A linear servo joint end vibration self-adaptive suppression method and system

Abstract

The application discloses a linear servo joint end vibration self-adaptive suppression method and system, and the method comprises the following steps: performing feature decoupling on a multidimensional signal to obtain a preliminarily decoupled dynamic force signal, and performing component extraction on the preliminarily decoupled dynamic force signal to obtain a high-frequency signal component; determining a cross-correlation coefficient of a linear servo joint theoretical linear acceleration and the high-frequency signal component; determining a vibration signal attenuation factor according to the high-frequency signal component of an adjacent moment; determining whether the current state is a mechanical residual vibration based on the cross-correlation coefficient and the attenuation factor; if yes, starting real-time tracking of a resonance frequency, determining a real-time resonance frequency of the linear servo joint under a current posture; updating a transfer function of a notch filter based on the real-time resonance frequency; inputting an original control instruction into the updated notch filter for feedforward shaping, and outputting a shaped compensation current instruction to a servo driver, so that the linear servo joint end vibration is suppressed.

Description

Technical Field

[0001] This invention relates to the field of robot control technology, and in particular to an adaptive method and system for suppressing vibration at the end of a linear servo joint. Background Technology

[0002] Linear servo joints, due to their high rigidity and high force-to-weight ratio, are now widely used in the hip, knee, and ankle joints of humanoid robots, improving their load-bearing and support capabilities. Linear servo joints typically consist of a frameless torque motor and a planetary roller screw. Their basic principle of motion is that the planetary roller screw converts the rotational motion generated by the frameless torque motor into a high-thrust linear reciprocating motion.

[0003] While planetary roller screws offer linear servo joints advantages such as high precision, high rigidity, and high load-bearing capacity, they also introduce some drawbacks. In pursuit of high power density, planetary roller screws in linear servo joints are typically designed with a slender structure and a large length-to-diameter ratio. When robots perform complex motion tasks, such as jumping, significant impacts are generated during start-up or landing, and the enormous axial load leads to substantial axial elastic deformation of the screw. Secondly, nonlinear transmission errors exist. Micrometer-level clearances exist between the planetary rollers, nuts, and screw, which can cause instantaneous impact oscillations during high-frequency reversal or torque-changing output. Finally, since linear servo joints typically drive multi-link mechanisms with large inertia, the amplification effect of multi-stage linkages amplifies the aforementioned axial elastic deformation and the minute displacements caused by nonlinear transmission errors, resulting in severe physical vibrations in the end effector (such as the foot or hand).

[0004] Traditional vibration suppression schemes often employ notch filters with fixed parameters. However, humanoid robots undergo continuous changes in spatial posture during motion, and their dynamic model and system natural frequencies drift in real time with joint configuration. Therefore, notch filters with fixed frequency parameters not only fail to effectively cover changing resonance points but may even introduce additional phase lag when there is frequency mismatch, leading to system control instability. Thus, how to achieve real-time identification and adaptive suppression of vibrations at the end faces of linear servo joints under complex changing posture conditions has become a key issue in improving the motion stability of humanoid robots. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an adaptive suppression method and system for end-effector vibration of a linear servo joint; it aims to solve the problem of residual end-effector oscillation of a planetary roller screw under varying attitude conditions, while effectively distinguishing between mechanical residual vibration and collisions with the external environment, and avoiding interference of the vibration reduction algorithm with normal force control sensing.

[0006] On the one hand, an adaptive suppression method for vibration at the end of a linear servo joint is provided, including: The multi-dimensional signals of the linear servo joint are acquired, and the multi-dimensional signals are decoupled by feature to obtain the preliminary decoupled dynamic force signal. The components of the preliminary decoupled dynamic force signal are extracted to obtain the high-frequency signal component. Determine the cross-correlation coefficient between the theoretical linear acceleration of the linear servo joint and the high-frequency signal component; determine the attenuation factor of the vibration signal based on the high-frequency signal components at adjacent time points; and determine whether the current state is mechanical residual vibration based on the cross-correlation coefficient and the attenuation factor. If the current state is mechanical residual vibration, then real-time tracking of the resonance frequency is initiated to determine the real-time resonance frequency of the linear servo joint in the current posture. Based on the real-time resonant frequency, the transfer function of the notch filter is updated; the original control command is input into the updated notch filter for feedforward shaping, and the shaped compensation current command is output to the servo driver to suppress physical jitter at the end of the linear servo joint.

