A synchronous belt drive mechanical arm start-stop swing suppression method and system

Abstract

The application provides a synchronous belt transmission mechanical arm start-stop swing suppression method and system, and relates to the technical field of mechanical arms. The synchronous belt transmission mechanical arm start-stop swing suppression method comprises the following steps: acquiring real-time vibration data of the synchronous belt transmission mechanical arm; acquiring a first swing amount-time correspondence relationship based on the real-time vibration data; acquiring a second swing amount-time correspondence relationship based on the first swing amount-time correspondence relationship; and adjusting the rotation mode of a driving wheel according to the first swing amount-time correspondence relationship and the second swing amount-time correspondence relationship, so as to suppress the swing of the synchronous belt transmission mechanical arm. The synchronous belt transmission mechanical arm start-stop swing suppression method and system solve the technical problem that the vibration value of the synchronous belt transmission mechanical arm is high when the synchronous belt transmission mechanical arm starts and stops due to the change of the tension and stretching amount of the synchronous belt.

Description

Technical Field

[0001] This invention relates to the field of robotic arm technology, specifically to a method and system for suppressing the start-stop swaying of a synchronous belt-driven robotic arm. Background Technology

[0002] As a key piece of equipment in the field of automation, the development of robotic arms began with the rise of industrial robot technology in the mid-20th century, initially used primarily in repetitive tasks such as automobile manufacturing. With breakthroughs in sensor technology, motion control algorithms, and artificial intelligence, modern robotic arms are now widely used in precision machining, medical surgery, logistics sorting, and hazardous environments. Synchronous belt drives, as the main transmission method for robotic arms, offer advantages such as high precision, low noise, and no need for lubrication. However, due to the elastic deformation characteristics of synchronous belts, the tension and stretching of the synchronous belt on both sides of the synchronous pulley are different during transmission.

[0003] Taking the stopping of a robotic arm as an example, when the robotic arm moves, the drive wheel rotates counterclockwise, driving the synchronous belt to rotate. At this time, the tension on the upper side of the synchronous belt is greater than that on the lower side. When the robotic arm receives a stop signal, the drive wheel stops rotating, while the driven wheel continues to rotate due to the inertia of the robotic arm. During this time, the tension and stretching on the lower side of the synchronous belt gradually increase, while the tension and stretching on the upper side gradually decrease. When the tension on the lower side of the synchronous belt gradually increases to exceed the inertia of the driven wheel, due to the principle of elasticity, the tension and stretching on the lower side of the synchronous belt gradually decrease, while the tension and stretching on the upper side gradually increase. This process continues until the tension and inertia of the synchronous belt reach equilibrium. Due to the change in the stretching of the synchronous belt, when the robotic arm receives a stop signal, the driven wheel does not immediately stop completely, but rather reciprocates with the change in the stretching of the synchronous belt, with the reciprocating angle gradually decreasing until it stops. The feedback to the robotic arm is that after receiving a stop signal, the arm does not stop immediately but continues to swing back and forth, with the swing angle gradually decreasing until it stops. The situation is similar when the robotic arm starts up, but because the driving wheel is in a state of uniform acceleration during startup, the driven wheel experiences both reciprocating swing motion and uniform acceleration. Both of these motions affect the final operating state of the robotic arm. In a synchronous belt-driven robotic arm, the rotation of the driven and driving wheels is not completely synchronized during startup and shutdown due to changes in the tension and stretching of the synchronous belt. This results in the robotic arm's motion during startup and shutdown not being in a state of uniform acceleration or deceleration. Consequently, the vibration value during startup and shutdown is higher than during uniform acceleration and deceleration, reaching its peak during the entire operation of the robotic arm.

[0004] Therefore, when a synchronous belt-driven robotic arm stops, it does not stop immediately but instead undergoes a reciprocating swinging motion, with the swing angle gradually decreasing until it comes to a complete stop. Upon startup, it is not a uniformly accelerated motion but a superposition of uniform acceleration and reciprocating swinging motion. During startup and shutdown, it is not a state of uniform acceleration or deceleration, and the vibration value of the robotic arm is highest during these times. In many precision machining industries, there are strict requirements for the vibration value of robotic arms during operation; reducing the vibration value of the robotic arm during startup and shutdown is equivalent to reducing the vibration value of the robotic arm throughout its entire operation.

