Robot stiffness identification method and system based on parallel rope drive loading multi-directional forces

Abstract

The invention discloses a robot stiffness identification method and system based on parallel rope-driven loading of multi-directional forces, belonging to the technical field of robot stiffness identification, comprising: selecting a plurality of different applied external loads as an initial load group, calculating the adaptability evaluation index of the initial load group and the condition number of the observation matrix corresponding to the initial load group, and then establishing an identification load evaluation index for system disturbance; wherein the magnitudes and directions of different external loads are different; taking minimization of the identification load evaluation index for system disturbance as the goal, combining the load constraint force limit of the robot as a constraint condition, constructing a variable load evaluation model for system disturbance, selecting the optimal load group according to the model solution result; based on the optimal load group, using a parallel rope-driven measurement and loading system to perform a stiffness identification test on the robot, measuring the coordinate values ​​of the target ball at the end of the robot under different loads, calculating the deformation vector, and then calculating the accurate robot joint stiffness.

Description

Technical Field

[0001] The invention belongs to the technical field of robot stiffness identification, and in particular relates to a robot stiffness identification method and system based on parallel rope drive loading multi-directional forces. Background Art

[0002] The statements herein merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] In high-end manufacturing, especially in precision machining fields such as aerospace, robots have gradually become important innovative equipment due to their high flexibility and spatial posture control capabilities. However, due to the inherent structural characteristics of their serial mechanisms, the stiffness of robots is usually much lower than that of traditional CNC machine tools. Especially when performing high-precision, high-load tasks, they are easily deformed by external forces. Therefore, stiffness becomes an important indicator to measure the robot's resistance to external force deformation, which directly affects the system's positioning accuracy, response speed and adaptability in complex environments. With the development of industrial robots towards higher precision and multi-tasking, the need for high-precision stiffness identification of robots has become more urgent.

[0004] Current robot stiffness identification methods are mostly based on static measurement, that is, by applying a static load in a specific direction and analyzing the displacement response of the end of the robot or the object being measured, its stiffness characteristics are calculated. However, this type of static measurement method can only reflect the stiffness information in a specific direction and it is difficult to cover complex, multi-directional dynamic force scenarios. In practical applications, robots are usually faced with multi-directional and multi-dimensional external interference, such as dynamic grasping, handling, and interactive operations. These application environments require more comprehensive stiffness characteristics. Static stiffness data in a single direction is often insufficient to accurately reflect the overall stiffness characteristics of the robot, further limiting the application effect of the robot in dynamic and complex scenarios.

[0005] In addition, traditional stiffness identification devices or systems also have certain limitations in the external load loading method. They can usually only apply loads in a single direction and it is difficult to ensure uniform distribution of force. Due to the limited structural design and directional arrangement of the loading device, it is difficult to simulate multi-directional and multi-angle force scenarios, resulting in the robot's response under complex loads being difficult to accurately capture. Moreover, this loading device is also insufficient in accuracy and control flexibility, which can easily cause uneven distribution of loading force and limited loading angles, which will further affect the accuracy of stiffness identification.

[0006] That is, the existing robot stiffness identification method has the limitation of a single force direction. Usually, the external force is applied to the end of the robot in only one direction, and the obtained stiffness model can only reflect the stiffness characteristics in that direction. Although this method is simple to operate, it has certain defects:

[0007] (1) The stiffness of a robot is usually anisotropic, that is, it exhibits different stiffness in different directions. For example, the stiffness of the end of the robot in the vertical direction may be much higher than that in the horizontal direction. The stiffness identification in a single direction cannot capture this difference, resulting in the model being unable to accurately reflect the actual situation and performing poorly in a multi-directional force environment.

[0008] (2) In multi-degree-of-freedom, high-dynamic motion operations, forces and deformations in different directions are often coupled with each other. If the analysis is based only on the stiffness identification results in a single direction, the error may increase with the increase in the complexity of the operation, especially in multi-joint operations and flexible joint systems, which can easily lead to unstable posture control and reduce the identification accuracy. Summary of the invention

[0009] In order to address the deficiencies of the above-mentioned prior art, the present invention provides a robot stiffness identification method and system based on parallel rope-driven loading of multi-directional forces. By constructing an anti-disturbance identification external load evaluation index, the optimal load group is selected to lay a data foundation for subsequent precise identification. A parallel rope-driven measurement and loading system is designed to identify the stiffness characteristics of the robot in multiple degrees of freedom, multiple angles and multiple directions, thereby achieving accurate multi-directional stiffness measurement and identification, and solving the problem of poor accuracy of traditional single-direction stiffness identification.

[0010] In a first aspect, the present invention provides a robot stiffness identification method based on parallel rope drive loading multi-directional forces.