[0007] On the other hand, a linear servo joint end-effector vibration adaptive suppression system is provided, comprising: The acquisition module is configured to: acquire multi-dimensional signals of the linear servo joint, perform feature decoupling on the multi-dimensional signals to obtain a preliminary decoupled dynamic force signal, and extract components from the preliminary decoupled dynamic force signal to obtain high-frequency signal components. The determination module is configured to: determine the cross-correlation coefficient between the theoretical linear acceleration of the linear servo joint and the high-frequency signal component; determine the attenuation factor of the vibration signal based on the high-frequency signal component at adjacent time points; and determine whether the current state is mechanical residual vibration based on the cross-correlation coefficient and the attenuation factor. The resonance frequency determination module is configured to: if the current state is mechanical residual vibration, start real-time tracking of resonance frequency to determine the real-time resonance frequency of the linear servo joint in the current posture. The vibration suppression module is configured to: update the transfer function of the notch filter based on the real-time resonant frequency; input the original control command into the updated notch filter for feedforward shaping; and output the shaped compensation current command to the servo driver to suppress physical jitter at the end of the linear servo joint.

[0008] Furthermore, an electronic device is also provided, including: Memory, used for non-transitory storage of computer-readable instructions; and Processor, for executing the computer-readable instructions, When the computer-readable instructions are executed by the processor, they perform the method described in the first aspect above.

[0009] In another aspect, a storage medium is also provided for non-transitory storage of computer-readable instructions, wherein when the non-transitory computer-readable instructions are executed by a computer, the method described in the first aspect is performed.

[0010] In another aspect, a computer program product is also provided, including a computer program that, when run on one or more processors, is used to implement the method described in the first aspect above.

[0011] The above technical solution has the following advantages or beneficial effects: (1) Significantly reduces the vibration generated at the end of the linear servo joint during robot movement. This invention uses adaptive notch filtering command shaping technology to accurately extract the high-frequency components that cause resonance in the motor drive command, cutting off the conditions for vibration generation from the physical source, and greatly improving the trajectory stability and positioning accuracy of the robot during movement.

[0012] (2) Significantly improves the safety of robot interaction with complex environments. With the help of the intent recognition mechanism, the system can accurately distinguish between internal mechanical residual vibration and collision with the external real physical environment, avoiding robot instability caused by misjudging the type of force signal.

[0013] (3) This invention solves the problem of natural frequency drift under varying attitudes and achieves adaptive vibration reduction under all operating conditions. The invention uses sliding window frequency domain analysis to track and dynamically update notch filter parameters in real time, ensuring the effectiveness of the algorithm in vibration reduction under any attitude.

[0014] (4) The algorithm does not require the introduction of additional physical hardware and has extremely high engineering application value. This invention is based entirely on closed-loop control command shaping at the software algorithm level, without the need to add bulky hardware such as dampers, effectively controlling the robot's weight and manufacturing cost. Attached Figure Description

[0015] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0016] Figure 1 This is a structural diagram of the linear servo joint described in this invention.

[0017] Figure 2 This is a flowchart of the overall system described in this invention. Detailed Implementation

[0018] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0019] Example 1 This embodiment provides an adaptive suppression method for vibration at the end of a linear servo joint. like Figure 2 As shown, an adaptive vibration suppression method for the end effector of a linear servo joint includes: S101: Acquire multi-dimensional signals of the linear servo joint, and perform feature decoupling on the multi-dimensional signals to obtain preliminary decoupled dynamic force signals. Extract components from the preliminary decoupled dynamic force signals to obtain high-frequency signal components. S102: Determine the cross-correlation coefficient between the theoretical linear acceleration of the linear servo joint and the high-frequency signal component; determine the attenuation factor of the vibration signal based on the high-frequency signal component at adjacent moments; and determine whether the current state is mechanical residual vibration based on the cross-correlation coefficient and the attenuation factor. S103: If the current state is mechanical residual vibration, start real-time tracking of the resonance frequency to determine the real-time resonance frequency of the linear servo joint in the current posture. S104: Based on the real-time resonant frequency, update the transfer function of the notch filter; input the original control command into the updated notch filter for feedforward shaping, and output the shaped compensation current command to the servo driver to suppress physical jitter at the end of the linear servo joint.