[0005] In existing technologies, conventional methods for suppressing start-stop sway in synchronous belt-driven robotic arms include reducing the weight the arm can transport or selecting a larger synchronous belt. However, these methods increase the size and manufacturing cost of the robotic arm when transporting the same weight. Alternatively, increasing the rigidity of the drive motor can suppress start-stop sway, but this requires a larger motor and increases manufacturing costs. Another method is to increase the tension of the synchronous belt, but higher tension reduces the lifespan of the synchronous belt.

[0006] Therefore, existing technologies need further development. Summary of the Invention

[0007] The purpose of this invention is to overcome the above-mentioned technical deficiencies and provide a method and system for suppressing the swaying of a synchronous belt driven robotic arm during start-up and stop, so as to solve the technical problem in the related art that the mechanical arm with synchronous belt driven robotic arm has a high vibration value during start-up and stop due to the changes in the tension and stretching of the synchronous belt.

[0008] To achieve the above technical objectives, the present invention adopts the following technical solution: a method for suppressing the swaying of a synchronous belt-driven robotic arm during start-stop, comprising: acquiring real-time vibration data of the synchronous belt-driven robotic arm; acquiring a first sway amount versus time correspondence based on the real-time vibration data; acquiring a second sway amount versus time correspondence based on the first sway amount versus time correspondence; and adjusting the rotation mode of the drive wheel according to the first sway amount versus time correspondence and the second sway amount versus time correspondence to suppress the swaying of the synchronous belt-driven robotic arm.

[0009] Furthermore, the synchronous belt driven robotic arm includes a first robotic arm and a second robotic arm; the first swing amount is the swing displacement y of the second robotic arm; the second swing amount is the swing angle θ of the second robotic arm.

[0010] Furthermore, the method for obtaining the correspondence between the first swing amount and time includes: repeatedly collecting the first swing amount y of the second robotic arm at preset time points. ij The first swing dataset Y is obtained.i Based on the first oscillation dataset Y i Obtain the first swing representative quantity y of the second robotic arm. i * Based on the first swing representative quantity and the preset time point corresponding to the first swing representative quantity, the correspondence between the first swing quantity and the time is obtained.

[0011] Furthermore, the acquisition of the first swing representative quantity y of the second robotic arm i * The method includes: defining the first oscillation dataset Y i Each of the first swing amounts y in ij Gaussian kernel density ρ j ; Calculate the Gaussian kernel density ρ for each of the above. j Greater than the first swing amount y ij The point and the first swing amount y ij Minimum distance δ between j Based on the minimum distance δ j Obtain each of the first swing amounts y ij The corresponding cluster center metric γ ij ; having the largest cluster center metric γ ij The first swing amount y ij The first oscillation representative quantity y i * .

[0012] Further, obtain the cluster center metric γ. ij The methods include: ;

[0013] Where, γ ij ρ is a measure of cluster center, where α is a weighting factor, α∈[0,1], and ρ ij Let be the local density, representing point y. ij The density of surrounding points, δ ij The relative distance represents point y. ij The minimum distance to points that are denser than it.

[0014] Furthermore, the method for obtaining the correspondence between the second swing amount and time includes: calculating the second swing amount based on the first swing amount; the method for calculating the second swing amount is θ=arcsiny / L, where θ is the second swing amount, y is the swing displacement y of the second robotic arm, and L is the length of the second robotic arm.

[0015] Furthermore, the method for adjusting the rotation mode of the drive wheel includes: obtaining the input oscillation amount θ1 based on the synchronous gear tooth ratio between the drive wheel and the driven wheel; obtaining the correspondence between the input oscillation amount and time based on the correspondence between the second oscillation amount and time; and inputting the correspondence between the input oscillation amount and time into a preset driver program to obtain a cam curve.

