CN-122008249-A - Method and device for identifying damping of self-adaptive cooperative mechanical arm based on different loads

CN122008249ACN 122008249 ACN122008249 ACN 122008249ACN-122008249-A

Abstract

The invention relates to a method and a device for identifying damping of a self-adaptive cooperative mechanical arm based on different loads, wherein the method comprises the following steps: the method comprises nine core steps of no-load moment reference calibration, end load calculation, load grading, self-adaptive excitation matching, resonance signal acquisition, whole machine resonance parameter extraction, double-window damping identification and stability verification, damping precision verification and load drift self-adaptive compensation, and finally, accurate and stable damping ratio can be obtained under different loads, and reliable parameter support is provided for vibration suppression and positioning precision improvement.

Inventors

HU ZHENWEI
XUE YAFEI
YANG YONGCHAO
YU GANG

Assignees

辰致汽车科技集团有限公司重庆创新研究分公司
辰致汽车科技集团有限公司

Dates

Publication Date: 20260512
Application Date: 20260408

Claims (10)

1. The self-adaptive cooperative mechanical arm damping identification method based on different loads is characterized by comprising the following steps of: The no-load moment reference calibration is to construct an no-load steady-state moment library by collecting no-load joint moment data of the typical gesture of the mechanical arm in the full working space, wherein the no-load steady-state moment library consists of joint reference moment generated by self gravity and steady-state friction force of the mechanical arm and is used for representing the dynamic change rule of the joint moment along with the operation gesture in the no-load state; The end load calculation is carried out, namely the end equivalent load mass under the current working condition is obtained through gesture matching, moment separation and dynamic model mapping based on an empty steady-state moment library obtained through the empty moment reference calibration; load classification, namely dividing working conditions into three types of light load, medium load and heavy load according to the end equivalent load mass obtained by calculating the end load; The adaptive excitation matching is carried out, namely, an excitation frequency band and an amplitude which are matched with the current load are set according to the load grading result and the offset characteristic of the first-order resonant frequency of the mechanical arm under the variable load, and an adaptive excitation condition is provided for the acquisition of the resonant signal; Selecting a dominant joint which is used for leading the vibration of the whole mechanical arm, applying the self-adaptive excitation which is matched with the self-adaptive excitation and is set, synchronously collecting joint moment and angle signals, and combining with separation of an empty steady moment library to obtain a high-frequency moment fluctuation component which is only caused by flexible resonance of the whole mechanical arm; the whole machine resonance parameter extraction, namely processing the high-frequency moment fluctuation component, extracting a resonance frequency and a resonance peak value, and calculating the equivalent damping ratio of the mechanical arm based on the resonance frequency and the resonance peak value; Double window damping identification and stability verification, namely carrying out double window synchronous damping identification on moment data of a dominant joint, combining resonance frequency and resonance peak value obtained by extracting resonance parameters of the whole machine, calculating damping ratio based on a half-power bandwidth method, and establishing quantized identification effectiveness judgment standards to determine final equivalent damping ratio, actual measurement resonance peak value, actual measurement resonance frequency and model prediction frequency; damping accuracy verification, namely calculating a model prediction resonance peak value based on the model prediction frequency and the final equivalent damping ratio, and comparing the model prediction frequency and the model prediction resonance peak value with the actual measurement resonance frequency and the actual measurement resonance peak value to judge the accuracy of a damping ratio calculation result; load drift self-adaptive compensation, namely detecting the load change of the tail end in real time, triggering a re-identification process and caching related data when the load is suddenly changed.
2. The method for identifying the damping of the self-adaptive cooperative mechanical arm based on different loads according to claim 1, wherein the specific implementation process of the no-load moment reference calibration and the end load calculation is as follows: Controlling the cooperative mechanical arm to be in an idle constant-speed operation working condition, collecting multiple groups of joint moment data corresponding to a plurality of typical operation postures in a full working space of the cooperative mechanical arm, and constructing an idle steady-state moment library based on the joint moment data; when the mechanical arm is in a loading operation state, acquiring joint moment signals and joint angle information in the current operation posture in real time, matching the joint moment signals and the joint angle information with the no-load steady-state moment library through a posture interpolation algorithm, and solving to obtain a no-load moment reference value completely corresponding to the current loading posture; subtracting the no-load moment reference value from the real-time acquired on-load joint moment to obtain a pure load moment component only caused by the end load; Substituting the pure load moment component into a Newton-Euler dynamics model, and combining the kinematics and dynamics parameters in the mechanical arm URDF file to finish the mapping calculation from the joint moment to the end load so as to obtain the end equivalent load mass under the current working condition.
3. The method for identifying the damping of the self-adaptive collaborative mechanical arm based on different loads according to claim 2 is characterized in that the typical operation gesture is selected to fully cover the main space direction of the conventional operation of the mechanical arm, the joint moment data corresponding to each group of typical gestures are acquired in a repeated acquisition mode, and the arithmetic average value of the acquired data is taken as the no-load reference moment of each joint in the gesture.
4. The method for identifying the damping of the self-adaptive collaborative mechanical arm based on different loads according to claim 1, wherein the specific criteria of matching the load classification with the self-adaptive excitation are as follows: The light load working condition adopts pseudo-random binary sequence micro-excitation with amplitude less than or equal to +/-0.005 rad and covering 25 Hz-50 Hz frequency band; the medium-load working condition adopts pseudo-random binary sequence micro-excitation with the amplitude being more than +/-0.005 rad and less than +/-0.01 rad and covering the frequency range of 10 Hz-30 Hz; The heavy load working condition adopts pseudo-random binary sequence micro-excitation with the amplitude of +/-0.01 rad and covering the frequency range of 5 Hz-25 Hz.