[0020] Further, step S101: acquiring multi-dimensional signals from the linear servo joint, and performing feature decoupling on the multi-dimensional signals to obtain preliminarily decoupled dynamic force signals, including: S101-1: At a preset sampling frequency, synchronously acquire the rotor position signal of the frameless torque motor and the original force signal output by the force sensor; the force sensor is installed at the end of the linear servo joint; S101-2: Based on the lead of the roller screw, the rotor position signal is mapped to the theoretical linear displacement of the linear servo joint. S101-3: Perform second differentiation and low-pass filtering on the theoretical linear displacement to obtain the theoretical linear acceleration with high-frequency quantization noise removed; S101-4: Based on the theoretical linear acceleration and load mass, determine the theoretical inertial force generated by the load mass when the linear servo joint accelerates. S101-5: Based on the original force signal and the theoretical inertial force, calculate the preliminary decoupled dynamic force signal.

[0021] like Figure 1 As shown, the linear servo joint includes a servo driver, a frameless torque motor, a planetary roller screw, and a force sensor. The linear servo joint employs a highly integrated mechatronic structure. The stator of its frameless torque motor is fixed to the inner wall of the joint housing, while the rotor is coaxially fixed to the outer surface of the planetary roller screw shaft for direct power coupling. Radial support and axial limiting are provided by bearing assemblies at both ends of the housing. When the motor drives the screw shaft to rotate, the screw nut utilizes its planetary motion to convert rotational torque into linear thrust, driving the front-end fixed push rod to reciprocate axially along the internal anti-rotation guide rail. Simultaneously, a force sensor is coaxially mounted at the output end of the push rod to sense the interaction force in real time. The servo driver, located within the housing, electrically connects the motor, encoder, and force sensor, thus physically constructing an integrated, fully closed-loop power chain from drive and transmission to end-effector sensing.

[0022] For example, step S101: acquiring multi-dimensional signals from the linear servo joint and performing feature decoupling on the multi-dimensional signals to obtain a preliminarily decoupled dynamic force signal, specifically includes: The rotor position signal of the frameless torque motor is synchronously acquired by a servo driver at a preset sampling frequency. and the raw force signal output by the force sensor installed at the end of the linear servo joint. ; Based on the lead of the roller screw , rotor position signal Theoretical linear displacement mapped to a linear servo joint The mapping relationship is as follows: ; For linear displacement By performing second-order differentiation and low-pass filtering, the theoretical linear acceleration, after removing high-frequency quantization noise, is obtained. The specific processing method is as follows: The theoretical linear displacement... Mapping to the complex frequency domain yields By including the second-order differential operator By performing a composite operation with the transfer function of the low-pass filter, high-frequency quantization noise is smoothed while extracting acceleration, thereby obtaining the theoretical linear acceleration signal. .

[0023]

[0024] Where T is the low-pass filter time constant, used to suppress encoder quantization noise.

[0025] Original force signal This includes the inertial force generated by the load mass during the accelerated motion of a linear servo joint; therefore, a load mass-based... The rigid body dynamics feedforward compensation term is used to calculate the theoretical inertial force. :

[0026] Obtain the dynamic force signal of preliminary decoupling for: .

[0027] Further, the step of extracting components from the initially decoupled dynamic force signal to obtain high-frequency signal components includes: The initially decoupled dynamic force signal is processed using a high-pass filter to obtain a high-frequency signal; The initially decoupled dynamic force signal is processed using a low-pass filter to obtain a low-frequency signal.

[0028] Furthermore, the processing using a high-pass filter specifically includes:

[0029] in, The representation of the initially decoupled dynamic force signal in the complex frequency domain; This represents the high-frequency force signal component separated by high-pass filtering; This is the filtering time constant of the high-pass filter.

[0030] Furthermore, the processing using a low-pass filter specifically includes:

[0031] in, The representation of the initially decoupled dynamic force signal in the complex frequency domain; This represents the low-frequency force signal component separated by low-pass filtering; This is the filtering time constant of the low-pass filter.