[0016] Furthermore, the method for obtaining the input swing amount θ1 specifically involves: obtaining the synchronous gear tooth ratio 1:n between the driving wheel and the driven wheel; the input swing amount θ1 = θ / n = (arcsiny / L) / n.

[0017] A system for suppressing the swaying of a synchronous belt-driven robotic arm during start-stop operation is also provided, comprising: a data acquisition module for acquiring real-time vibration data of the synchronous belt-driven robotic arm; a first calculation module for acquiring a first sway amount versus time relationship based on the real-time vibration data; a second calculation module for acquiring a second sway amount versus time relationship based on the first sway amount versus time relationship; and a suppression module for adjusting the rotation mode of the drive wheel according to the first sway amount versus time relationship and the second sway amount versus time relationship, thereby suppressing the swaying of the synchronous belt-driven robotic arm.

[0018] A computer-readable storage medium is also provided, on which computer-readable instructions are stored, which, when executed by a processor, implement the various steps of the synchronous belt driven robotic arm start-stop swing suppression method or the synchronous belt driven robotic arm start-stop swing suppression system as described in any of the preceding claims.

[0019] Beneficial effects

[0020] 1. The synchronous belt drive robotic arm start-stop swing suppression method of the present invention can predict the swing trend based on high-precision representative displacement data and issue instructions in advance to adjust the drive wheel, so as to achieve advanced and precise suppression, rather than a lagging and passive response, which greatly improves the efficiency and effect of vibration suppression, significantly reduces the time required for the swing to converge to a stable state, and improves the working cycle and efficiency of the robotic arm.

[0021] 2. The synchronous belt drive robotic arm start-stop swing suppression method of the present invention, based on a precise representative value curve, allows the control system to plan a smooth and optimal vibration suppression trajectory. The motion command of the drive wheel changes continuously and smoothly. Its purpose is to allow the robotic arm to smoothly return to a standstill along the ideal decay curve, rather than abruptly "brake". It does not require the robotic arm to adjust the tension of the synchronous belt to a high state for operation, reducing the impact on the entire mechanical structure, helping to reduce component wear, extend equipment life, and improve motion accuracy. At the same time, the motion process is also quieter and smoother.

[0022] 3. The synchronous belt drive robotic arm start-stop swing suppression method of the present invention can increase the transmission weight of robotic arms of the same size without affecting transmission stability or increasing transmission vibration; it can also enable the selection of smaller synchronous belts for robotic arms transmitting the same weight, reducing the size of the robotic arm without affecting transmission stability or increasing transmission vibration. Attached Figure Description

[0023] Figure 1 This is a flowchart of the method for suppressing the start-stop swing of a synchronous belt-driven robotic arm used in an embodiment of the present invention;

[0024] Figure 2 This is a schematic diagram of the start-stop swing suppression system for a synchronous belt-driven robotic arm used in an embodiment of the present invention;

[0025] Figure 3 This is a schematic diagram of the structure of the synchronous belt drive robotic arm used in an embodiment of the present invention;

[0026] Figure 4 This is a schematic diagram illustrating the effect of suppressing the start-stop swing of the synchronous belt-driven robotic arm used in an embodiment of the present invention.

[0027] The above figures include the following reference numerals:

[0028] 1. First robotic arm; 2. Second robotic arm; 3. Drive wheel; 4. Synchronous belt; 5. Swinging measurement point. Detailed Implementation

[0029] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0030] According to an embodiment of the present invention, a method for suppressing the start-stop sway of a synchronous belt-driven robotic arm is provided. Please refer to [link to relevant documentation]. Figures 1 to 4 ,include:

[0031] S100 acquires the real-time vibration data of the synchronous belt driven robotic arm;

[0032] Specifically, the synchronous belt driven robotic arm includes a first robotic arm and a second robotic arm;

[0033] Methods for obtaining real-time vibration data of synchronous belt driven robotic arms include:

[0034] In this embodiment, see Figure 3 The X-axis represents the position of the second robotic arm 2 in its natural state. The second robotic arm 2 has a swing measurement point 5. The relative distance between the swing measurement point 5 and the X-axis after swinging is y, and the distance between the swing measurement point 5 and the rotation point of the second robotic arm 2 is L.

