Industrial robot and control method, device, storage medium and program product thereof

Abstract

The application discloses a control method and device of an industrial robot, the industrial robot, a storage medium and a computer program product. The method comprises the following steps: designing an observer module, and setting the observer module in a feedback part of a servo control system; under the condition that the industrial robot is running, observing, by the observer module, based on a speed feedback signal of a servo motor and a torque feedback signal of a harmonic reducer, to obtain a torque disturbance signal of the industrial robot; and compensating the torque disturbance signal of the industrial robot to torque control of the harmonic reducer by the servo motor, so as to suppress the end vibration of the industrial robot. According to the scheme, the adaptive reduced dimension state observer is set in the feedback part of the servo control system of the industrial robot, the fluctuating torque signal observed by the adaptive reduced dimension state observer is compensated to the torque control of the harmonic reducer by the servo motor, the vibration caused by the harmonic reducer is suppressed, and the dynamic performance of the industrial robot is provided.

Description

Technical Field

[0001] This invention belongs to the field of industrial robot technology, specifically relating to a control method, device, industrial robot, storage medium, and computer program product for an industrial robot, and particularly to a vibration suppression method, device, industrial robot, storage medium, and computer program product based on an adaptive dimensionality reduction state observer for a servo control system of an industrial robot. Background Technology

[0002] With the development of intelligent and industrial robot technologies, industrial robots have been widely used in industrial production, assembly, and other fields. High speed and high precision have always been crucial technical indicators for industrial robots. However, because industrial robot servo systems use harmonic reducers (such as harmonic drive reducers) to connect the motor and the articulated arm to achieve speed reduction and torque amplification, these harmonic reducers can cause end-effector vibrations, affecting the robot's dynamic performance.

[0003] The above content is only used to help understand the technical solution of the present invention and does not represent an admission that the above content is prior art. Summary of the Invention

[0004] The purpose of this invention is to provide a control method, device, industrial robot, storage medium, and computer program product for an industrial robot, in order to solve the problem that harmonic reducers in industrial robots cause end-effector vibrations, affecting the dynamic performance of the industrial robot. This is achieved by setting an adaptive dimensionality-reduced state observer in the feedback section of the industrial robot's servo control system, compensating the fluctuating torque signal observed by the adaptive dimensionality-reduced state observer into the control of the servo motor on the harmonic reducer, thereby suppressing the vibrations caused by the harmonic reducer and improving the dynamic performance of the industrial robot.

[0005] This invention provides a control method for an industrial robot, the industrial robot having a servo motor, a harmonic reducer, and a servo control system; the control method for the industrial robot includes: designing an observer module; and setting the observer module in the feedback part of the servo control system; acquiring the speed feedback signal of the servo motor and the torque feedback signal of the harmonic reducer while the industrial robot is running; obtaining the torque disturbance signal of the industrial robot by observing the speed feedback signal of the servo motor and the torque feedback signal of the harmonic reducer through the observer module; and compensating the torque disturbance signal of the industrial robot into the torque control of the servo motor on the harmonic reducer to suppress the end effector vibration of the industrial robot.

[0006] In some embodiments, the observer module includes a PI module and an integrator, the PI module and the integrator being connected in series. The observer module observes the speed feedback signal of the servo motor and the torque feedback signal of the harmonic reducer to obtain the torque disturbance signal of the industrial robot. This includes: using the PI module to perform PI processing on the difference between the speed feedback signal of the servo motor and the estimated rotor speed of the servo motor to obtain the estimated load torque of the industrial robot, which serves as the torque disturbance signal; using the integrator to perform integrator processing on the estimated load torque of the industrial robot and the torque feedback signal of the harmonic reducer to obtain a new estimated rotor speed of the servo motor; again using the PI module to perform PI processing on the difference between the speed feedback signal of the servo motor and the new estimated rotor speed of the servo motor to obtain a new estimated load torque of the industrial robot, which serves as a new torque disturbance signal; and repeating this process to obtain a dynamic estimated load torque of the industrial robot, which serves as the dynamic torque disturbance signal of the industrial robot.

[0007] In some implementations, the observer module is expressed as follows:

[0008]

[0009] Where d is the differential operator and t is time. Estimate the speed of the rotor of the servo motor. The estimated load torque of the harmonic reducer is given by T, where B is the preset friction coefficient, J is the system moment of inertia of the servo control system, and T is the torque of the harmonic reducer. e ω represents the electromagnetic torque of the harmonic reducer, k1 is the proportional gain parameter of the observer module, k2 is the integral gain parameter of the observer module, and ω m The rotor mechanical angular velocity of the servo motor is... Estimate the speed of the rotor of the servo motor.

[0010] In some implementations, k1 and k2 are calculated using gradient descent in the expression of the observer module, and the expressions for k1 and k2 are as follows:

[0011]

[0012] Where, ω c δ is the cutoff frequency of the observer module, and δ is the stability margin of the observer module.

[0013] In some embodiments, the servo control system includes: a position regulator, a speed regulator, and a current regulator; wherein the position regulator is located in the speed loop of the servo control system; the speed regulator and the current regulator are located in the current loop of the servo control system; the output signal of the current regulator is capable of controlling the action of the servo motor; and the observer module is set in the feedback part of the servo control system, including: in the feedback part of the servo control system, controlling the harmonic reducer to operate based on the torque disturbance signal of the industrial robot output by the observer module and the output signal of the servo motor.

[0014] In some embodiments, the torque disturbance signal of the industrial robot is compensated into the torque control of the harmonic reducer by the servo motor to suppress the end effector vibration of the industrial robot. This includes: adjusting the position of the servo motor based on the difference between the position command signal and the position feedback signal of the servo motor sent by the host computer, using the position regulator, to obtain a speed command signal of the servo motor; adjusting the speed of the servo motor based on the speed command signal and the speed feedback signal of the servo motor, using the speed regulator, to obtain a current command signal of the servo motor; controlling the operation of the servo motor based on the current command signal of the servo motor using the current regulator; obtaining the torque command signal of the servo motor while it is running; obtaining the torque disturbance signal of the industrial robot through the observer module in the feedback part of the servo control system; and performing torque control on the harmonic reducer based on the difference between the torque command signal and the torque disturbance signal of the industrial robot to suppress the end effector vibration of the industrial robot.

[0015] In conjunction with the above method, another aspect of the present invention provides a control device for an industrial robot, the industrial robot having a servo motor, a harmonic reducer, and a servo control system; the control device for the industrial robot includes: a control unit configured to design an observer module; and the observer module being disposed in the feedback section of the servo control system; an acquisition unit configured to acquire the speed feedback signal of the servo motor and the torque feedback signal of the harmonic reducer when the industrial robot is running; the control unit is further configured to observe, through the observer module, based on the speed feedback signal of the servo motor and the torque feedback signal of the harmonic reducer to obtain a torque disturbance signal of the industrial robot; the control unit is further configured to compensate the torque disturbance signal of the industrial robot into the torque control of the servo motor on the harmonic reducer to suppress the end effector vibration of the industrial robot.

[0016] In some embodiments, the observer module includes a PI module and an integrator, the PI module and the integrator being connected in series; wherein, the control unit, through the observer module, observes the speed feedback signal of the servo motor and the torque feedback signal of the harmonic reducer to obtain the torque disturbance signal of the industrial robot, including: using the PI module to perform PI processing on the difference between the speed feedback signal of the servo motor and the estimated rotor speed of the servo motor to obtain the estimated load torque of the industrial robot, which serves as the torque disturbance signal of the industrial robot; using the integrator to perform integrator processing on the estimated load torque of the industrial robot and the torque feedback signal of the harmonic reducer to obtain a new estimated rotor speed of the servo motor; again using the PI module to perform PI processing on the difference between the speed feedback signal of the servo motor and the new estimated rotor speed of the servo motor to obtain a new estimated load torque of the industrial robot, which serves as a new torque disturbance signal of the industrial robot; repeating this process to obtain a dynamic estimated load torque of the industrial robot, which serves as the dynamic torque disturbance signal of the industrial robot.

[0017] In some implementations, the observer module is expressed as follows:

[0018]

[0019] Where d is the differential operator and t is time. Estimate the speed of the rotor of the servo motor. The estimated load torque of the harmonic reducer is given by T, where B is the preset friction coefficient, J is the system moment of inertia of the servo control system, and T is the torque of the harmonic reducer. e ω represents the electromagnetic torque of the harmonic reducer, k1 is the proportional gain parameter of the observer module, k2 is the integral gain parameter of the observer module, and ω m The rotor mechanical angular velocity of the servo motor is... Estimate the speed of the rotor of the servo motor.