5. The method for identifying the damping of the self-adaptive collaborative mechanical arm based on different loads according to claim 4, wherein the amplitude of the self-adaptive excitation is subjected to linear self-adaptive adjustment according to the end equivalent load mass, and a specific adjustment rule is as follows: The excitation amplitude is linearly increased by 20% every 5kg of the end load mass, and the upper limit threshold of the excitation amplitude is set to be +/-0.015 rad.
6. The method for identifying the damping of the self-adaptive cooperative mechanical arm based on different loads according to claim 1, wherein the specific flow of the resonant signal acquisition and the complete machine resonant parameter extraction is as follows: Selecting a dominant joint which has the greatest contribution to the vibration of the tail end of the mechanical arm and is used for leading the flexible vibration of the whole machine as an excitation and signal acquisition object, applying self-adaptive excitation to the dominant joint, and synchronously acquiring real-time joint moment signals and joint angle signals; Obtaining the no-load steady-state moment of the dominant joint under the current gesture through interpolation of the no-load steady-state moment library, and subtracting the no-load steady-state moment from the real-time acquired moment of the loaded joint to obtain a high-frequency moment fluctuation component only caused by flexible resonance vibration of the whole machine; processing the high-frequency moment fluctuation component, and extracting resonance parameters of the whole machine, wherein the resonance parameters comprise resonance frequency and resonance peak value; if only a single dominant resonance peak exists in the frequency spectrum, calculating a damping ratio by adopting a half-power bandwidth method based on the resonance peak; If a plurality of obvious resonance peaks formed by superposition of multi-order vibration modes exist in the frequency spectrum, the resonance frequency and the resonance peak corresponding to each vibration mode are extracted, the damping ratio of each mode is calculated by a half-power bandwidth method, and normalization weighting processing is carried out on the resonance peak of each mode as a basis, so that the final equivalent damping ratio is obtained.
7. The method for identifying the damping of the self-adaptive cooperative mechanical arm based on different loads according to claim 1, wherein the specific operations of identifying the damping of the double windows and checking the stability are as follows: After excitation is applied to a single dominant joint, continuously acquiring steady-state moment data of the dominant joint under the condition that the gesture, load and excitation parameters of the mechanical arm are kept constant, equally dividing the acquired steady-state moment data into two continuous sliding windows, and respectively and independently executing damping identification on the two sliding windows; the two sliding windows adopt completely consistent signal preprocessing and frequency domain analysis processes, respectively extract the resonant frequency and the resonant peak value of each window, and calculate the damping ratio of the corresponding window by a half-power bandwidth method; calculating the relative deviation of the resonant frequencies and the damping ratio deviation of the two sliding windows, when the relative deviation of the resonant frequencies is less than 2% and the damping ratio deviation is less than 0.01, taking the arithmetic average value of the damping ratios of the two sliding windows to obtain the final equivalent damping ratio, taking the average value of the resonant peaks of the two sliding windows as the actual measurement resonant peak value, taking the average value model prediction frequency of the frequencies of the two sliding windows, taking the average value of the frequencies of the two sliding windows or the single sliding window resonant frequency as the actual measurement resonant frequency, and if the deviation threshold value is not met, reapplying excitation and collecting moment data until the effectiveness judgment condition is met.
8. The method for identifying the damping of the self-adaptive collaborative mechanical arm based on different loads according to claim 7, wherein in the step of identifying the damping of the double windows and verifying the stability, the method further comprises the step of self-adaptively adjusting the length of the window according to the movement speed of the mechanical arm, in particular, when the movement speed of the tail end of the mechanical arm is greater than 0.5m/s, the length of the sliding window is automatically shortened by 10%.
9. The method for identifying the damping of the self-adaptive collaborative mechanical arm based on different loads according to claim 7, wherein the specific process of damping accuracy verification and load drift self-adaptive compensation is as follows: Based on the inherent characteristics of the low-frequency flexible vibration of the cooperative mechanical arm, the low-frequency flexible vibration is equivalent to a typical second-order vibration system which is a dynamic equivalent model for analyzing the vibration characteristics of the cooperative mechanical arm, the model prediction frequency and the final equivalent damping ratio are substituted into a preset second-order vibration system transfer function, and the model prediction resonance peak value is calculated after the transfer function is converted to a frequency domain; When the actually measured resonant frequency is the resonant frequency of a single sliding window, if the relative frequency deviation is less than 2% and the relative amplitude deviation is less than 5%, judging that the damping ratio calculation result is accurate; When the measured resonance frequency takes the average value of the double sliding windows, if the relative deviation of the amplitude is less than 5%, the damping ratio calculation result is determined to be accurate; When the change amount of the load at the tail end of the mechanical arm is detected to be more than or equal to 0.5kg, the load matching identification process is immediately triggered, and multi-dimensional data of at least two sliding window lengths are synchronously cached.
10. The device for identifying the damping of the self-adaptive cooperative mechanical arm based on different loads is characterized by comprising a memory and a controller, wherein a computer readable program is stored in the memory, and the computer readable program can execute the method for identifying the damping of the self-adaptive cooperative mechanical arm based on different loads according to any one of claims 1 to 9 when the computer readable program is called by the controller.