[0032] To achieve accurate extraction of vibration characteristics while avoiding interference with normal force control, a frequency domain decoupling strategy was adopted. Through different filtering stages, the dynamic force signal was decomposed into low-frequency components representing steady-state interaction forces and high-frequency components representing abnormal vibrations.

[0033] For example, the step of extracting components from the initially decoupled dynamic force signal to obtain high-frequency signal components includes: For the obtained dynamic force signal The system processes the signals to separate the robot's normal force sensing signals from the target force signals suspected of vibration. Specifically, it uses low-pass and high-pass filters to separate the dynamic force signals. The low-frequency and high-frequency signal components are extracted.

[0034] The low-frequency signal component represents the gravity, static pressure of the load, and the actual contact force of the robot's smooth interaction with the outside world, which is the normal force sensing signal. It is directly retained and sent to the robot's underlying posture balance and impedance control algorithm to ensure the accuracy of force control.

[0035] The high-frequency signal component may be an abnormal vibration characteristic caused by the elastic deformation and mechanical backlash of the planetary roller screw, which is the target vibration reduction signal. This signal is then output to the subsequent intent recognition and status determination module.

[0036] To prevent the vibration reduction algorithm from misinterpreting normal foot contact with the ground or collisions with the external environment as mechanical residual vibrations, thus causing force control failure, an intent recognition and state determination mechanism is introduced here.

[0037] The beneficial effects of the above technical solution are: it has carried out multi-dimensional signal acquisition and feature decoupling, separated low-frequency effective force signals from high-frequency suspected vibration signals, and laid the data foundation that takes into account both the underlying force control and the upper vibration reduction.

[0038] Further, S102: determining the cross-correlation coefficient between the theoretical linear acceleration of the linear servo joint and the high-frequency signal component specifically includes: Within the sliding time window T, the theoretical linear acceleration of the joint is calculated in real time. High-frequency signal components cross-correlation coefficient : ; in, The variable represents the integral variable, which represents the continuous time quantity within the observation window [tT, t].

[0039] Cross-correlation number To determine whether the vibration signal is correlated with the motor motion, The closer the correlation is to 1 or -1, the stronger the correlation.

[0040] Further, step S102: determining the attenuation factor of the vibration signal based on the high-frequency signal components at adjacent times, specifically includes: The signal generated by mechanical residual vibration decays periodically. If the mechanical residual vibration is an impact, the generated signal is a single pulse; calculate the high-frequency signal components. The ratio of two adjacent envelope peaks is defined as the attenuation factor. :

[0041] in, and For two adjacent peak times, express Peak envelope value at time 10:00 express The peak value of the envelope at any given time.

[0042] Further, step S102: determining whether the current state is mechanical residual vibration based on the cross-correlation coefficient and the attenuation factor, specifically includes: Preset cross-correlation threshold and attenuation factor threshold Make a judgment: if and This indicates that the force fluctuation is highly correlated with the motor's operation and exhibits periodic decay, thus determining the current state as mechanical residual vibration; if Or the force signal does not decay periodically, for example... This indicates that the current state is not mechanical residual vibration, the sudden change in force is unrelated to the motor's action, and it is a real foot contact with the ground or a collision with the environment.

[0043] When the current state is determined to be mechanical residual vibration, real-time resonant frequency tracking is initiated. This is because when the robot is in different postures, the posture and forces on the linear servo joints also change, causing their natural frequencies to drift. To accurately eliminate vibration, the real-time resonant frequency must be determined.

[0044] The beneficial effects of the above technical solution are: it performs intent recognition and state determination, accurately distinguishes between internal mechanical residual vibration and external environmental collision, prevents the vibration reduction algorithm from being triggered erroneously, and ensures the safety of robot control.

[0045] Further, in step S103: if the current state is mechanical residual vibration, then real-time tracking of the resonance frequency is initiated to determine the real-time resonance frequency of the linear servo joint in the current posture, specifically including: Extracting high-frequency signal components of length N within a sliding time window Using Fourier transform to calculate high-frequency signal components at different frequencies Amplitude spectrum below The core conversion formula is:

[0046] in, Sampling frequency, It is the imaginary unit.