[0020] In some implementations, k1 and k2 are calculated using gradient descent in the expression of the observer module, and the expressions for k1 and k2 are as follows:

[0021]

[0022] Where, ω c δ is the cutoff frequency of the observer module, and δ is the stability margin of the observer module.

[0023] In some embodiments, the servo control system includes: a position regulator, a speed regulator, and a current regulator; wherein the position regulator is located in the speed loop of the servo control system; the speed regulator and the current regulator are located in the current loop of the servo control system; the output signal of the current regulator is capable of controlling the action of the servo motor; the control unit sets the observer module in the feedback part of the servo control system, including: in the feedback part of the servo control system, controlling the harmonic reducer to operate based on the torque disturbance signal of the industrial robot output by the observer module and the output signal of the servo motor.

[0024] In some embodiments, the control unit compensates for the torque disturbance signal of the industrial robot in the torque control of the servo motor on the harmonic reducer to suppress the end effector vibration of the industrial robot. This includes: adjusting the position of the servo motor based on the difference between the position command signal and the position feedback signal sent by the host computer, obtained by the position regulator, to obtain a speed command signal for the servo motor; adjusting the speed based on the speed command signal and the speed feedback signal of the servo motor, obtained by the speed regulator, to obtain a current command signal for the servo motor; controlling the operation of the servo motor based on the current command signal of the servo motor, by the current regulator; acquiring the torque command signal of the servo motor while it is running; obtaining the torque disturbance signal of the industrial robot through the observer module in the feedback part of the servo control system; and performing torque control on the harmonic reducer based on the difference between the torque command signal and the torque disturbance signal of the industrial robot to suppress the end effector vibration of the industrial robot.

[0025] In conjunction with the above-described device, the present invention further provides an industrial robot, comprising: the control device for the industrial robot described above.

[0026] In conjunction with the above method, the present invention further provides a storage medium comprising a stored program, wherein, when the program is executed, the device containing the storage medium is controlled to perform the steps of the control method for the industrial robot described above.

[0027] In conjunction with the above method, the present invention further provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described industrial robot control method.

[0028] Therefore, the present invention addresses the two types of vibrations caused by harmonic reducers in industrial robots (i.e., end-effector vibration and resonance occurring when the natural frequency of the harmonic reducer is close to the natural frequency of the industrial robot). An adaptive dimension-reduced state observer is designed to address these vibrations. This adaptive dimension-reduced state observer is added to the feedback section of the industrial robot's servo control system. The fluctuating torque signal output by the adaptive dimension-reduced state observer is used to compensate the torque command of the servo motor in the servo control system to control the harmonic reducer, thereby suppressing the two types of vibrations caused by the harmonic reducer. Thus, by setting an adaptive dimension-reduced state observer in the feedback section of the industrial robot's servo control system and compensating the fluctuating torque signal observed by the adaptive dimension-reduced state observer into the servo motor's control of the harmonic reducer, the vibration caused by the harmonic reducer is suppressed, improving the dynamic performance of the industrial robot.

[0029] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention.

[0030] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0031] Figure 1 This is a flowchart illustrating an embodiment of the industrial robot control method of the present invention;

[0032] Figure 2 This is a schematic flowchart of an embodiment of the method of the present invention, in which the torque disturbance signal of the industrial robot is obtained through the observer module.

[0033] Figure 3 This is a schematic flowchart of an embodiment of the method of the present invention, which compensates the torque disturbance signal of the industrial robot for the control of the harmonic reducer by the servo motor.

[0034] Figure 4 This is a schematic diagram of a structure of an embodiment of the control device for an industrial robot of the present invention;

[0035] Figure 5 This is a schematic flowchart of a vibration suppression method based on an adaptive dimensionality reduction state observer according to the present invention;

[0036] Figure 6 This is a schematic diagram of the vibration suppression structure of the servo control system of the industrial robot of the present invention;

[0037] Figure 7 A simulation diagram of torque fluctuations observed by a torque observer;

[0038] Figure 8This is a schematic diagram of the torque observer.

[0039] Referring to the accompanying drawings, the reference numerals in the embodiments of the present invention are as follows:

[0040] 102 - Acquisition unit; 104 - Control unit. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0042] Considering that harmonic reducers in industrial robots can cause end-effector vibrations, affecting their dynamic performance, and given the unique dual-wave transmission characteristics of harmonic reducers—where the flexible gears undergo two meshing processes within a complete rotational cycle—resulting in even-frequency components in the output speed. This generates even-frequency fluctuations in the robot's rotational speed, causing end-effector vibrations and impacting the accuracy of the robot's trajectory. Furthermore, the assembly process of harmonic reducers for industrial robots requires high precision. During operation, the periodic meshing of the flexible and rigid gears inevitably generates vibrations. When the reducer's speed reaches a certain range, and its natural vibration frequency approaches that of the industrial robot, resonance occurs, exacerbating the vibration amplitude and causing end-effector jitter, severely affecting the robot's motion accuracy.

[0043] In related solutions, since the end-effector vibration caused by the harmonic reducer of the industrial robot is unavoidable, the design of the flexible gear is often improved to reduce the nonlinear effect of its deformation, or the servo compensation circuit is added to suppress the vibration.

[0044] Some solutions disclose an optimized joint drive structure for industrial robots that reduces the weight at the joint end and mitigates speed fluctuations caused by harmonic reducers. However, this approach makes the transmission structure of the industrial robot more complex, reduces the control efficiency of the drive on the joint arm, and does not completely eliminate the vibration caused by the harmonic reducer.

[0045] Other solutions disclose a vibration suppression method and device, a servo driver, and a servo drive system. One approach determines the current compensation value based on the position difference between the rotor position feedback value and the position command of the servo motor. A current regulator then controls the servo motor based on the current compensation value, the current feedback value, and the current command to suppress resonance caused by the harmonic reducer. However, this method uses differential calculation of the current compensation value, resulting in a certain lag, and only suppresses resonance caused by the harmonic reducer, failing to suppress even-frequency vibrations caused by the dual-wave transmission characteristics of the harmonic reducer.

[0046] Other solutions disclose an anti-interference control method for a frame system driven by a harmonic reducer. First, high-frequency vibrations caused by rotor dynamic imbalance are estimated in real time based on a mathematical model. Then, an extended state observer is designed to estimate the total interference of the harmonic reducer's transmission torque at the drive motor end in real time. Finally, a backstepping controller is designed to compensate for and suppress multi-source interference. This method uses an extended observer to observe various types of interference and suppress vibrations across multiple frequency bands. However, it does not consider that the even-numbered vibration frequencies generated by the harmonic reducer due to speed changes also change in real time, making effective suppression difficult.

[0047] In relevant solutions, industrial robots employ three main methods to suppress vibrations in their transmission structures, particularly those caused by harmonic reducers: The first method optimizes the joint connection structure to avoid or reduce structural vibrations; the second compensates for vibrations based on sensor feedback current or speed deviations; and the third involves designing observers to monitor various vibration signals and compensate for or even eliminate vibrations caused by interference signals. The first method requires modifying the industrial robot structure, is complex in design, has high assembly requirements, and can only suppress or eliminate single vibrations, not completely eliminating interference vibrations from harmonic reducers. The second method is simple to design but suffers from significant hysteresis, failing to effectively compensate for interference fluctuations in torque or current signals. The third method uses observers to monitor vibration signals, but often involves setting up multiple observers for overall interference signal monitoring, failing to consider how changes in the interference source can increase the observation error of existing observers, thus affecting the vibration suppression effect of the servo controller.

[0048] Therefore, the present invention proposes a control method for industrial robots, specifically a vibration suppression method based on an adaptive dimensionality reduction state observer. This method analyzes two types of vibration caused by the harmonic reducer in the industrial robot and designs an adaptive dimensionality reduction state observer as a torque observer. After observing the two types of vibration caused by the harmonic reducer in the industrial robot through the torque observer, the vibration signals are inversely compensated to the servo control system, which can effectively suppress the vibration caused by the harmonic reducer and improve the dynamic performance of the industrial robot.