[0035] In practice, based on the swing displacement of the second robotic arm 2 at the swing measurement point 5 when it stops, as detected by the vibration sensor, the curve of the swing displacement y of the swing measurement point 5 as a function of time t can be obtained, and then the corresponding relationship table of y as a function of time t can be output by the vibration sensor.

[0036] Preferably, in this embodiment, the sampling frequency of the vibration sensor is set to 10 times per second, so that the value of the swing displacement y is output every 0.1 seconds. As a large, high-inertia device, the residual swing of the robotic arm after it stops is usually a low-frequency, decaying vibration. Using a sampling rate of 10 times per second can effectively capture the typical swing frequency after the robotic arm stops, ensuring that key motion information is not lost. This ensures that the swing process can be completely described, while avoiding information loss caused by too low a sampling rate.

[0037] Based on the real-time vibration data, S200 obtains the correspondence between the first oscillation amount and time;

[0038] Specifically, the first swing amount is the swing displacement y of the second robotic arm.

[0039] In this embodiment, the Density Peak Clustering (DPC) algorithm is used to process multiple sets of correspondence tables of the swing displacement y changing with time t obtained from multiple measurements, so as to obtain the final correspondence table of the first swing amount and time.

[0040] Specifically, the methods for obtaining the correspondence between the first swing amount and time include:

[0041] The first swing amount y of the second robotic arm at preset time points was collected multiple times. ij The first swing dataset Y is obtained. i ;

[0042] Based on the first swing dataset Y i Obtain the first swing representative quantity y of the second robotic arm. i * ;

[0043] Wherein, the first swing representative quantity y of the second robotic arm is obtained i * The methods include:

[0044] Define the first oscillation dataset Y i Each of the first swing amounts y in ij Gaussian kernel density ρj ;

[0045] Calculate the Gaussian kernel density ρ for each of the above. j Greater than the first swing amount y ij The point and the first swing amount y ij Minimum distance δ between j ;

[0046] Based on the minimum distance δ j Obtain each of the first swing amounts y ij The corresponding cluster center metric γ ij ;

[0047] Among them, the cluster center metric is obtained. The methods include:

[0048] ;

[0049] Where, γ ij ρ is a measure of cluster center, where α is a weighting factor, α∈[0,1], and ρ ij Let be the local density, representing point y. ij The density of surrounding points, δ ij The relative distance represents point y. ij The minimum distance to points that are denser than it.

[0050] Specifically, α is a weighting factor used to control density and distance, α∈[0,1]. When α=1, only density is considered, which is equivalent to "selecting the most crowded point". When α=0, only distance is considered, which is equivalent to "selecting the most isolated point". The multiplication of local density and relative distance means: only points that are simultaneously high in density and far from other high-density points are considered. Only then will the cluster be large, thus automatically eliminating "high-density but attached points" and "isolated noise points" and obtaining the cluster center.

[0051] To measure "liveliness / clustering" (how many neighbors are around). The product measures "independence / significance" (whether it is far from other higher density points, whether it is an independent peak). The significance of the product is that it satisfies both conditions simultaneously. Multiplication (especially the geometric mean property) amplifies points where both are large, while the product is suppressed when either indicator is small (i.e., both conditions must be met). The parameter α is introduced to control ρ. ij and δ ij The weights of α and γ differ: a larger α emphasizes density (preferring more crowded points); a smaller α emphasizes independence (preferring more isolated / more peak-like points); in short, only points possessing both "high density" and "independence" are considered suitable. ij The largest cluster is the one that is identified as the cluster center.

[0052] Based on the first swing representative quantity and the preset time point corresponding to the first swing representative quantity, the correspondence between the first swing quantity and the time is obtained.