Description

Method and device for identifying damping of self-adaptive cooperative mechanical arm based on different loads Technical Field The invention relates to the technical field of mechanical arms, in particular to a method and a device for identifying damping of a self-adaptive cooperative mechanical arm based on different loads. Background The cooperative mechanical arm is widely applied to the fields of intelligent manufacturing, precision machining, logistics sorting and the like. The current collaborative mechanical arm gradually develops to the one-machine multifunctional flexible operation direction, various end execution devices such as clamping jaws, sucking discs and execution tools are required to be frequently replaced for adapting to different operation scenes, the end load quality and inertia of the mechanical arm are irregularly changed, a fixed load state cannot be maintained, low-frequency flexible resonance is extremely easy to generate in the operation process of the mechanical arm, the problems of end shake, overlarge track deviation, operation precision reduction and the like are caused, and the operation efficiency and application scene expansion of the collaborative mechanical arm are severely restricted. The damping parameters are core parameters for flexible vibration suppression and resonance active control of the cooperative mechanical arm, and the vibration suppression effect is directly determined by the accuracy and stability of damping identification. The conventional damping identification technology at present has obvious short plates in the aspect of the adaptability of the variable load working condition of the cooperative mechanical arm, and has the following specific defects: 1) The amplitude and the frequency of the excitation signal are not adaptively adjusted according to the load, the excitation is too large in light load and easy to introduce interference, and the resonance peak cannot be effectively excited due to insufficient excitation in heavy load; 2) The lack of an identification accuracy checking mechanism under the condition of no reference can not judge whether the damping identification result is reliable; 3) In actual operation, the load of the mechanical arm has transient mutation, the current excitation strategy and the damping identification algorithm do not form a cooperative optimization mechanism, and high-precision damping identification is difficult to continuously guarantee under variable load working conditions. What is needed is a new identification method that can accurately measure and calculate the end load, adaptively match excitation signals based on the end load, stably identify damping parameters, and have load drift compensation. Disclosure of Invention The invention aims to provide a method and a device for identifying damping of an adaptive collaborative mechanical arm based on different loads, which can accurately measure and calculate the end load, adaptively match excitation signals based on the end load, stably identify damping parameters and have load drift compensation. In a first aspect, the method for identifying damping of the adaptive collaborative mechanical arm based on different loads comprises the following steps: The no-load moment reference calibration is to construct an no-load steady-state moment library by collecting no-load joint moment data of the typical gesture of the mechanical arm in the full working space, wherein the no-load steady-state moment library consists of joint reference moment generated by self gravity and steady-state friction force of the mechanical arm and is used for representing the dynamic change rule of the joint moment along with the operation gesture in the no-load state; The end load calculation is carried out, namely the end equivalent load mass under the current working condition is obtained through gesture matching, moment separation and dynamic model mapping based on an empty steady-state moment library obtained through the empty moment reference calibration; load classification, namely dividing working conditions into three types of light load, medium load and heavy load according to the end equivalent load mass obtained by calculating the end load; The adaptive excitation matching is carried out, namely, an excitation frequency band and an amplitude which are matched with the current load are set according to the load grading result and the offset characteristic of the first-order resonant frequency of the mechanical arm under the variable load, and an adaptive excitation condition is provided for the acquisition of the resonant signal; Selecting a dominant joint which is used for leading the vibration of the whole mechanical arm, applying the self-adaptive excitation which is matched with the self-adaptive excitation and is set, synchronously collecting joint moment and angle signals, and combining with separation of an empty steady moment library to obtain a high-frequency moment fluctu