[0049] In servo systems, the output of the servo motor is connected to the load and driven by a harmonic reducer. The industrial robot servo system receives position commands from the host computer and outputs control signals to drive the load. Because the harmonic reducer, acting as a transmission device in the industrial robot, has a natural low-frequency vibration due to assembly reasons, when the operating frequency of the motor matches the natural frequency of the harmonic reducer, the servo motor and the harmonic reducer will resonate. Simultaneously, this vibration is transmitted to the connected end effector of the industrial robot load, causing vibration and a decrease in the robot's dynamic performance. Furthermore, due to the unique dual-wave transmission characteristics of the harmonic reducer, the flexible gear undergoes two meshing processes within a complete rotation cycle. The harmonic reducer generates even-harmonic frequency fluctuations in the industrial robot's rotational speed, causing vibration at the robot's end effector and affecting its dynamic performance. The present invention proposes a vibration suppression scheme for an industrial robot servo control system based on an adaptive dimensionality-reduced state observer. The main effect is to solve the end effector vibration caused by the structure of the harmonic reducer in related solutions, thereby improving the dynamic performance of the industrial robot.

[0050] According to embodiments of the present invention, a control method for an industrial robot is provided, such as... Figure 1 The diagram shows a flowchart of an embodiment of the method of the present invention. The industrial robot includes a servo motor, a harmonic reducer, and a servo control system; in the solution of the present invention, as... Figure 1 As shown, the control method for the industrial robot includes steps S110 to S140.

[0051] In step S110, an observer module, such as an adaptive dimensionality reduction state observer, is designed in advance; and the observer module is set in the feedback part of the servo control system.

[0052] In step S120, while the industrial robot is running, the speed feedback signal of the servo motor and the torque feedback signal of the harmonic reducer are acquired.

[0053] In step S130, the torque disturbance signal of the industrial robot is obtained by observing the speed feedback signal of the servo motor and the torque feedback signal of the harmonic reducer through the observer module.

[0054] In step S140, the torque disturbance signal of the industrial robot is compensated into the torque control of the servo motor on the harmonic reducer to suppress the end vibration of the industrial robot, that is, to suppress the vibration caused by the harmonic reducer and improve the dynamic performance of the industrial robot.

[0055] This invention addresses the vibration of industrial robots caused by disturbances in their transmission structures. It proposes a vibration suppression scheme for the servo control system of an industrial robot based on an adaptive dimensionality-reduced state observer. The scheme analyzes two types of vibration caused by the harmonic reducer in the industrial robot and designs an adaptive dimensionality-reduced state observer as a torque observer. By adding a torque observer to the servo control system of the industrial robot, the observed values ​​of this adaptive dimensionality-reduced state observer can better match the actual disturbance waveform, significantly improving the dynamic performance of the industrial robot. It can specifically suppress even-frequency vibrations and resonances caused by the harmonic reducer in the transmission structure of the industrial robot, improving the control accuracy of the servo controller. This scheme is highly practical in engineering, flexible in structure, and has strong anti-interference capabilities.

[0056] Figure 5 This is a flowchart illustrating a vibration suppression method based on an adaptive dimensionality-reduced state observer according to the present invention. The solution of this invention, by incorporating an adaptive dimensionality-reduced state observer into the servo control system, suppresses the end effector vibration of the industrial robot from the harmonic reducer, effectively improving the dynamic performance of the industrial robot. The adaptively adjusting observer can adapt to various operating environments and exhibits better stability. The vibration suppression flowchart is shown below. Figure 5 As shown. It should be noted that the solution of this invention is not only applicable to vibrations caused by harmonic reducers in industrial robots, but also to vibrations caused by other interference sources, which can be suppressed using the solution of this invention and the servo system. In principle, the adaptive dimensionality reduction state observer designed in this invention observes not only the vibrations caused by the harmonic reducer, but also the vibrations caused by various possible interference sources affecting the industrial robot. It simply performs special processing on the vibrations caused by the harmonic reducer to adapt and adjust accordingly. Therefore, for other vibrations, the adaptive dimensionality reduction state observer designed in this invention can also observe them and feed them back into the servo control system for suppression.

[0057] In this invention, the adaptive aspect is manifested in the use of gradient descent to update the system's state observer parameters based on speed, thereby suppressing even-harmonic frequency fluctuations in the harmonic reducer and improving the dynamic performance and stability of the industrial robot. Other vibrations include mechanical resonances generated when the robot system's operating frequency approaches its natural frequency; robot vibrations caused by external disturbances such as load changes; the adaptive dimensionality-reduced state observer in this invention estimates the actual torque of the motor in real time; feeding back the estimated actual torque of the motor to the control system helps adjust the motor's driving torque to avoid entering the resonant frequency range, thus reducing the resulting vibrations. The torque observer can estimate the impact of external disturbances on the motor torque; by monitoring torque changes in real time, it can detect changes in external disturbances and feed them back to the torque control in real time to offset the effects of external disturbances and reduce vibrations.

[0058] In some implementations, the observer module includes a PI module and an integrator, wherein the PI module and the integrator are connected in series.

[0059] In step S130, the specific process of obtaining the torque disturbance signal of the industrial robot by observing the speed feedback signal of the servo motor and the torque feedback signal of the harmonic reducer through the observer module is illustrated in the following exemplary description.

[0060] The following is combined Figure 2 The schematic diagram shown is a flowchart of an embodiment of the method of the present invention in which the torque disturbance signal of the industrial robot is obtained through the observer module. The specific process of obtaining the torque disturbance signal of the industrial robot through the observer module in step S130 is further explained, including steps S210 to S240.

[0061] Step S210: Using the PI module, the difference between the speed feedback signal of the servo motor and the estimated rotor speed of the servo motor is processed to obtain the estimated load torque of the industrial robot, which is used as the torque disturbance signal of the industrial robot.

[0062] Step S220: Based on the estimated load torque of the industrial robot and the torque feedback signal of the harmonic reducer, the integrator processes the data to obtain a new estimated rotor speed of the servo motor.

[0063] Step S230: Again, the PI module performs PI processing based on the speed feedback signal of the servo motor and the difference between the new estimated rotor speed of the servo motor to obtain the new estimated load torque of the industrial robot, which serves as the new torque disturbance signal of the industrial robot.

[0064] Step S240, repeating this process, obtains the estimated load torque of the dynamic industrial robot, which serves as the torque disturbance signal of the dynamic industrial robot.

[0065] An adaptive state observer is an observer that can estimate the state online and automatically adjust its parameters. The observer designed in this invention consists of a PI circuit and an integrator connected in series. Figure 8 This is a schematic diagram of the torque observer.

[0066] like Figure 8 As shown, the rotor mechanical angular velocity ω m and the estimated rotor speed at the previous moment. The speed difference is obtained after passing through a comparator; this speed difference is then processed by the k1J module to obtain a first processed value, and by the k1 / s module to obtain a second processed value; the first and second processed values ​​are then compared with a comparator to obtain the estimated load torque. The estimated load torque Electromagnetic torque T e The torque difference is obtained after passing through the comparator; this torque difference is then processed by the 1 / (sJ+B) module to obtain the current estimated rotor speed.

[0067] The present invention proposes a vibration suppression scheme for a servo control system of an industrial robot based on an adaptive dimensionality reduction state observer. It analyzes two types of vibration caused by the harmonic reducer in the industrial robot and designs a dimensionality reduction state observer to suppress these two types of vibration, thereby reducing computational complexity. Furthermore, the present invention does not require external sensors, reducing costs and simplifying the design.

[0068] Considering that the even-frequency vibrations of harmonic reducers are difficult to observe effectively with a single observer, the present invention incorporates a nonlinear optimization-gradient descent method to tune the observer. This allows the adaptive observer to adjust the optimal state observer parameters based on dynamic parameters such as speed, obtaining a more realistic disturbance signal. As a result, the servo control system can effectively suppress the vibrations of industrial robots caused by harmonic reducers, improve the accuracy and reliability of industrial robot operation, and extend the lifespan of industrial robots.

[0069] The shaper method used in related solutions can only suppress vibrations with known amplitude and period, and the shaper needs to be readjusted for different trajectory protocols. The solution of this invention uses an observer to observe the vibration signal in real time, thereby compensating the servo control system. It is highly adaptable and has a good vibration suppression effect for different trajectories and even working conditions, thus enhancing the robustness and efficiency of industrial robots.