[0053] Using the above settings, the trend of oscillation is predicted based on high-precision representative displacement data. By assigning different weights to data points of different quality and reliability, the clustering results are more inclined to trust high-quality data, thus obtaining more accurate and representative cluster centers. The weighted cluster centers are no longer simple "geometric centers" but "weighted centroids," which are closer to the regions of high-weight, high-reliability data points, making the final representative displacement value y more accurate. i * The calculations are more scientific and precise.

[0054] Furthermore, traditional clustering algorithms use density to eliminate outliers, but this is not precise enough. This embodiment introduces a weighting factor α to provide a second line of defense. The weighting rules are set according to the swing pattern of the robotic arm: α=1 means only density is considered, α=0 means only distance is considered, and α=0.5 means both density and distance are considered. α flexibly adapts to different types of distributions and noise, and its influence in subsequent iterations will be greatly reduced. By automatically assigning extremely low weights to potential outliers, their negative impact on the clustering results is fundamentally suppressed, improving the robustness of the algorithm. Example

[0055] In this embodiment, a total of m time points t1, t2, t3, ..., t are set. m The time interval is set to Δt = 0.1s.

[0056] Specifically, each time t i The data set Y for that time point was composed of n displacements (i.e., the first oscillation y). i ={y i1 ,y i2 ,...y in}, i=1,2,...,n.

[0057] It should be noted that this embodiment requires each time t i Find the corresponding representative displacement y. i * This forms the table showing the relationship between the final first swing amount and time. .

[0058] Specifically, for each point in time Each corresponding sampling point y ij ∈Y i Its Gaussian kernel density is defined as ρ j :

[0059] ;

[0060] Where k is the k-th displacement measured at the same time point, y ij For the tth i The j-th displacement value collected within a time interval, y ik For the tth i The k-th displacement value collected within a time interval, d c This represents the cutoff distance (controlling the width of the density kernel, typically taken as 10%-20% of the data standard deviation).

[0061] Furthermore, calculate y for each sampling point. ij Minimum distance to all points with a density greater than its own:

[0062] If there exist points with a higher density than the current point: find all points with higher density; calculate their distances to the current point; take the minimum value, which is δ. ij If the current point is the point with the highest density: let its δ... ij This represents the global maximum distance.

[0063] It should be noted that if ρ j If it is not the point of maximum density, then δ j It is equal to the distance of the nearest point with a higher density than it, used to measure y. ij Point and a point in the "more crowded area" (y ik The nearest distance; otherwise δ j If it is the point of maximum density, then δ j It is equal to the maximum distance to all points, since there are no points denser than it, so it is defined as the global maximum value to ensure that it can be used as a candidate cluster center.

[0064] Where, δ j Representing point y ij The minimum distance to a higher density point.

[0065] It should be noted that if the sampling point y ij If it is the point of highest density, then δ j Let this be the distance between this point and the point furthest from it.

[0066] Furthermore, a "cluster center metric" γ is defined for each sampling point. ij : Cluster centers are those with high density (close to many points). Here, α is a weighting factor, and α∈[0,1]. α=1 means only density is considered, α=0 means only distance is considered, and α=0.5 means both density and distance are considered. It can flexibly adapt to different types of distribution and noise.

[0067] Furthermore, the representative displacement y is determined by the cluster center metric. i * .

[0068] Specifically, take the one with the largest γ j The point is used as the representative value at that time point: y i * = y ij* ,

[0069] Among them, y ij* In order to be in All displacements y within a given time point ij Importance index γ in (1≤j≤n) j Maximum displacement. .

[0070] In practice, for each time point t i We obtained a cluster center y i * For representative values, see Table 1.

[0071] Table 1: Correspondence between the first swing amount and time

[0072]

[0073] S300 obtains the correspondence between the second swing amount and time based on the correspondence between the first swing amount and time;

[0074] Specifically, the second swing amount is the swing angle θ of the second robotic arm.

[0075] In practice, the method for obtaining the correspondence between the second swing amount and time includes: calculating the second swing amount based on the first swing amount;

[0076] The method for calculating the second oscillation is θ = arcsiny / L.

[0077] Where θ is the second swing amount, y is the swing displacement y of the second robotic arm, and L is the length of the second robotic arm.

