Mechanical arm joint servo module torque compensation control method and system

Abstract

The invention discloses a mechanical arm joint servo module torque compensation control method and system, and the method comprises the steps: S1, building a dynamic model of a mechanical arm joint servo module, and obtaining an expected torque, S2, comparing the expected torque with an actual torque monitored in real time, and obtaining a torque fluctuation value, S3, automatically compensating the recognized torque fluctuation value through a PID control algorithm, and carrying out the compensation control of the mechanical arm joint servo module. Adjusting the actual output torque of the joint to be close to the expected torque; according to the method, a dynamic model containing nonlinear factors and a real-time monitoring system are established, accurate prediction and compensation of the torque fluctuation of the mechanical arm joint servo module are achieved, empirical mode decomposition and Hilbert transformation are utilized, the system deeply analyzes the time-frequency characteristics of the torque fluctuation, key physical phenomena such as friction and back clearance are recognized, and therefore the torque fluctuation of the mechanical arm joint servo module is accurately predicted and compensated. Based on the analysis, the system dynamically generates an adjustment control instruction, it is ensured that the mechanical arm joint accurately follows the preset track, and the control precision, stability and reliability of the mechanical arm are improved.

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 the field of modern industrial automation and robotics applications, the control accuracy of the robot arm joint servo module is directly related to the performance and operation quality of the entire system. However, due to the influence of factors such as nonlinear friction, backlash, and load changes in the mechanical structure, the joint servo module is prone to torque fluctuations when performing precision operations, resulting in inaccurate positioning, unstable movements, and other problems. Especially for application scenarios that require high-precision force control, such as assembly, grinding, and handling lightweight and fragile objects, traditional control methods are difficult to achieve accurate torque compensation and control, which seriously affects the efficiency and reliability of the operation. 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, the model takes into account influencing factors such as nonlinear friction, backlash effect, load change, etc., and provides a theoretical basis for torque compensation; integrate high-precision torque sensors at the joints to monitor the difference between the actual output torque of the joints and the expected torque in real time to ensure the high accuracy and real-time nature of the feedback information; dynamically adjust the control instructions according to the joint dynamic model and the torque difference fed back in real time to achieve uncertainty in friction, load disturbances, etc.

[0004] The purpose of the present invention can be achieved through the following technical solutions: The present application provides a method for controlling torque compensation of a robot arm joint servo module, comprising: 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; S2. Integrate high-precision torque sensors at key positions of 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 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; S3. According to the control adjustment instruction, the PID control algorithm is used to automatically compensate for the identified torque fluctuation value, adjust the actual output torque of the joint to be close to the expected torque, and send the adjusted control instruction to the servo driver in real time to drive the robot arm joint to move according to the expected trajectory.

[0005] Furthermore, 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.

[0006] Furthermore, the dynamic model of the robot arm joint servo module includes a nonlinear friction model, a backlash model and a load change model. 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 speed, is the Stribeck speed, 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 saturating function; 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.

[0007] Furthermore, the torque generated by the motor and the dynamic model of the robot joint servo module are integrated, and the dynamic equation of the robot arm is established 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: ; Substitute the control input vector u into the dynamic equation and solve it to obtain the desired torque.

[0008] Further, the IMFs obtained by analysis 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 S21-S23 are repeated until the IMF is extracted when the conditions are met; 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 satisfies the IMF condition, wherein the remaining signal is used as the residual term of the decomposition. Each IMF obtained by the decomposition is analyzed to identify the physical meaning and time scale, and dynamically adjusted control instructions are generated according to the analysis results of the IMFs to compensate for the torque fluctuation.

[0009] Furthermore, dynamically adjusted control instructions are generated, including: S221, perform Hilbert transform on each IMF, obtain instantaneous frequency and amplitude, and perform time-frequency analysis; S222, analyzing the frequency characteristics and vibration intensity of the signal at different time points by acquiring the instantaneous frequency and amplitude of each IMF; 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; S224, matching the features of the IMFs with the physical characteristics of the robot arm, identifying the physical phenomenon represented by each IMF, and generating control adjustment instructions based on the results of feature extraction and physical phenomenon identification.