[0070] In some implementations, the observer module is expressed as follows:

[0071]

[0072] Where d is the differential operator and t is time. Estimate the speed of the rotor of the servo motor. The estimated load torque of the harmonic reducer is given by T, where B is the preset friction coefficient, J is the system moment of inertia of the servo control system, and T is the torque of the harmonic reducer. e ω represents the electromagnetic torque of the harmonic reducer, k1 is the proportional gain parameter of the observer module, k2 is the integral gain parameter of the observer module, and ω m The rotor mechanical angular velocity of the servo motor is... Estimate the speed of the rotor of the servo motor.

[0073] In the scheme of this invention, the adaptive observer used in the servo system is a dimension-reduced torque state observer based on nonlinear optimization. First, the dimension-reduced torque observer is designed. According to the principles of dynamics, the mechanical motion equation of the motor can be obtained as follows:

[0074]

[0075] In the formula, J is the system's moment of inertia; T e For electromagnetic torque; ω m θ is the rotor mechanical angular velocity; B is the coefficient of friction; m For mechanical angle; T L d represents the load torque; d represents the differential operator; and t represents time.

[0076] The controller sampling period is on the order of microseconds, which can be approximated as the load torque T within one sampling period. L Since it is a constant value, the torque observer can be constructed as follows:

[0077]

[0078] In the formula, k1 and k2 are the proportional and integral gain parameters of the designed observer. Estimate the speed of the rotor of the servo motor. To estimate the load torque, d is the differential operator, and the load torque can be obtained by simplification. The estimated equation is:

[0079]

[0080] In the formula, s is a complex frequency variable.

[0081] In this invention, a torque observer is added to the servo control system of the industrial robot. After observing the resonance generated by the harmonic reducer and the natural frequency of the industrial robot, the vibration signal is used to compensate the servo control system in reverse. Furthermore, because the harmonic reducer generates vibration signals with even-frequency harmonics due to its dual-wave transmission characteristics, the torque observer in the servo control system obtains its optimal parameters using a nonlinear optimization-gradient descent method. This approach provides better observation of varying harmonic reducer disturbances, effectively suppressing vibrations caused by the harmonic reducer and improving the dynamic performance of the industrial robot.

[0082] In some implementations, k1 and k2 are calculated using gradient descent in the expression of the observer module, and the expressions for k1 and k2 are as follows:

[0083]

[0084] Where, ω c δ is the cutoff frequency of the observer module, k1 is the proportional gain parameter of the observer module, k2 is the integral gain parameter of the observer module, and J is the system rotational inertia of the servo control system.

[0085] In the solution of this invention, the parameters k1 and k2 of the observer shown in formula (3) or formula (4) are calculated as follows:

[0086]

[0087] In the formula, ω c δ is the cutoff frequency of the load torque observer; δ is the stability margin of the load torque observer; J is the system moment of inertia.

[0088] Figure 7 This is a simulation diagram of torque fluctuations observed by the torque observer. The load torque observer design is complete, and simulations verify the accuracy of the observer in observing torque fluctuations caused by the harmonic reducer. Figure 7 As shown, Figure 7 The curves in the upper right corner, from top to bottom, represent the simulation diagrams of the actual torque, the fluctuations of the actual torque observed when the load torque observer gain k1 = 1500, the fluctuations of the actual torque observed when the load torque observer gain k1 = 3000, the fluctuations of the actual torque observed when the load torque observer gain k1 = 5000, and the fluctuations of the actual torque observed when the load torque observer gain k1 = 10000. Figure 7 The simulation diagram showing the fluctuation of the observed real torque at a load torque observer gain of k1 illustrates that the larger the load torque observer gain k1, the better the torque fluctuation is observed. However, excessively high gain can lead to high-frequency interference in the observed signal. Compensating this interference into the servo motor control torque introduces new vibrations. Therefore, a suitable observer parameter needs to be set to suppress vibrations. Furthermore, because the harmonic reducer, in addition to its natural frequency resonance, also exhibits even-harmonic frequency fluctuations that vary with motor speed, different frequency speed harmonic signals require different observer gain parameters k1. Therefore, the gradient descent method is used to obtain the optimal observer gain for the corresponding motor speed, thereby suppressing even-harmonic frequency fluctuations in the harmonic reducer and improving the dynamic performance of the industrial robot. The core parameter update formula for the gradient descent method is as follows:

[0089]

[0090] In the formula, β t With β t-1These represent the parameters at the current time step and the parameters at the previous time step, respectively, i.e., the observer gain k1 and the velocity; η is the learning rate, i.e., the step size controlling the update, ▽f(β) t-1 ) represents the gradient (i.e., the partial derivative vector) of the observer torque error at the previous moment.

[0091] The steps for designing an adaptive dimensionality reduction state observer with adaptive parameters are as follows:

[0092] Step 1: Initialization. Input motor parameters and initial speed to obtain the initial parameter β. t-1 And connect the observer to the servo control.

[0093] Step 2: Calculate the gradient. After the industrial robot starts running, calculate the gradient ▽f(β) of the vibration error at the end of the industrial robot at the previous parameter. t-1 ).

[0094] Step 3: Update the parameters and use formula (6) to obtain the new parameter β. t Update the observer parameters again.

[0095] Step 4: Repeat steps 2 and 3 until the torque fluctuation observed by the observer approximates the end-effector vibration of the industrial robot caused by the harmonic reducer. The resulting observer is as follows: Figure 8 As shown.

[0096] In the solution of this invention, the above-mentioned nonlinearly optimized adaptive dimension reduction state observer is connected to the servo control system to compensate for the torque fluctuation error observed by the observer, thereby eliminating the two types of vibration caused by the harmonic reducer. The compensation effect is independent and controllable, not affected by other regulator parameters, easy to implement, and greatly improves the dynamic performance of the industrial robot servo control system.

[0097] In some embodiments, the servo control system includes: a position regulator, a speed regulator, and a current regulator.

[0098] The position regulator is located in the speed loop of the servo control system; the speed regulator and the current regulator are located in the current loop of the servo control system; the output signal of the current regulator can control the action of the servo motor.

[0099] In step S110, setting the observer module in the feedback section of the servo control system includes: in the feedback section of the servo control system, controlling the harmonic reducer to operate based on the torque disturbance signal of the industrial robot output by the observer module and the output signal of the servo motor.

[0100] Figure 6 This is a schematic diagram of the vibration suppression structure of the servo control system for the industrial robot of the present invention. Figure 6 As shown, the difference between the position command issued by the host computer and the position feedback from the servo motor, obtained through a comparator, is used to obtain the speed command after passing through the position regulator. The difference between the speed command and the speed feedback from the servo motor, obtained through a comparator, is used to obtain the current command after passing through the speed regulator. The current command, after passing through a comparator, is output to the current regulator, which adjusts the current of the servo motor. The adaptive state observer outputs a torque disturbance signal based on the speed feedback from the servo motor and the torque fluctuation of the harmonic reducer. The difference between the torque command output by the servo motor and the torque disturbance signal output by the adaptive state observer, obtained through a comparator, is output to the harmonic reducer, which controls the end load. Currently, there is a direct relationship between current and torque; the torque fluctuation of the harmonic reducer is fed back into the current signal. By processing the feedback current fluctuation signal, the torque disturbance signal can be obtained.

[0101] In the above embodiments, an adaptive dimensionality-reduced state observer is incorporated into the industrial robot servo control system to suppress two types of vibrations caused by the harmonic reducer in the industrial robot servo motor drive. This embodiment addresses the vibration suppression caused by the harmonic reducer in the industrial robot; other vibrations in the industrial robot can also be suppressed using this design. Furthermore, this embodiment can observe system torque disturbances from other similar vibration sources in the industrial robot and feed their torque fluctuations back to the servo system control, thereby suppressing vibrations and achieving the effect of reducing end-effector vibration and improving the dynamic performance of the industrial robot.

[0102] In the above embodiments, the servo motor output commands, speed feedback, position feedback signals, and torque disturbance signals output by the observer may all contain high-frequency disturbance noise when connected to the servo system. These signals can all be filtered, including but not limited to low-pass or band-pass filters. Using a low-pass filter is simple to design and can effectively filter out high-frequency noise in the signal, resulting in a smoother system response and improved system reliability and dynamic performance. Using a band-pass filter can specifically remove high-frequency noise from the signal, reducing the impact on the useful signal compared to a low-pass filter, increasing the accuracy of the servo control signal, and enhancing system reliability.