[0078] Example 2:

[0079] See Figure 3 The swing displacement y, the swing angle θ of the second robotic arm 2 relative to the X-axis (i.e., the position of the second robotic arm 2 in its natural state), according to Figure 3 The relationship between them can be concluded as follows:

[0080] Sinθ=y / L, that is, θ=arcsiny / L;

[0081] From the formula θ=arcsiny / L and the table of the first oscillation quantity and time, we can obtain the table of the second oscillation quantity and time (i.e., the table of the second oscillation quantity and time).

[0082] It should be noted that since the swing of the robotic arm is essentially a rigid body rotational motion around a point, its dynamic equations, vibration models, and control are usually based on angles and angular velocities. The swing angle θ, relative to the swing displacement y, can more fundamentally describe the swing state of the robotic arm, providing the most direct input for generating high-precision swing suppression control commands. This avoids the errors caused by using an approximate linear relationship to replace a nonlinear rotational relationship, thus significantly improving the calculation accuracy of the swing suppression algorithm and the final control effect. When controlling the drive wheel 3, the command can directly correspond to the "angle difference" that needs to be compensated, making the swing suppression more precise.

[0083] S400 adjusts the rotation mode of the drive wheel according to the correspondence between the first swing amount and time and the second swing amount and time, so as to suppress the swing of the synchronous belt driven mechanical arm.

[0084] It should be noted that the method for adjusting the rotation of the drive wheel includes:

[0085] The input oscillation θ1 is obtained based on the tooth ratio of the synchronous gears of the driving wheel and the driven wheel;

[0086] The method for obtaining the input swing amount θ1 is specifically as follows:

[0087] Obtain the tooth ratio of the synchronous gears of the driving wheel and the driven wheel as 1:n;

[0088] The input swing amount θ1 = θ / n = (arcsiny / L) / n.

[0089] Based on the correspondence between the second swing amount and time, the correspondence between the input swing amount and time is obtained;

[0090] The correspondence between the input oscillation amount and time is input into a preset driver program to obtain the cam curve.

[0091] It should be noted that the cam curve generated in this embodiment is a swing trend predicted based on high-precision representative displacement data. This curve can instruct the drive wheel 3 to make a precise compensation motion curve that is opposite to the swing of the second robotic arm 2 in advance, so as to achieve advanced and precise suppression, rather than a delayed and passive response.

[0092] Specifically, traditional robotic arm swing methods only begin to calculate the response after vibration is detected, resulting in inherent delays. In this embodiment, the generated cam curve is a motion curve that represents the actual swing attenuation law of the robotic arm, extracted based on a clustering algorithm. By using high-precision representative displacement data, the motion law that the drive wheel 3 needs to execute to precisely counteract the swing is derived, thereby greatly improving the efficiency and effect of vibration suppression, significantly shortening the swing convergence time, and improving the working efficiency of the robotic arm.

[0093] Example 3:

[0094] In this embodiment, see Figure 3 When the drive wheel 3 rotates counterclockwise to stop the robotic arm, the second robotic arm 2 swings counterclockwise due to inertia. The tension and stretch of the left synchronous belt gradually decrease, while the tension and stretch of the right synchronous belt gradually increase. At this time, controlling the drive wheel 3 to rotate in the direction where the tension and stretch of the synchronous belt gradually increase allows the left and right synchronous belts to reach balance more quickly, thereby reducing the swing amplitude of the second robotic arm 2.

[0095] Specifically, a table showing the correspondence between a specific angle θ1 and its change with time t is imported into the servo driver program (i.e., the preset driver program) of drive wheel 3. By using the method of suppressing the rotation of the driven wheel with the driving wheel, the free swing time and swing amplitude of the second robotic arm 2 can be reduced. The specific method is as follows:

[0096] Import the time t data into the input axis and the second type of data into the output axis. The second type of data can be selected as stroke, speed, or acceleration vibration. In this way, the cam curve of the drive wheel 3 is obtained through the preset driver program.

[0097] Preferably, this embodiment uses the second data selection process.

