a desired torque by calculating the joint torque required under a specific motion trajectory; Specifically, it includes: integrating the torque generated by the motor and the dynamic model of the robot arm joint servo module, and establishing the dynamic equation of the robot arm through the Newton-Raphson iteration method: ； Where M is the inertia matrix, C is the centrifugal force and Coriolis force terms, G is the gravity vector, u is the control input vector, and q is the joint variable vector; The torque obtained by the dynamic model of the robot joint servo module and the torque generated by the motor are integrated into the control input vector u, which is specifically expressed as: ； in, is the torque generated by the motor, is the friction torque, is the backlash torque, is the load torque, substitute the control input vector u into the dynamic equation and solve it to get the desired torque; Monitor and record the actual output torque in real time, compare the expected torque with the actual torque monitored in real time, calculate the difference between the actual torque and the expected torque, and obtain the torque fluctuation value; Among them, the moment fluctuation value is processed by EMD to decompose it into multiple IMFs to identify and quantify the fluctuation characteristics on different time scales; Based on IMFs, control adjustment instructions are generated. According to the control adjustment instructions, the identified torque fluctuation value is automatically compensated through the PID control algorithm, the actual output torque of the joint is adjusted to be close to the expected torque, and the adjusted control instructions are sent to the servo drive in real time to drive the robot arm joint to move according to the expected trajectory.

2. A method for controlling torque compensation of a robot arm joint servo module according to claim 1, characterized in that: The dynamic model of the robot arm joint servo module includes: Collect initial data of the servo module, including the motor's torque constant, resistance, inductance, gear ratio, connecting rod length, joint and connecting rod mass distribution, center of mass position, and moment of inertia; Determine the joint structure and kinematic constraints of the robot arm, establish a kinematic model using the DH parameter method, and define the position and orientation of each joint; The specific torque generated by the motor is expressed as: ； Where, E is the back electromotive force, which is proportional to the angular velocity ω of the motor, R is the resistance of the motor, L is the inductance of the motor, It is the rate of change of current over time, indicating the dynamic change of current.

3. A method for controlling torque compensation of a robot arm joint servo module according to claim 2, characterized in that: The nonlinear friction model is used to describe the friction force in joint motion, which is related to the static friction coefficient, dynamic friction coefficient and Stribeck velocity. The friction torque is expressed as: ； in, and represent the static and dynamic friction coefficients, respectively, It is positive pressure. is the coefficient of the Stribeck effect, Joint velocity or angular velocity, is the Stribeck speed.

4. A method for controlling torque compensation of a robot arm joint servo module according to claim 3, characterized in that: The backlash model is used to describe the gap effect of the joint during movement, indicating the nonlinear behavior at the start and stop of the movement. The additional torque caused by the backlash is expressed as: ； in, is the backlash stiffness, which indicates the system's sensitivity to backlash, sat( ) is a saturation function that describes the effect of backlash when the joint velocity exceeds a certain threshold; The load change model is used to describe the change in dynamic behavior caused by load change, including changes in load mass, center of mass position and moment of inertia. The load torque is expressed as: ； in, is the moment of inertia of the load, is the load mass, g is the acceleration due to gravity, is the distance from the load center of mass to the joint, is the angular acceleration of the joint.

5. A method for controlling torque compensation of a robot arm joint servo module according to claim 1, characterized in that: The IMFs obtained through analysis and decomposition include: S211, obtaining the torque fluctuation value, preprocessing the torque fluctuation value, and finding all extreme value points of the torque fluctuation value sequence; S212, constructing upper and lower limit functions using extreme value points, using cubic spline interpolation, and calculating the local average values ​​of the upper and lower limit functions as the envelope of the signal; S213, subtracting the envelope from the original signal to obtain the first IMF candidate, and checking whether the IMF condition is met; S214. When the IMF candidate meets the conditions, it is regarded as the first IMF; when the conditions are not met, steps S211-S213 are repeated until the IMF is extracted when the conditions are met.

6. A method for controlling torque compensation of a robot arm joint servo module according to claim 5, characterized in that: The extracted IMF is subtracted from the original signal, and the extraction is repeated on the remaining signal until the remaining signal becomes monotonic or no longer meets the IMF condition, where the remaining signal is used as the residual term of the decomposition. Each IMF obtained by decomposition is analyzed to identify the physical meaning and time scale. Based on the analysis results of the IMFs, dynamically adjusted control instructions are generated to compensate for torque fluctuations.

7. A method for controlling torque compensation of a robot arm joint servo module according to claim 6, characterized in that: The generating of the dynamically adjusted control instruction comprises: Perform Hilbert transform on each IMF to obtain instantaneous frequency and amplitude and perform time-frequency analysis; By obtaining the instantaneous frequency and amplitude of each IMF, the frequency characteristics and vibration intensity of the signal at different time points are analyzed; Calculate the energy of each IMF by squaring the magnitude of the IMF and multiplying it by the corresponding instantaneous frequency, integrating the result over time; The features of IMFs are matched with the physical characteristics of the robot arm, the physical phenomenon represented by each IMF is identified, and control adjustment instructions are generated based on the results of feature extraction and physical phenomenon identification.

8. A method for controlling torque compensation of a robot arm joint servo module according to claim 5, characterized in that: According to the current torque fluctuation value, the compensation torque is calculated and the control instruction is generated, the error signal is determined, and the output is controlled by the PID control algorithm, which is specifically expressed as: ； in, is the proportional gain, is the integral gain, is the differential gain; Through the PID control output, the compensation torque is calculated and added to the actual torque to reduce the error, which is expressed as , the final torque command is the sum of the actual measured torque and the compensation torque, expressed as ; The final torque command is sent to the servo driver, and the driver adjusts the motor output according to the command to compensate for torque fluctuations.

9. A mechanical arm joint servo module torque compensation control system, applied to a mechanical arm joint servo module torque compensation control method as claimed in any one of claims 1 to 8, characterized in that: It includes data acquisition and dynamic modeling module, torque monitoring and fluctuation analysis module and control instruction generation and optimization module. The data acquisition and dynamic modeling module is used to collect the initial data of the servo module, establish the kinematic model of the robot arm using the DH parameter method, and establish the dynamic model of the robot arm in combination with the dynamic characteristics; The torque monitoring and fluctuation analysis module monitors and records the actual output torque, compares the expected torque with the actual torque monitored in real time, calculates the torque fluctuation value, performs empirical mode decomposition on the torque fluctuation value, decomposes it into multiple intrinsic mode functions, performs Hilbert transform to obtain instantaneous frequency and amplitude, and performs time-frequency analysis; The control instruction generation and optimization module extracts features that are helpful for diagnosis from IMFs, identifies the physical phenomenon represented by each IMF, generates control adjustment instructions based on the results of feature extraction and physical phenomenon identification, applies the PID control algorithm, dynamically adjusts control parameters, calculates the compensation torque, generates the final control instruction, sends the control instruction to the servo driver in real time, executes the compensation action, and adjusts the actual output torque of the robot arm joint.

Images