[0047] The energy of mechanical vibration is highly concentrated at the system's natural frequency. Based on the amplitude spectrum, a frequency scan is performed within the system's set high-frequency sensitive region (e.g., 20Hz to 50Hz) to find the peak point. The frequency corresponding to the peak point with the highest energy is the real-time resonant frequency of the linear servo joint in its current posture. .

[0048] For the searched real-time resonant frequency A first-order smoothing filter is applied to remove frequency anomalies caused by slight noise from the sensor, resulting in a smooth output resonant frequency. .

[0049] The beneficial effects of the above technical solution are: real-time tracking of the resonant frequency, dynamic locking of the instantaneous resonant frequency under joint posture changes, overcoming the problem of inherent frequency drift, and providing precise control parameters for subsequent vibration reduction control.

[0050] Further, S104: updating the transfer function of the notch filter based on the real-time resonant frequency, specifically includes: The resonant frequency of the smooth output Convert to angular frequency And based on angular frequency Real-time update of the transfer function of the notch filter :

[0051] in, The molecular damping ratio of the notch filter determines the notch depth. The damping ratio in the denominator determines the notch bandwidth. Let denote the Laplace operator, and represent the complex frequency variable in the complex frequency domain.

[0052] When the instantaneous resonant frequency is determined Then, the real-time resonant frequency will be... As a time-varying parameter, it is input into the servo driver to dynamically suppress residual mechanical vibration.

[0053] Further, S104: inputting the original control command into the updated notch filter for feedforward shaping includes: In the current loop of a frameless torque motor, the original control command It contains both low-frequency motion control components and high-frequency harmonic components that cause mechanical resonance; the original control command Transfer function of notch filter Perform a multiplication operation to obtain the shaped compensation current command.

[0054] Notch filter At its center frequency It provides a very small amplitude gain, thereby selectively attenuating the high-frequency harmonic components that cause mechanical resonance while preserving the integrity of the low-frequency control components.

[0055] Further, step S104: outputting the shaped compensation current command to the servo driver to suppress physical jitter at the end of the linear servo joint, specifically including: The shaped compensation current command is output to the servo driver.

[0056] The servo driver performs the following physical actions to achieve vibration reduction control: First, the servo driver maps the shaped compensation current to the target setpoint of the motor current loop; second, the current controller inside the driver performs closed-loop calculations based on the target setpoint and the real-time collected feedback current to generate the corresponding voltage control signal; finally, the power drive unit converts the control signal into the physical current that actually acts on the motor.

[0057] Since the compensation current after shaping has avoided the mechanical resonance point of the system in the frequency domain, the torque output by the motor does not have the energy characteristics to excite mechanical resonance at the source. Thus, the active suppression of residual vibration at the end of the linear joint is achieved without reducing the system response bandwidth.

[0058] Because the reshaped compensation current command has precisely removed the high-frequency components that cause resonance in the linear motor, the motor output torque no longer excites resonance in the mechanical system. The physical jitter at the end of the linear servo joint then rapidly decays, thus ensuring stable end-effector trajectory and high-precision positioning of the humanoid robot under dynamic posture changes.

[0059] The beneficial effects of the above technical solution are as follows: It performs notch filter command shaping and vibration suppression, dynamically updates the notch filter parameters using real-time frequency, accurately filters out high-frequency excitation components in the motor command, and achieves rapid vibration reduction at the end of the linear servo joint. It also performs notch filter command shaping and closed-loop compensation, updates the frequency parameters of the notch filter based on the solved real-time resonant frequency, and shapes the current command to achieve adaptive suppression of end-effector vibration.

[0060] This invention mainly comprises the following four stages: The first stage involves multi-dimensional signal acquisition and feature decoupling, separating low-frequency effective force signals from high-frequency suspected vibration signals, laying a data foundation that balances underlying force control and upper-level vibration reduction. The second stage involves intent recognition and state determination, accurately distinguishing between internal mechanical residual vibration and external environmental collisions, preventing false triggering of the vibration reduction algorithm, and ensuring the safety of robot control. The third stage involves real-time tracking of the resonant frequency, dynamically locking the instantaneous resonant frequency under joint posture changes, overcoming the problem of inherent frequency drift, and providing precise control parameters for subsequent vibration reduction control. The fourth stage involves notch filter command shaping and vibration suppression, dynamically updating the notch filter parameters using real-time frequency to accurately filter out high-frequency excitation components in joint commands, achieving rapid vibration reduction at the end of the linear servo joint.