[0103] In some embodiments, the specific process of compensating the torque disturbance signal of the industrial robot in step S140 into the torque control of the servo motor on the harmonic reducer to suppress the end effector vibration of the industrial robot is described in the following exemplary description.

[0104] The following is combined Figure 3The diagram shows a flowchart of an embodiment of the method of the present invention in which the torque disturbance signal of the industrial robot is compensated to the control of the harmonic reducer by the servo motor. The specific process of compensating the torque disturbance signal of the industrial robot to the control of the harmonic reducer by the servo motor in step S140 is further explained, including steps S310 to S340.

[0105] Step S310: The position is adjusted by the position adjuster based on the difference between the position command signal of the servo motor sent by the host computer and the position feedback signal of the servo motor, so as to obtain the speed command signal of the servo motor.

[0106] Step S320: The speed is adjusted by the speed regulator based on the speed command signal and the speed feedback signal of the servo motor to obtain the current command signal of the servo motor.

[0107] Step S330: The current regulator controls the operation of the servo motor based on the current command signal of the servo motor; while the servo motor is running, the torque command signal of the servo motor can be obtained.

[0108] In step S340, in the feedback section of the servo control system, the torque disturbance signal of the industrial robot is obtained through the observer module; based on the difference between the torque command signal of the servo motor and the torque disturbance signal of the industrial robot, torque control is performed on the harmonic reducer to suppress the end vibration of the industrial robot.

[0109] like Figure 5 As shown, the specific implementation process of a vibration suppression method based on an adaptive dimensionality reduction state observer is as follows:

[0110] Step S1: The industrial robot's servo control system receives the position command from the host computer. The servo control system then controls a motor with a harmonic reducer to move the robot's end effector. The rotor inside the servo motor is a permanent magnet. The three-phase electricity controlled by the servo driver in the servo control system forms an electromagnetic field. The rotor of the servo motor rotates under the influence of this magnetic field. Simultaneously, the encoder built into the servo motor collects the rotor's position and generates a feedback signal to the industrial robot's servo control system.

[0111] Step S2: During servo motor control, in addition to sending torque control signals to the transmission device—harmonic reducer—the servo motor also uses an encoder to provide feedback on the motor's position and speed for more precise control of the industrial robot's end effector's motion. The position feedback is subtracted from the position command from the host computer and sent to the position regulator in the servo system. The speed feedback is subtracted from the speed command output by the position regulator and then sent to the speed regulator. After receiving the current command from the speed regulator (resulting from the combined effect of the command and feedback), the current regulator outputs torque / current to the servo motor. The servo motor ultimately controls the movement of the industrial robot's end effector through the harmonic reducer, where the harmonic reducer amplifies torque and reduces speed. Figure 6 The command, feedback, and torque control sections of the servo control system.

[0112] Step S3: When the industrial robot end effector operates under the control of the servo control system described above, the output of the servo motor is connected to the end effector and driven through a transmission device—a harmonic reducer. Because the control system driven by the harmonic reducer generates two parts of industrial robot vibration: 1) One part of the vibration is due to a low-frequency natural frequency generated during the industrial robot assembly process. When the operating frequency of the motor is the same as the natural frequency of the harmonic reducer, the servo motor and the harmonic reducer will resonate. 2) In addition, due to the unique dual-wave transmission characteristics of the harmonic reducer, the flexible gear undergoes two meshing processes within a complete rotation cycle, resulting in even-harmonic frequency fluctuations in the output speed, causing industrial robot vibration. Both of these vibration signals are transmitted to the connected industrial robot end effector. If the industrial robot is controlled solely by the control unit described in Step 2 above, the vibrations caused by these two parts of the harmonic reducer cannot be effectively eliminated, resulting in vibration of the industrial robot end effector and affecting the dynamic performance of the industrial robot.

[0113] To address these two types of vibration, the present invention incorporates an adaptive dimensionality-reduced state observer module into the feedback section of the servo control system. Related solutions only observe the overall state of the interference experienced by the servo motor of the industrial robot, with few analyzing and designing state observers specifically for the vibration interference source. Some studies have analyzed the vibration in harmonic reducers, but these are often categorized as a single type of vibration, ignoring the even-harmonic frequency fluctuations caused by the dual-wave characteristics of the harmonic reducer. Therefore, most studies assume the vibration is a single-frequency fluctuation, using a single state observer to observe the error, failing to consider that the error vibration frequency changes with the motor speed. This approach cannot effectively suppress the vibration caused by the harmonic reducer. Therefore, an adaptive dimensionality-reduced state observer needs to be designed and added to the servo control system. Its output fluctuating torque signal compensates for the servo motor's control of the harmonic reducer to suppress these two types of vibration. Figure 6 As shown.

[0114] Step S4: The adaptive observer used in the servo system of the present invention is a dimension-reduced torque state observer based on nonlinear optimization.

[0115] In this invention, the designed adaptive dimensionality-reduced state observer can be applied to the servo control system described in this invention, and can also be used in other industrial robot servo motor control systems with harmonic reducers. The vibration sources are resonance and even-frequency fluctuations in velocity, and the vibration suppression method is the adaptive dimensionality-reduced state observer. This invention designs a servo control system for an industrial robot, and in the servo control system of the industrial robot, it designs an adaptive state observer with better performance for the even-frequency vibrations and resonances caused by the harmonic reducer of the industrial robot. This solution has strong engineering practicality, high reliability, and high stability. The designed dimensionality-reduced observer has a simple structure and strong adaptability, and can effectively observe torque fluctuations in the control system. Furthermore, the adaptive observer designed using the gradient descent method can effectively suppress the even-frequency fluctuations in the harmonic reducer that change with speed, greatly improving the dynamic performance of the servo controller for the industrial robot.

[0116] The technical solution of this embodiment addresses two types of vibrations caused by harmonic reducers in industrial robots (i.e., end-effector vibration and resonance occurring when the natural frequency of the harmonic reducer is close to the natural frequency of the industrial robot). An adaptive dimension-reduced state observer is designed to counter these vibrations. This observer is added to the feedback section of the industrial robot's servo control system. The fluctuating torque signal output by the observer is used to compensate the torque command of the servo motor in the servo control system to control the harmonic reducer, thereby suppressing the two types of vibrations caused by the harmonic reducer. Thus, by setting an adaptive dimension-reduced state observer in the feedback section of the industrial robot's servo control system and compensating the fluctuating torque signal observed by the observer into the servo motor's control of the harmonic reducer, the vibrations caused by the harmonic reducer are suppressed, improving the dynamic performance of the industrial robot.

[0117] According to an embodiment of the present invention, a control device for an industrial robot corresponding to a control method for an industrial robot is also provided. See also Figure 4 The diagram shows a structural schematic of an embodiment of the device of the present invention. The industrial robot includes a servo motor, a harmonic reducer, and a servo control system; in the solution of the present invention, as... Figure 4 As shown, the control device of the industrial robot includes: an acquisition unit 102 and a control unit 104.

[0118] The control unit 104 is configured to pre-design an observer module, such as an adaptive dimensionality reduction state observer, and to place the observer module in the feedback section of the servo control system. The specific functions and processing of the control unit 104 are described in step S110.

[0119] The acquisition unit 102 is configured to acquire the speed feedback signal of the servo motor and the torque feedback signal of the harmonic reducer when the industrial robot is running. The specific functions and processing of the acquisition unit 102 are described in step S120.

[0120] The control unit 104 is further configured to observe, through the observer module, the speed feedback signal of the servo motor and the torque feedback signal of the harmonic reducer to obtain the torque disturbance signal of the industrial robot. The specific functions and processing of the control unit 104 are further described in step S130.

[0121] The control unit 104 is further configured to compensate the torque disturbance signal of the industrial robot into the torque control of the servo motor on the harmonic reducer, so as to suppress the end effector vibration of the industrial robot, that is, to suppress the vibration caused by the harmonic reducer and improve the dynamic performance of the industrial robot. The specific functions and processing of the control unit 104 are further described in step S140.

[0122] This invention addresses the vibration caused by disturbances in the transmission structure of industrial robots. It proposes a vibration suppression scheme for the servo control system of an industrial robot based on an adaptive dimensionality-reduced state observer. The scheme analyzes two types of vibration caused by the harmonic reducer in the industrial robot and designs an adaptive dimensionality-reduced state observer as a torque observer. By adding a torque observer to the servo control system of the industrial robot, the observed values ​​of this adaptive dimensionality-reduced state observer can better match the actual disturbance waveform, significantly improving the dynamic performance of the industrial robot. It can specifically suppress even-frequency vibrations and resonances caused by the harmonic reducer in the transmission structure of the industrial robot, improving the control accuracy of the servo controller. This scheme is highly practical in engineering, flexible in structure, and has strong anti-interference capabilities.