[0010] Furthermore, the compensation torque is calculated and the control command 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: ; 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 the torque fluctuation so that the actual output torque is close to the expected torque.

[0011] A mechanical arm joint servo module torque compensation control system, including a data acquisition and dynamic modeling module, a torque monitoring and fluctuation analysis module and a 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.

[0012] The beneficial effects of the present invention are: 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. Then, the empirical mode decomposition technique is used to decompose the torque fluctuation signal, identify and quantify the fluctuation characteristics on different time scales, and perform time-frequency analysis through Hilbert transform to deeply understand the internal mechanism of torque fluctuation, extract diagnostic features from the decomposed intrinsic mode function, and identify the physical phenomena represented by each IMF, such as friction, backlash, load change, etc. According to 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 robot 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, grinding, and handling of lightweight and fragile objects. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] For better understanding and implementation, the technical solution of the present application is described in detail below with reference to the accompanying drawings.

[0014] Figure 1 A schematic flow chart of a torque compensation control method for a robot arm joint servo module provided in Embodiment 1 of the present application; 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 the first embodiment of the present application; 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 Embodiment 1 of the present application; Figure 4 This is a structural schematic diagram of a torque compensation control system of a robotic arm joint servo module provided in Example 2 of the present application. DETAILED DESCRIPTION

[0015] In order to further explain the technical means and effects taken by the present invention to achieve the predetermined invention purpose, exemplary embodiments will be described in detail here, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are only examples of methods and systems consistent with some aspects of the present application as detailed in the attached claims.

[0016] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0017] The specific implementation methods, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0018] Embodiment 1 See also Figure 1-Figure 3 This embodiment provides a torque compensation control method and system for a robot arm joint servo module, establishes a dynamic model of the robot 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, and realizes the uncertainty of friction, load disturbance, etc.

[0019] The present invention provides a method for controlling torque compensation of a mechanical arm joint servo module, comprising the following steps: S1. Collect the 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. The 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; The specific dynamic model of the robot arm joint servo module includes: Collect the 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 robot arm, use the DH parameter method to establish a kinematic model, define the position and direction of each joint, achieve precise control of the robot arm joints, and ensure that the robot arm moves along the predetermined trajectory.

[0020] Among them, the motor parameters include the torque constant, resistance, inductance, etc. of the motor, which have a direct impact on the response speed and output torque of the motor; 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, which affects the dynamic characteristics and carrying capacity of the robotic arm.

[0021] The 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.

[0022] Furthermore, the dynamic model of the robot arm joint servo module includes a nonlinear friction model, a backlash model and a load change model. 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: ; in, and represent the static and dynamic friction coefficients, respectively, It is positive pressure. is the coefficient of the Stribeck effect, Joint speed, is the Stribeck speed, 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.

[0023] 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.

[0024] Furthermore, the torque generated by the motor and the dynamic model of the robot joint servo module are integrated, and factors such as nonlinear friction, backlash and load change are considered. Then, the dynamic equation of the robot arm is established through the Newton-Raphson iteration method: ; 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.

[0025] 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: ; Substitute the control input vector u into the dynamic equation and solve it to obtain the desired torque.

[0026] Specifically, the expected torque calculated by the dynamic model of the robot joint servo module integrates multiple factors such as motor characteristics, friction model, backlash effect, and load changes. In the calculation process, according to the dynamic equation of the robot, the inertia matrix, centrifugal force and Coriolis force, gravity vector and other dynamic parameters are combined with the control input vector; then the expected torque is calculated by integrating the torque generated by the motor model and the theoretical torque required by the robot joint under specific operating conditions. This expected torque not only takes into account the influence of nonlinear friction (including static and dynamic friction) and backlash, but also takes into account the changes in load mass, center of mass position and moment of inertia, making the final expected torque more accurate.