[0061] Multidimensional signal acquisition and feature decoupling are employed to acquire dynamic force signals at the end effector of a linear servo joint, and suspected vibration signals are decoupled based on vibration characteristics. The acquisition of residual mechanical vibration signals is determined based on time-series correlation coefficients and waveform attenuation characteristics. The resonant frequency of the vibration signal is calculated in real time using Fourier transform. The frequency parameters of the notch filter are updated in real time, and the current control signal of the linear servo joint is shaped using the notch filter to achieve adaptive suppression of vibration at the end effector of the linear servo joint. After repeated experiments, the adaptive vibration suppression method for the end effector of a linear servo joint proposed in this invention can effectively suppress residual vibration at the end effector of the linear servo joint and has certain practical value.

[0062] Example 2 This embodiment provides a linear servo joint end-effector vibration adaptive suppression system, including: The acquisition module is configured to: acquire multi-dimensional signals of the linear servo joint, perform feature decoupling on the multi-dimensional signals to obtain a preliminary decoupled dynamic force signal, and extract components from the preliminary decoupled dynamic force signal to obtain high-frequency signal components. The determination module is configured to: determine the cross-correlation coefficient between the theoretical linear acceleration of the linear servo joint and the high-frequency signal component; determine the attenuation factor of the vibration signal based on the high-frequency signal component at adjacent time points; and determine whether the current state is mechanical residual vibration based on the cross-correlation coefficient and the attenuation factor. The resonance frequency determination module is configured to: if the current state is mechanical residual vibration, start real-time tracking of resonance frequency to determine the real-time resonance frequency of the linear servo joint in the current posture. The vibration suppression module is configured to: update the transfer function of the notch filter based on the real-time resonant frequency; input the original control command into the updated notch filter for feedforward shaping; and output the shaped compensation current command to the servo driver to suppress physical jitter at the end of the linear servo joint.

[0063] It should be noted that the acquisition module, determination module, resonance frequency determination module, and vibration suppression module described above correspond to steps S101 to S104 in Embodiment 1. The examples and application scenarios implemented by these modules and their corresponding steps are the same, but they are not limited to the content disclosed in Embodiment 1. It should also be noted that these modules, as part of the system, can be executed in a computer system, such as a set of computer-executable instructions.

[0064] The descriptions of each embodiment in the above embodiments have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0065] The proposed system can be implemented in other ways. For example, the system embodiments described above are merely illustrative, and the division of modules described above is only a logical functional division. In actual implementation, there may be other division methods. For example, multiple modules may be combined or integrated into another system, or some features may be ignored or not executed.

[0066] Example 3 This embodiment also provides an electronic device, including: one or more processors, one or more memories, and one or more computer programs; wherein, the processor is connected to the memory, and the one or more computer programs are stored in the memory. When the electronic device is running, the processor executes the one or more computer programs stored in the memory to cause the electronic device to perform the method described in Embodiment 1.

[0067] It should be understood that in this embodiment, the processor can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc.

[0068] Memory may include read-only memory and random access memory, and provides instructions and data to the processor. A portion of memory may also include non-volatile random access memory. For example, memory may also store information about the device type.

[0069] In the implementation process, each step of the above method can be completed by the integrated logic circuits in the processor hardware or by software instructions.

[0070] The method in Embodiment 1 can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor. The software modules can reside in readily available storage media in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory; the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method. To avoid repetition, a detailed description is not provided here.

[0071] Those skilled in the art will recognize that the units and algorithm steps described in connection with the various examples of this embodiment can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this invention.

[0072] Example 4 This embodiment also provides a computer-readable storage medium for storing computer instructions, which, when executed by a processor, complete the method described in Embodiment 1.

[0073] Example 5 This embodiment also provides a computer program product, including a computer program that, when executed by a processor, implements the method in Embodiment 1.