[0123] The solution of this invention, by incorporating an adaptive dimensionality-reduced state observer into the servo control system, suppresses the end effector vibration of the industrial robot from the harmonic reducer, effectively improving the dynamic performance of the industrial robot. The adaptive adjustment observer can adapt to various operating environments and has better stability. The vibration suppression flowchart is shown below. Figure 5As shown. It should be noted that the solution of this invention is not only applicable to vibrations caused by harmonic reducers in industrial robots, but also to vibrations caused by other interference sources, which can be suppressed using the solution of this invention and the servo system. In principle, the adaptive dimensionality reduction state observer designed in this invention observes not only the vibrations caused by the harmonic reducer, but also the vibrations caused by various possible interference sources affecting the industrial robot. It simply performs special processing on the vibrations caused by the harmonic reducer to adapt and adjust accordingly. Therefore, for other vibrations, the adaptive dimensionality reduction state observer designed in this invention can also observe them and feed them back into the servo control system for suppression.

[0124] In some implementations, the observer module includes a PI module and an integrator, wherein the PI module and the integrator are connected in series.

[0125] The control unit 104, through the observer module, observes the speed feedback signal of the servo motor and the torque feedback signal of the harmonic reducer to obtain the torque disturbance signal of the industrial robot, including:

[0126] The control unit 104 is further configured to perform PI processing on the difference between the speed feedback signal of the servo motor and the estimated rotor speed of the servo motor through the PI module to obtain the estimated load torque of the industrial robot, which serves as the torque disturbance signal of the industrial robot. The specific functions and processing of this control unit 104 are further described in step S210.

[0127] The control unit 104 is further configured to perform integrator processing based on the estimated load torque of the industrial robot and the torque feedback signal of the harmonic reducer to obtain a new estimated rotor speed of the servo motor. The specific functions and processing of the control unit 104 are further described in step S220.

[0128] The control unit 104 is further configured to perform PI processing again through the PI module based on the difference between the speed feedback signal of the servo motor and the estimated rotor speed of the new servo motor, to obtain a new estimated load torque of the industrial robot, which serves as a new torque disturbance signal for the industrial robot. The specific functions and processing of this control unit 104 are further described in step S230.

[0129] The control unit 104 is further configured to cyclically obtain the estimated load torque of the industrial robot as a dynamic torque disturbance signal for the industrial robot. The specific functions and processing of the control unit 104 are further described in step S240.

[0130] An adaptive state observer is an observer that can estimate the state online and automatically adjust its parameters. The observer designed in this invention consists of a PI circuit and an integrator connected in series.

[0131] like Figure 8 As shown, the rotor mechanical angular velocity ω m and the estimated rotor speed at the previous moment. The speed difference is obtained after passing through a comparator; this speed difference is then processed by the k1J module to obtain a first processed value, and by the k1 / s module to obtain a second processed value; the first and second processed values ​​are then compared with a comparator to obtain the estimated load torque. The estimated load torque Electromagnetic torque T e The torque difference is obtained after passing through the comparator; this torque difference is then processed by the 1 / (sJ+B) module to obtain the current estimated rotor speed.

[0132] The present invention proposes a vibration suppression scheme for a servo control system of an industrial robot based on an adaptive dimensionality reduction state observer. It analyzes two types of vibration caused by the harmonic reducer in the industrial robot and designs a dimensionality reduction state observer to suppress these two types of vibration, thereby reducing computational complexity. Furthermore, the present invention does not require external sensors, reducing costs and simplifying the design.

[0133] Considering that the even-frequency vibrations of harmonic reducers are difficult to observe effectively with a single observer, the present invention incorporates a nonlinear optimization-gradient descent method to tune the observer. This allows the adaptive observer to adjust the optimal state observer parameters based on dynamic parameters such as speed, obtaining a more realistic disturbance signal. As a result, the servo control system can effectively suppress the vibrations of industrial robots caused by harmonic reducers, improve the accuracy and reliability of industrial robot operation, and extend the lifespan of industrial robots.

[0134] The shaper method used in related solutions can only suppress vibrations with known amplitude and period, and the shaper needs to be readjusted for different trajectory protocols. The solution of this invention uses an observer to observe the vibration signal in real time, thereby compensating the servo control system. It is highly adaptable and has a good vibration suppression effect for different trajectories and even working conditions, thus enhancing the robustness and efficiency of industrial robots.

[0135] In some implementations, the observer module is expressed as follows:

[0136]

[0137] Where d is the differential operator and t is time. Estimate the speed of the rotor of the servo motor. The estimated load torque of the harmonic reducer is given by T, where B is the preset friction coefficient, J is the system moment of inertia of the servo control system, and T is the torque of the harmonic reducer. e ω represents the electromagnetic torque of the harmonic reducer, k1 is the proportional gain parameter of the observer module, k2 is the integral gain parameter of the observer module, and ω m The rotor mechanical angular velocity of the servo motor is... Estimate the speed of the rotor of the servo motor.

[0138] In the scheme of this invention, the adaptive observer used in the servo system is a dimension-reduced torque state observer based on nonlinear optimization. First, the dimension-reduced torque observer is designed. According to the principles of dynamics, the mechanical motion equation of the motor can be obtained as follows:

[0139]

[0140] In the formula, J is the system's moment of inertia; T e For electromagnetic torque; ω m θ is the rotor mechanical angular velocity; B is the coefficient of friction; m For mechanical angle; T L d represents the load torque; d represents the differential operator; and t represents time.

[0141] The controller sampling period is on the order of microseconds, which can be approximated as the load torque T within one sampling period. L Since it is a constant value, the torque observer can be constructed as follows:

[0142]

[0143] In the formula, k1 and k2 are the proportional and integral gain parameters of the designed observer. Estimate the speed of the rotor of the servo motor. To estimate the load torque, d is the differential operator, and the load torque can be obtained by simplification. The estimated equation is:

[0144]

[0145] In the formula, s is a complex frequency variable.

[0146] In this invention, a torque observer is added to the servo control system of the industrial robot. After observing the resonance generated by the harmonic reducer and the natural frequency of the industrial robot, the vibration signal is used to compensate the servo control system in reverse. Furthermore, because the harmonic reducer generates vibration signals with even-frequency harmonics due to its dual-wave transmission characteristics, the torque observer in the servo control system obtains its optimal parameters using a nonlinear optimization-gradient descent method. This approach provides better observation of varying harmonic reducer disturbances, effectively suppressing vibrations caused by the harmonic reducer and improving the dynamic performance of the industrial robot.

[0147] In some implementations, k1 and k2 are calculated using gradient descent in the expression of the observer module, and the expressions for k1 and k2 are as follows:

[0148]

[0149] Where, ω c δ is the cutoff frequency of the observer module, k1 is the proportional gain parameter of the observer module, k2 is the integral gain parameter of the observer module, and J is the system rotational inertia of the servo control system.

[0150] In the solution of this invention, the parameters k1 and k2 of the observer shown in formula (3) or formula (4) are calculated as follows:

[0151]

[0152] In the formula, ω c δ is the cutoff frequency of the load torque observer; δ is the stability margin of the load torque observer; J is the system moment of inertia.

[0153] The load torque observer has been designed, and simulations have verified its accuracy in observing torque fluctuations caused by harmonic reducers. Figure 7 As shown, Figure 7 The curves in the upper right corner, from top to bottom, represent the simulation diagrams of the actual torque, the fluctuations of the actual torque observed when the load torque observer gain k1 = 1500, the fluctuations of the actual torque observed when the load torque observer gain k1 = 3000, the fluctuations of the actual torque observed when the load torque observer gain k1 = 5000, and the fluctuations of the actual torque observed when the load torque observer gain k1 = 10000. Figure 7The simulation diagram showing the fluctuation of the observed real torque at a load torque observer gain of k1 illustrates that the larger the load torque observer gain k1, the better the torque fluctuation is observed. However, excessively high gain can lead to high-frequency interference in the observed signal. Compensating this interference into the servo motor control torque introduces new vibrations. Therefore, a suitable observer parameter needs to be set to suppress vibrations. Furthermore, because the harmonic reducer, in addition to its natural frequency resonance, also exhibits even-harmonic frequency fluctuations that vary with motor speed, different frequency speed harmonic signals require different observer gain parameters k1. Therefore, the gradient descent method is used to obtain the optimal observer gain for the corresponding motor speed, thereby suppressing even-harmonic frequency fluctuations in the harmonic reducer and improving the dynamic performance of the industrial robot. The core parameter update formula for the gradient descent method is as follows:

[0154]

[0155] In the formula, β t With β t-1 These represent the parameters at the current time step and the parameters at the previous time step, respectively, i.e., the observer gain k1 and the velocity; η is the learning rate, i.e., the step size controlling the update, ▽f(β) t-1 ) represents the gradient (i.e., the partial derivative vector) of the observer torque error at the previous moment.