[0027] S2. Integrate high-precision torque sensors at key positions of 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 torque and the expected torque, and obtain the torque fluctuation value; Furthermore, the moment fluctuation value is processed by EMD to decompose it into multiple IMFs, which are intrinsic mode functions, to identify and quantify the fluctuation characteristics on different time scales. Analyze the IMFs obtained by decomposition, extract features that are helpful for diagnosis, and generate dynamic adjustment control instructions based on the EMD analysis results. Specifically: S211, obtaining the torque fluctuation value, preprocessing the torque fluctuation value, and finding all extreme value points of the torque fluctuation value sequence; S212. Use extreme points to construct upper and lower limit functions, use cubic spline interpolation, calculate the local average of the upper and lower limit functions, and use them 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. 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 maximum and minimum values ​​is equal in the entire data set, and the average value of the envelope is zero), where the original signal refers to the torque fluctuation value sequence, that is, the sequence of torque data measured during the operation of the robot arm joint servo module over time; S214. If the IMF candidate meets the conditions, it is regarded as the first IMF; if not, steps S21-S23 are repeated until the conditions are met; Furthermore, the extracted IMF is subtracted from the original signal, and the above process is repeated for the remaining signal, and the next IMF is extracted, and the above process is repeated until the remaining signal becomes monotonic or no longer satisfies the IMF condition. At this time, the remaining signal serves as the residual term of the decomposition; wherein 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, and dynamically adjusted control instructions are generated according to the analysis results of the IMFs. These instructions will be used to compensate for the torque fluctuation.

[0028] Further dynamically adjusted control instructions are generated, including: S221, perform Hilbert transform on each IMF to obtain instantaneous frequency and amplitude, so as to perform time-frequency analysis. Specifically, each IMF is subjected to Hilbert transform to obtain an analytical signal, which is expressed as: ; 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. , yes The phase of , where a(t) represents the instantaneous amplitude and w(t) represents the instantaneous frequency.

[0029] Through time-frequency analysis, it is possible to distinguish and identify fluctuations in the torque signal caused by different sources, such as friction, backlash, and load changes. This precise identification can help determine the specific impact of each factor on the torque fluctuation, and provide 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.

[0030] S222. By obtaining the instantaneous frequency and amplitude of each IMF, the frequency characteristics and vibration intensity of the signal at different time points can be analyzed, which helps to understand the time-varying characteristics and dynamic behavior of the signal. In the torque signal analysis of the robot arm joint servo module, the internal mechanism of torque fluctuation, such as friction, backlash, load change, etc., is revealed, providing a basis for precise control and compensation.

[0031] S223, calculating the energy of each IMF by squaring the magnitude of the IMF and multiplying it by the corresponding instantaneous frequency, and then integrating the result over time. The energy distribution helps to identify which IMFs have a significant impact on the torque fluctuation; S224, matching the features of the IMFs with the physical characteristics of the robot arm, identifying the physical phenomenon represented by each IMF, such as friction, backlash, external impact or load change; and generating control adjustment instructions based on the results of feature extraction and physical phenomenon identification.

[0032] It should be explained that empirical mode decomposition (EMD) is a method for nonlinear and non-stationary signal analysis. It is suitable for processing complex signals, such as the torque signal of the servo module of the robot joint. 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 effect or load changes.

[0033] By decomposing the complex torque fluctuation signal into a series of intrinsic mode functions (IMFs), we can not only analyze the frequency components and energy distribution of the torque fluctuation in detail, but also accurately identify the periodic and non-periodic changes in the signal, providing key input information for the control algorithm. Finally, the control algorithm is dynamically adjusted based on the input information, which can achieve accurate compensation for the torque fluctuation, significantly improve the control accuracy and stability of the robot arm, and ensure that it moves stably and efficiently according to the predetermined trajectory.