[0098] Furthermore, the servo input axis in the preset driver is set to axis 1, and the output axis is set to axis 2. At the same time, the cam curve of axis 2 is set to cam number 1. After the settings are completed, the program only needs to drive axis 1, and axis 2 will move according to the cam curve.

[0099] Specifically, after adopting the synchronous belt drive robotic arm start-stop swing suppression method of this embodiment, the swing displacement of the robotic arm when it stops is shown in [reference]. Figure 4 As can be seen, the swing displacement of the second robotic arm 2 is significantly improved after adopting active suppression.

[0100] In practice, refer to Table 2 to monitor and statistically analyze the vibration values ​​of the robotic arm before and after using this method.

[0101] Table 2. Statistical table of vibration values ​​of the synchronous belt driven robotic arm during the stopping process.

[0102]

[0103] Where G=9.8m / s², as shown in the table above, the synchronous belt drive robotic arm start-stop swing suppression method of this embodiment can effectively reduce the vibration value of the robotic arm, making the transmission of the robotic arm more stable.

[0104] In this way, the cam curve generated based on real swing data ensures that the speed and acceleration changes of the drive wheel 3 are smooth, avoiding the impact and secondary vibration generated when using simple vibration suppression strategies such as emergency stop and emergency start. The motion process is smoother, reducing the impact on the mechanical structure, which helps to reduce wear, extend equipment life, improve positioning accuracy, and at the same time, the operating noise is also lower.

[0105] Furthermore, the cam curve in this embodiment is not fixed. The start-stop swing characteristics (frequency and decay rate) are different depending on the load of the robotic arm. As a result, the second swing amount obtained by the clustering algorithm has a different correspondence with time, and the generated cam curve is also different. This realizes customized sway suppression, completely avoids the tedious process of manually adjusting control parameters for different working conditions, and has a very high level of intelligence and strong generalization ability.

[0106] In practice, the system can form a closed loop of "curve generation - oscillation suppression - effect evaluation - feedback optimization". If there is still a small residual vibration after a certain vibration suppression, this information can be fed back to fine-tune the clustering weights or curve generation algorithm, so that the next generated cam curve is better. Its vibration suppression performance will continuously improve and enhance itself with the increase of usage time.

[0107] It should be noted that the start-stop sway suppression method of the synchronous belt driven robotic arm in this embodiment can be used not only for the start-stop sway suppression of synchronous belt driven robotic arms, but also for the start-stop sway suppression of chain driven robotic arms and similar structures.

[0108] See Figure 2 This embodiment also provides a start-stop sway suppression system for a synchronous belt-driven robotic arm, comprising: a data acquisition module for acquiring real-time vibration data of the synchronous belt-driven robotic arm; a first calculation module for acquiring a first sway amount versus time relationship based on the real-time vibration data; a second calculation module for acquiring a second sway amount versus time relationship based on the first sway amount versus time relationship; and a suppression module for adjusting the rotation mode of the drive wheel according to the first sway amount versus time relationship and the second sway amount versus time relationship, thereby suppressing the sway of the synchronous belt-driven robotic arm.

[0109] By adopting the above settings, the transmission weight of a robotic arm of the same size can be increased without affecting transmission stability or increasing transmission vibration; and a smaller synchronous belt can be selected for robotic arms transmitting the same weight, reducing the size of the robotic arm without affecting transmission stability or increasing transmission vibration.

[0110] This embodiment also provides a computer-readable storage medium storing computer-readable instructions. When executed by a processor, the computer-readable instructions implement various steps of the synchronous belt drive robotic arm start-stop swing suppression method or the synchronous belt drive robotic arm start-stop swing suppression system as described in any of the preceding embodiments.

[0111] This invention can take the form of a computer program product implemented on one or more storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing program code. Computer-readable storage media include permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer-readable storage media include, but are not limited to: novel memories such as phase-change memory / resistive random access memory / magnetic memory / ferroelectric memory (PRAM / RRAM / MRAM / FeRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, read-only optical disc read-only memory (CD-ROM), digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0112] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0113] Optionally, specific examples in this embodiment can refer to the examples described in the above embodiments, and will not be repeated here.