[0074] The present invention also provides at least one computer program product tangibly stored on a non-transitory computer-readable storage medium. The computer program product includes computer-executable instructions, such as instructions included in program modules, which execute in a device on a target real or virtual processor to perform the processes / methods described above. Typically, program modules include routines, programs, libraries, objects, classes, components, data structures, etc., that perform specific tasks or implement specific abstract data types. In various embodiments, the functionality of program modules can be combined or divided among program modules as needed. The machine-executable instructions for the program modules can execute within a local or distributed device. In a distributed device, the program modules can reside in both local and remote storage media.

[0075] The computer program code used to implement the methods of the present invention may be written in one or more programming languages. This computer program code may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the computer or other programmable data processing device, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a computer, partially on a computer, as a stand-alone software package, partially on a computer and partially on a remote computer, or entirely on a remote computer or server.

[0076] In the context of this invention, computer program code or related data may be carried by any suitable carrier to enable a device, apparatus, or processor to perform the various processes and operations described above. Examples of carriers include signals, computer-readable media, and the like. Examples of signals may include electrical, optical, radio, sound, or other forms of propagation signals, such as carrier waves, infrared signals, etc.

[0077] Those skilled in the art will recognize that the units and algorithm steps described in conjunction with the embodiments herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0078] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An adaptive suppression method for vibration at the end of a linear servo joint, characterized in that, include: The multi-dimensional signals of the linear servo joint are acquired, and the multi-dimensional signals are decoupled by feature to obtain the preliminary decoupled dynamic force signal. The components of the preliminary decoupled dynamic force signal are extracted to obtain the high-frequency signal component. Determine the cross-correlation coefficient between the theoretical linear acceleration of the linear servo joint and the high-frequency signal component; determine the attenuation factor of the vibration signal based on the high-frequency signal components at adjacent time points; and determine whether the current state is mechanical residual vibration based on the cross-correlation coefficient and the attenuation factor. If the current state is mechanical residual vibration, then real-time tracking of the resonance frequency is initiated to determine the real-time resonance frequency of the linear servo joint in the current posture. The transfer function of the notch filter is updated based on the real-time resonant frequency. The original control command is input into the updated notch filter for feedforward shaping, and the shaped compensation current command is output to the servo driver to suppress physical jitter at the end of the linear servo joint.

2. The adaptive vibration suppression method for the end effector of a linear servo joint as described in claim 1, characterized in that, Acquire multi-dimensional signals from the linear servo joint, and perform feature decoupling on the multi-dimensional signals to obtain a preliminarily decoupled dynamic force signal, including: The rotor position signal of the frameless torque motor and the raw force signal output by the force sensor are acquired synchronously at a preset sampling frequency; the force sensor is installed at the end of the linear servo joint. Based on the lead of the roller screw, the rotor position signal is mapped to the theoretical linear displacement of the linear servo joint. The theoretical linear displacement is subjected to second differentiation and low-pass filtering to obtain the theoretical linear acceleration with high-frequency quantization noise removed. Based on the theoretical linear acceleration and load mass, the theoretical inertial force generated by the load mass during the accelerated motion of the linear servo joint is determined. Based on the original force signal and the theoretical inertial force, the preliminary decoupled dynamic force signal is calculated.