[0156] The steps for designing an adaptive dimensionality reduction state observer with adaptive parameters are as follows:

[0157] Step 1: Initialization. Input motor parameters and initial speed to obtain the initial parameter β. t-1 And connect the observer to the servo control.

[0158] Step 2: Calculate the gradient. After the industrial robot starts running, calculate the gradient ▽f(β) of the vibration error at the end of the industrial robot at the previous parameter. t-1 ).

[0159] Step 3: Update the parameters and use formula (6) to obtain the new parameter β. t Update the observer parameters again.

[0160] Step 4: Repeat steps 2 and 3 until the torque fluctuation observed by the observer approximates the end-effector vibration of the industrial robot caused by the harmonic reducer. The resulting observer is as follows: Figure 8 As shown.

[0161] In the solution of this invention, the above-mentioned nonlinearly optimized adaptive dimension reduction state observer is connected to the servo control system to compensate for the torque fluctuation error observed by the observer, thereby eliminating the two types of vibration caused by the harmonic reducer. The compensation effect is independent and controllable, not affected by other regulator parameters, easy to implement, and greatly improves the dynamic performance of the industrial robot servo control system.

[0162] In some embodiments, the servo control system includes: a position regulator, a speed regulator, and a current regulator.

[0163] The position regulator is located in the speed loop of the servo control system; the speed regulator and the current regulator are located in the current loop of the servo control system; the output signal of the current regulator can control the action of the servo motor.

[0164] The control unit 104, which sets the observer module in the feedback part of the servo control system, includes: the control unit 104 is further configured to control the harmonic reducer to operate in the feedback part of the servo control system based on the torque disturbance signal of the industrial robot output by the observer module and the output signal of the servo motor.

[0165] like Figure 6 As shown, the difference between the position command issued by the host computer and the position feedback of the servo motor, obtained after being compared by a comparator, is used to obtain the speed command after being processed by the position regulator; the difference between the speed command and the speed feedback of the servo motor, obtained after being compared by a comparator, is used to obtain the current command after being processed by the speed regulator; the current command is output to the current regulator after being compared by a comparator, and the current regulator adjusts the current of the servo motor; the adaptive state observer outputs a torque disturbance signal based on the speed feedback of the servo motor and the torque fluctuation of the harmonic reducer; the difference between the torque command output by the servo motor and the torque disturbance signal output by the adaptive state observer, obtained after being compared by a comparator, is output to the harmonic reducer, which controls the end load.

[0166] In the above embodiments, an adaptive dimensionality-reduced state observer is incorporated into the industrial robot servo control system to suppress two types of vibrations caused by the harmonic reducer in the industrial robot servo motor drive. This embodiment addresses the vibration suppression caused by the harmonic reducer in the industrial robot; other vibrations in the industrial robot can also be suppressed using this design. Furthermore, this embodiment can observe system torque disturbances from other similar vibration sources in the industrial robot and feed their torque fluctuations back to the servo system control, thereby suppressing vibrations and achieving the effect of reducing end-effector vibration and improving the dynamic performance of the industrial robot.

[0167] In the above embodiments, the servo motor output commands, speed feedback, position feedback signals, and torque disturbance signals output by the observer may all contain high-frequency disturbance noise when connected to the servo system. These signals can all be filtered, including but not limited to low-pass or band-pass filters. Using a low-pass filter is simple to design and can effectively filter out high-frequency noise in the signal, resulting in a smoother system response and improved system reliability and dynamic performance. Using a band-pass filter can specifically remove high-frequency noise from the signal, reducing the impact on the useful signal compared to a low-pass filter, increasing the accuracy of the servo control signal, and enhancing system reliability.

[0168] In some embodiments, the control unit 104 compensates for the torque disturbance signal of the industrial robot in the torque control of the servo motor on the harmonic reducer to suppress the end effector vibration of the industrial robot, including:

[0169] The control unit 104 is further configured to adjust the position of the servo motor based on the difference between the position command signal of the servo motor sent by the host computer and the position feedback signal of the servo motor, thereby obtaining the speed command signal of the servo motor. The specific functions and processing of this control unit 104 are further described in step S310.

[0170] The control unit 104 is further configured to adjust the speed based on the speed command signal and the speed feedback signal of the servo motor through the speed regulator, thereby obtaining the current command signal of the servo motor. The specific functions and processing of the control unit 104 are further described in step S320.

[0171] The control unit 104 is further configured to control the operation of the servo motor based on the current command signal of the servo motor through the current regulator; and to acquire the torque command signal of the servo motor when the servo motor is running. The specific functions and processing of the control unit 104 are further described in step S330.

[0172] Specifically, the control unit 104 is further configured to obtain the torque disturbance signal of the industrial robot through the observer module in the feedback section of the servo control system; and to perform torque control on the harmonic reducer based on the difference between the torque command signal of the servo motor and the torque disturbance signal of the industrial robot, so as to suppress the end effector vibration of the industrial robot. The specific functions and processing of the control unit 104 are further described in step S340.

[0173] like Figure 5 As shown, the specific implementation process of a vibration suppression method based on an adaptive dimensionality reduction state observer is as follows:

[0174] Step S1: The industrial robot's servo control system receives the industrial robot's position command from the host computer. The servo control system then controls a motor with a harmonic reducer to move the robot's end effector. The rotor inside the servo motor is a permanent magnet. The three-phase electricity controlled by the servo driver in the industrial robot's servo control system forms an electromagnetic field. The rotor of the servo motor rotates under the influence of this magnetic field. Simultaneously, the encoder built into the servo motor collects the rotor's position and generates a feedback signal to the industrial robot's servo control system.

[0175] Step S2: During servo motor control, in addition to sending torque control signals to the transmission device—harmonic reducer—the servo motor also uses an encoder to provide feedback on the motor's position and speed for more precise control of the industrial robot's end effector's motion. The position feedback is subtracted from the position command from the host computer and sent to the position regulator in the servo system. The speed feedback is subtracted from the speed command output by the position regulator and then sent to the speed regulator. After receiving the current command from the speed regulator (resulting from the combined effect of the command and feedback), the current regulator outputs torque / current to the servo motor. The servo motor ultimately controls the movement of the industrial robot's end effector through the harmonic reducer, where the harmonic reducer amplifies torque and reduces speed. Figure 6 The command, feedback, and torque control sections of the servo control system.

[0176] Step S3: When the industrial robot end effector operates under the control of the servo control system described above, the output of the servo motor is connected to the end effector and driven through a transmission device—a harmonic reducer. Because the control system driven by the harmonic reducer generates two parts of industrial robot vibration: 1) One part of the vibration is due to a low-frequency natural frequency generated during the industrial robot assembly process. When the operating frequency of the motor is the same as the natural frequency of the harmonic reducer, the servo motor and the harmonic reducer will resonate. 2) In addition, due to the unique dual-wave transmission characteristics of the harmonic reducer, the flexible gear undergoes two meshing processes within a complete rotation cycle, resulting in even-harmonic frequency fluctuations in the output speed, causing industrial robot vibration. Both of these vibration signals are transmitted to the connected industrial robot end effector. If the industrial robot is controlled solely by the control unit described in Step 2 above, the vibrations caused by these two parts of the harmonic reducer cannot be effectively eliminated, resulting in vibration of the industrial robot end effector and affecting the dynamic performance of the industrial robot.