[0034] Among them, the adaptability of the EMD method does not require a preset basis function, but directly extracts intrinsic mode functions (IMFs) from the signal itself, which enables it to handle various complex, nonlinear and non-stationary signals, while also allowing the study of torque fluctuation characteristics on different time scales.

[0035] S3. According to the control adjustment instruction, the identified torque fluctuation value is automatically compensated by the control algorithm, and the actual output torque of the joint is adjusted to be close to the expected torque to compensate for the torque fluctuation caused by factors such as nonlinear friction, backlash effect, and load change. The adjusted control instruction is sent to the servo driver in real time to drive the robot arm joint to move according to the expected trajectory.

[0036] Furthermore, the compensation torque is calculated and the control command is generated according to the current torque fluctuation value, and the error signal, that is, the error signal value, 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, which is ; The final torque command is sent to the servo drive, which adjusts the motor output accordingly to compensate for torque fluctuations.

[0037] This embodiment collects initial data of the servo module, including motor parameters, gear ratio, connecting rod size and mass distribution, and establishes a dynamic model that comprehensively considers kinematic and dynamic characteristics. The model can accurately predict the joint torque required under a specific motion trajectory, that is, the expected torque. Furthermore, by considering factors such as motor characteristics, nonlinear friction, backlash effect and load change, the calculation of the expected torque is more accurate, providing a theoretical basis for the precise control of the robot arm. In practical applications, high-precision torque sensors are integrated at the 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 on different time scales. The Hilbert transform is used to perform time-frequency analysis on the IMFs to deeply understand the intrinsic mechanism of torque fluctuation, such as friction, backlash and load changes.

[0038] Combining feature extraction and physical phenomenon recognition, the control system can dynamically generate adjustment control instructions, which 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.

[0039] 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 that require high-precision operations.

[0040] Embodiment 2 This embodiment is applied to a mechanical arm joint servo module torque compensation control method in embodiment 1, such as Figure 4 As shown, it is specifically a mechanical arm joint servo module torque compensation control system, including: 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 robot arm, and combining the dynamic characteristics such as inertia matrix, centrifugal force and Coriolis force terms, and gravity vector to establish the dynamic model of the robot arm.

[0041] 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.

[0042] The control instruction generation and optimization module extracts features that are helpful for diagnosis from IMFs and identifies the physical phenomena represented by each IMF, such as friction, backlash, load change, etc. Based on the results of feature extraction and physical phenomenon identification, it generates control adjustment instructions, applies the PID control algorithm, dynamically adjusts the control parameters, calculates the compensation torque, generates the final control instructions, sends the control instructions to the servo drive in real time, executes the compensation action, and adjusts the actual output torque of the robot arm joint.

[0043] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. 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: Establish a 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; 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; 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 expected torque, and sends the adjusted control instructions to the servo driver 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 dynamic model of the robot arm joint servo module also includes a nonlinear friction model, a backlash model and a load change model. 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 speed, 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 saturating function; 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 2, characterized in that: Integrate the torque generated by the motor and the dynamic model of the robot joint servo module, and establish the dynamic equation of the robot 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: ； Substitute the control input vector u into the dynamic equation and solve it to obtain the desired torque.

6. 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: Obtain the torque fluctuation value, pre-process the torque fluctuation value, and find all extreme value points of the torque fluctuation value sequence; The upper and lower limit functions are constructed using extreme points, and cubic spline interpolation is used to calculate the local average of the upper and lower limit functions as the envelope of the signal; Subtract the envelope from the original signal to obtain the first IMF candidate and check whether it meets the IMF conditions; When the IMF candidate meets the conditions, it is regarded as the first IMF; when the conditions are not met, steps S21-S23 are repeated until the IMF is extracted when the conditions are met.

7. A method for controlling torque compensation of a robot arm joint servo module according to claim 6, 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.

8. A method for controlling torque compensation of a robot arm joint servo module according to claim 7, 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.

9. A method for controlling torque compensation of a robot arm joint servo module according to claim 6, 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.

10. 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 9, 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