3. The adaptive vibration suppression method for the end effector of a linear servo joint as described in claim 1, characterized in that, Acquire multi-dimensional signals from the linear servo joint, and perform feature decoupling on the multi-dimensional signals to obtain a preliminary decoupled dynamic force signal, specifically including: The rotor position signal of the frameless torque motor is synchronously acquired by a servo driver at a preset sampling frequency. and the raw force signal output by the force sensor installed at the end of the linear servo joint. ; Based on the lead of the roller screw , rotor position signal Theoretical linear displacement mapped to a linear servo joint The mapping relationship is as follows: ； For linear displacement By performing second-order differentiation and low-pass filtering, the theoretical linear acceleration, after removing high-frequency quantization noise, is obtained. ; Original force signal This includes the inertial force generated by the load mass during the accelerated motion of a linear servo joint; therefore, a load mass-based... The rigid body dynamics feedforward compensation term is used to calculate the theoretical inertial force. : ； Obtain the dynamic force signal of preliminary decoupling for: 。 4. The adaptive vibration suppression method for the end effector of a linear servo joint as described in claim 1, characterized in that, Determine the cross-correlation coefficient between the theoretical linear acceleration of the linear servo joint and the high-frequency signal component, specifically including: Within the sliding time window T, the theoretical linear acceleration of the joint is calculated in real time. High-frequency signal components cross-correlation coefficient : ； in, This represents the integral variable, which represents the continuous time quantity within the observation window [tT, t]. The attenuation factor of the vibration signal is determined based on the high-frequency signal components at adjacent times, specifically including: The signal generated by mechanical residual vibration decays periodically. If the mechanical residual vibration is an impact, the generated signal is a single pulse; calculate the high-frequency signal components. The ratio of two adjacent envelope peaks is defined as the attenuation factor. : ； in, and For two adjacent peak times, express Peak envelope value at time 10:00 express The peak value of the envelope at any given time.

5. The adaptive vibration suppression method for the end effector of a linear servo joint as described in claim 1, characterized in that, Based on the cross-correlation coefficient and the attenuation factor, determining whether the current state is mechanical residual vibration specifically includes: Preset cross-correlation threshold and attenuation factor threshold Make a judgment: if and This indicates that the force fluctuation is highly correlated with the motor's operation and exhibits periodic decay, thus determining the current state as mechanical residual vibration; if If the force signal does not decay periodically, it indicates that the current state is not mechanical residual vibration, the sudden change in force is unrelated to the motor action, and it is a real foot contact with the ground or a collision with the environment.

6. The adaptive vibration suppression method for the end effector of a linear servo joint as described in claim 1, characterized in that, If the current state is mechanical residual vibration, then real-time tracking of the resonance frequency is initiated to determine the real-time resonance frequency of the linear servo joint in the current posture, specifically including: Extracting high-frequency signal components of length N within a sliding time window Using Fourier transform to calculate high-frequency signal components at different frequencies Amplitude spectrum below The core conversion formula is: ； in, Sampling frequency, It is the imaginary unit.

7. The adaptive vibration suppression method for the end effector of a linear servo joint as described in claim 1, characterized in that, Based on the real-time resonant frequency, the transfer function of the notch filter is updated, specifically including: The resonant frequency of the smooth output Convert to angular frequency And based on angular frequency Real-time update of the transfer function of the notch filter : ； in, The molecular damping ratio of the notch filter determines the notch depth. The damping ratio in the denominator determines the notch bandwidth. Let denote the Laplace operator, and represent the complex frequency variable in the complex frequency domain.

8. A linear servo joint end-effector vibration adaptive suppression system, characterized in that, include: The acquisition module is configured to: acquire multi-dimensional signals of the linear servo joint, perform feature decoupling on the multi-dimensional signals to obtain a preliminary decoupled dynamic force signal, and extract components from the preliminary decoupled dynamic force signal to obtain high-frequency signal components. The determination module is configured to: determine the cross-correlation coefficient between the theoretical linear acceleration of the linear servo joint and the high-frequency signal component; determine the attenuation factor of the vibration signal based on the high-frequency signal component at adjacent time points; and determine whether the current state is mechanical residual vibration based on the cross-correlation coefficient and the attenuation factor. The resonance frequency determination module is configured to: if the current state is mechanical residual vibration, start real-time tracking of resonance frequency to determine the real-time resonance frequency of the linear servo joint in the current posture. The vibration suppression module is configured to: update the transfer function of the notch filter based on the real-time resonant frequency; input the original control command into the updated notch filter for feedforward shaping; and output the shaped compensation current command to the servo driver to suppress physical jitter at the end of the linear servo joint.

9. An electronic device, characterized in that it comprises: Memory is used to store computer-readable instructions in a non-transitory manner. as well as Processor, for executing the computer-readable instructions, When the computer-readable instructions are executed by the processor, they perform the method described in any one of claims 1-7.

10. A storage medium, characterized in that, Non-transitory storage of computer-readable instructions, wherein when the non-transitory computer-readable instructions are executed by a computer, the method of any one of claims 1-7 is performed.

Images