[0177] To address these two types of vibration, the present invention incorporates an adaptive dimensionality-reduced state observer module into the feedback section of the servo control system. Related solutions only observe the overall state of the interference experienced by the servo motor of the industrial robot, with few analyzing and designing state observers specifically for the vibration interference source. Some studies have analyzed the vibration in harmonic reducers, but these are often categorized as a single type of vibration, ignoring the even-harmonic frequency fluctuations caused by the dual-wave characteristics of the harmonic reducer. Therefore, most studies assume the vibration is a single-frequency fluctuation, using a single state observer to observe the error, failing to consider that the error vibration frequency changes with the motor speed. This approach cannot effectively suppress the vibration caused by the harmonic reducer. Therefore, an adaptive dimensionality-reduced state observer needs to be designed and added to the servo control system. Its output fluctuating torque signal compensates for the servo motor's control of the harmonic reducer to suppress these two types of vibration. Figure 6 As shown.

[0178] Step S4: The adaptive observer used in the servo system of the present invention is a dimension-reduced torque state observer based on nonlinear optimization.

[0179] In this invention, the designed adaptive dimensionality-reduced state observer can be applied to the servo control system described in this invention, and can also be used in other industrial robot servo motor control systems with harmonic reducers. The vibration sources are resonance and even-frequency fluctuations in velocity, and the vibration suppression method is the adaptive dimensionality-reduced state observer. This invention designs a servo control system for an industrial robot, and in the servo control system of the industrial robot, it designs an adaptive state observer with better performance for the even-frequency vibrations and resonances caused by the harmonic reducer of the industrial robot. This solution has strong engineering practicality, high reliability, and high stability. The designed dimensionality-reduced observer has a simple structure and strong adaptability, and can effectively observe torque fluctuations in the control system. Furthermore, the adaptive observer designed using the gradient descent method can effectively suppress the even-frequency fluctuations in the harmonic reducer that change with speed, greatly improving the dynamic performance of the servo controller for the industrial robot.

[0180] Since the processing and functions implemented by the device in this embodiment are basically the same as the embodiments, principles and examples of the aforementioned methods, any details not covered in the description of this embodiment can be found in the relevant descriptions in the aforementioned embodiments, and will not be repeated here.

[0181] According to an embodiment of the present invention, an industrial robot corresponding to a control device for an industrial robot is also provided. This industrial robot may include the control device for the industrial robot described above.

[0182] Since the processing and functions implemented by the industrial robot in this embodiment are basically the same as the embodiments, principles and examples of the aforementioned device, any details not covered in this embodiment can be found in the relevant descriptions in the aforementioned embodiments, and will not be repeated here.

[0183] According to an embodiment of the present invention, a computer program product corresponding to the control method for an industrial robot is also provided, comprising a computer program that, when executed by a processor, implements the steps of the control method for the industrial robot described above.

[0184] Since the processing and functions implemented by the product in this embodiment are basically the same as the embodiments, principles and examples of the aforementioned methods, any details not covered in the description of this embodiment can be found in the relevant descriptions in the aforementioned embodiments, and will not be repeated here.

[0185] According to an embodiment of the present invention, a storage medium corresponding to a control method for an industrial robot is also provided. The storage medium includes a stored program, wherein, when the program is executed, the device where the storage medium is located executes the steps of the control method for the industrial robot described above.

[0186] Since the processing and functions implemented by the storage medium in this embodiment are basically the same as the embodiments, principles and examples of the aforementioned methods, any details not covered in the description of this embodiment can be found in the relevant descriptions in the aforementioned embodiments, and will not be repeated here.

[0187] In summary, it is readily understood by those skilled in the art that, without conflict, the aforementioned advantageous methods can be freely combined and superimposed.

[0188] The above description is merely an embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the invention by those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the invention should be included within the scope of the claims.

Claims

1. A control method for an industrial robot, characterized in that, The industrial robot has a servo motor, a harmonic reducer, and a servo control system; the control method of the industrial robot includes: Design an observer module; and set the observer module in the feedback part of the servo control system; the observer module includes: a PI module and an integrator, the PI module and the integrator being connected in series; While the industrial robot is in operation, the speed feedback signal of the servo motor and the torque feedback signal of the harmonic reducer are acquired. The observer module observes the speed feedback signal of the servo motor and the torque feedback signal of the harmonic reducer to obtain the torque disturbance signal of the industrial robot. This includes: using the PI module to perform PI processing on the difference between the speed feedback signal of the servo motor and the estimated rotor speed of the servo motor to obtain the estimated load torque of the industrial robot, which serves as the torque disturbance signal; using the integrator to perform integrator processing on the estimated load torque of the industrial robot and the torque feedback signal of the harmonic reducer to obtain a new estimated rotor speed of the servo motor; again using the PI module to perform PI processing on the difference between the speed feedback signal of the servo motor and the new estimated rotor speed of the servo motor to obtain a new estimated load torque of the industrial robot, which serves as a new torque disturbance signal; and repeating this process to obtain a dynamic estimated load torque of the industrial robot, which serves as the dynamic torque disturbance signal of the industrial robot. The torque disturbance signal of the industrial robot is compensated into the torque control of the servo motor on the harmonic reducer to suppress the end effector vibration of the industrial robot.

2. The control method for an industrial robot according to claim 1, characterized in that, The expression for the observer module is as follows: ； in, d For differential operators, t For time, Estimate the speed of the rotor of the servo motor. The estimated load torque of the harmonic reducer. B The preset coefficient of friction, J The system rotational inertia of the servo control system is given. T e The electromagnetic torque of the harmonic reducer is... k 1 The proportional gain parameter of the observer module. k 2 The integral gain parameter of the observer module. ω m The rotor mechanical angular velocity of the servo motor is... Estimate the speed of the rotor of the servo motor.

3. The control method for an industrial robot according to claim 2, characterized in that, The expression for the observer module is calculated using the gradient descent method. The expression is as follows: ； in, ω c The cutoff frequency of the observer module is... δ This represents the stability margin of the observer module.

4. The control method for an industrial robot according to any one of claims 1 to 3, characterized in that, The servo control system includes: a position regulator, a speed regulator, and a current regulator; wherein, The position regulator is located in the speed loop of the servo control system; the speed regulator and the current regulator are located in the current loop of the servo control system; the output signal of the current regulator can control the action of the servo motor. Setting the observer module in the feedback section of the servo control system includes: In the feedback section of the servo control system, the harmonic reducer is controlled to operate based on the torque disturbance signal of the industrial robot output by the observer module and the output signal of the servo motor.

5. The control method for an industrial robot according to claim 4, characterized in that, The torque disturbance signal of the industrial robot is compensated for in the torque control of the servo motor to the harmonic reducer to suppress the end effector vibration of the industrial robot, including: The position is adjusted based on the difference between the position command signal of the servo motor sent by the host computer and the position feedback signal of the servo motor, and the speed command signal of the servo motor is obtained by using the position adjuster. The speed is adjusted based on the speed command signal and the speed feedback signal of the servo motor by the speed regulator, and the current command signal of the servo motor is obtained. The current regulator controls the operation of the servo motor based on the current command signal of the servo motor; while the servo motor is running, the torque command signal of the servo motor can be obtained. In the feedback section of the servo control system, the torque disturbance signal of the industrial robot is obtained through the observer module; based on the difference between the torque command signal of the servo motor and the torque disturbance signal of the industrial robot, torque control is performed on the harmonic reducer to suppress the end effector vibration of the industrial robot.

6. A control device for an industrial robot, wherein the industrial robot is controlled using the control method for any one of claims 1 to 5, characterized in that, The industrial robot has a servo motor, a harmonic reducer, and a servo control system; the control device of the industrial robot includes: The control unit is configured to design an observer module; and the observer module is located in the feedback section of the servo control system; The acquisition unit is configured to acquire the speed feedback signal of the servo motor and the torque feedback signal of the harmonic reducer when the industrial robot is running. The control unit is further configured to observe, through the observer module, the speed feedback signal of the servo motor and the torque feedback signal of the harmonic reducer to obtain the torque disturbance signal of the industrial robot. The control unit is further configured to compensate the torque disturbance signal of the industrial robot into the torque control of the servo motor on the harmonic reducer, so as to suppress the end effector vibration of the industrial robot.

7. An industrial robot, characterized in that, include: The control device for an industrial robot as described in claim 6.

8. A storage medium, characterized in that, The storage medium includes a stored program, wherein, when the program is executed, the device containing the storage medium is controlled to perform the control method of the industrial robot as described in any one of claims 1 to 5.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the control method for the industrial robot according to any one of claims 1 to 5.

Images