[0114] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0115] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0116] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for suppressing the swaying during start-stop operation of a synchronous belt-driven robotic arm, characterized in that, include: Obtain real-time vibration data of the synchronous belt driven robotic arm; The synchronous belt drive robotic arm includes a first robotic arm (1) and a second robotic arm (2); Based on the real-time vibration data, the correspondence between the first swing amount and time is obtained, wherein the first swing amount is the swing displacement y of the second robotic arm; Based on the correspondence between the first swing amount and time, the correspondence between the second swing amount and time is obtained, wherein the method for obtaining the correspondence between the first swing amount and time includes: The first swing amount y of the second robotic arm (2) at preset time points was collected multiple times. ij The first swing dataset Y is obtained. i ; Based on the first swing dataset Y i Obtain the first swing representative quantity y of the second robotic arm (2). i * Wherein, the first swing representative quantity y of the second robotic arm (2) is obtained. i * The methods include: Define the first oscillation dataset Y i Each of the first swing amounts y in ij The corresponding Gaussian kernel density ρ j ; Calculate the Gaussian kernel density ρ for each of the above. j Greater than the first swing amount y ij The point and the first swing amount y ij The minimum distance δ between j ; Based on the minimum distance δ j Obtain each of the first swing amounts y ij The corresponding cluster center metric γ ij Wherein, the cluster center metric γ ij The first swing amount y corresponding to the maximum point ij The first oscillation representative quantity y i * Obtain the cluster center metric γ ij The methods include: ； Where, γ ij ρ is a measure of cluster center, where α is a weighting factor, α∈[0,1], and ρ ij Let be the local density, representing point y. ij The density of surrounding points, δ ij The relative distance represents point y. ij The minimum distance to points that are denser than it; Based on the first swing representative quantity y i * The first swing amount y is obtained by using a preset time point corresponding to the first swing amount. ij The correspondence with time t; According to the correspondence between the first swing amount and time and the correspondence between the second swing amount and time, the rotation mode of the drive wheel (3) is adjusted to suppress the swing of the synchronous belt driven mechanical arm. The second swing amount is the swing angle θ of the second mechanical arm (2). The method for obtaining the correspondence between the second swing amount and time includes: calculating the second swing amount based on the first swing amount. The method for calculating the second oscillation is θ = arcsiny / L. Where θ is the second swing amount, y is the swing displacement y of the second robotic arm (2), and L is the length of the second robotic arm (2); The method for adjusting the rotation mode of the drive wheel (3) includes: The input oscillation amount θ1 is obtained based on the synchronous gear tooth ratio between the driving wheel (3) and the driven wheel, wherein the method for obtaining the input oscillation amount θ1 is as follows: Obtain the tooth ratio of the synchronous gear of the drive wheel (3) to the driven wheel as 1:n; The input swing amount θ1 = θ / n = (arcsiny / L) / n; Based on the correspondence between the second swing amount and time, the correspondence between the input swing amount and time is obtained; The correspondence between the input oscillation amount and time is input into a preset driver program to obtain the cam curve.

2. A start-stop sway suppression system for a synchronous belt-driven robotic arm, used to operate the start-stop sway suppression method for a synchronous belt-driven robotic arm as described in claim 1, characterized in that, The synchronous belt drive robotic arm start / stop swing suppression system includes: The acquisition module is used to acquire real-time vibration data of the synchronous belt driven robotic arm; The first calculation module is used to obtain the correspondence between the first oscillation amount and time based on the real-time vibration data; The second calculation module is used to obtain the correspondence between the second swing amount and time based on the correspondence between the first swing amount and time. The suppression module is used to adjust the rotation mode of the drive wheel (3) according to the correspondence between the first swing amount and time and the second swing amount and time, so as to suppress the swing of the synchronous belt driven mechanical arm.

3. A computer-readable storage medium storing computer-readable instructions, characterized in that, When the computer-readable instructions are executed by the processor, they implement the various steps of the synchronous belt drive robotic arm start-stop swing suppression method as described in claim 1.

Images