[0011] A robot stiffness identification method based on parallel rope drive loading multi-directional force, comprising:

[0012] Select any multiple different applied external loads as the initial load group, calculate the adaptability evaluation index of the initial load group and the condition number of the observation matrix corresponding to the initial load group, and then establish the identification load evaluation index for system disturbance; wherein the size and direction of different external loads are different;

[0013] With the goal of minimizing the identification load evaluation index for system disturbance, the load constraint force limit of the joint robot is used as a constraint condition, a variable load evaluation model for system disturbance is constructed, and the optimal load group is selected according to the model solution results.

[0014] Based on the optimal load group, the parallel rope-driven measurement and loading system is used to carry out stiffness identification test on the robot. The coordinates of the target ball at the end of the robot under different loads are measured, the deformation vector is calculated, and then the joint stiffness of the robot is calculated.

[0015] In a second aspect, the present invention provides a robot stiffness identification system based on parallel rope drive loading multi-directional forces.

[0016] A robot stiffness identification system based on parallel rope drive loading multi-directional force, comprising:

[0017] The identification load evaluation index construction module is used to select multiple different applied external loads as the initial load group, calculate the adaptability evaluation index of the initial load group and the condition number of the observation matrix corresponding to the initial load group, and then establish the identification load evaluation index for the system disturbance; wherein the size and direction of different external loads are different;

[0018] The optimal load group selection module is used to minimize the identification load evaluation index facing the system disturbance, and the load constraint force limit of the joint robot is used as a constraint condition to build a variable load evaluation model facing the system disturbance, and select the optimal load group according to the model solution results;

[0019] The stiffness test and identification module is used to perform stiffness identification tests on the robot based on the optimal load group using a parallel rope-driven measurement and loading system, measure the coordinate values ​​of the target ball at the end of the robot under different loads, calculate the deformation vector, and then calculate the joint stiffness of the robot.

[0020] In a third aspect, the present invention further provides an electronic device comprising a memory and a processor, and computer instructions stored in the memory and executed on the processor, wherein when the computer instructions are executed by the processor, the steps of the method described in the first aspect are completed.

[0021] In a fourth aspect, the present invention further provides a computer-readable storage medium for storing computer instructions, which, when executed by a processor, complete the steps of the method described in the first aspect.

[0022] One or more of the above technical solutions have the following beneficial effects:

[0023] 1. The present invention proposes a robot stiffness identification method and system based on parallel rope-driven loading of multi-directional forces. By constructing an anti-disturbance identification external load evaluation index, the optimal load group is selected to lay a data foundation for subsequent accurate identification; the parallel rope-driven experimental measurement loading system is used to identify the stiffness characteristics of the robot under multiple degrees of freedom, multiple angles and multiple directions, to achieve accurate multi-directional stiffness measurement, and solve the problem of poor accuracy of traditional single-direction stiffness identification; the present invention constructs a stiffness model with more generalization ability through the above-mentioned method, significantly improves the positioning accuracy and operational adaptability of the robot in complex tasks, eliminates the uncertainty caused by the coupling effects in different directions, thereby avoiding the accumulation of errors in complex operations, and providing more accurate data support for the stiffness compensation algorithm of the robot, so that the reliability of the stiffness model and compensation algorithm is greatly improved, ensuring the adaptability of the robot operation.

[0024] 2. The present invention proposes a quantitative characterization method for identification accuracy based on joint stiffness identification error, establishes the correlation between joint stiffness identification error and observation matrix condition number, and proposes an anti-disturbance identification external load evaluation index in combination with adaptive identification load evaluation index, thereby achieving the integrity and accuracy of robot stiffness modeling.

[0025] 3. The present invention designs a parallel rope-driven measurement and loading system, in which a parallel rope-driven force loading platform is provided, that is, a force loading test platform which is combined with a driving device such as a motor or a cylinder and ropes that cooperate with each other, and a winding drum, an adjustment rope and a universal pulley that can control multiple directions are provided. The length of the rope is controlled by the driving device to change the size of the loading force, thereby realizing the application and adjustment of external force loads with multiple degrees of freedom, multiple angles and multiple directions. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0027] Figure 1 This is a flow chart of robot stiffness identification based on parallel rope drive loading multi-directional forces in an embodiment of the present invention;

[0028] Figure 2 Schematic diagram of a parallel rope-driven measurement and loading system in an embodiment of the present invention;

[0029] Figure 3 A schematic diagram of a force-bearing structure at the end of a robot in an embodiment of the present invention;

[0030] Figure 4 Schematic diagram of the force loading point at the end of the robot in an embodiment of the present invention;

[0031] Figure 5 The flowchart of the stiffness identification test of the robot using the parallel rope drive test loading system in the embodiment of the present invention.

[0032] Among them, 1. Laser tracker; 2. Driving device; 3. Three-pulley tension sensor; 4. Universal pulley; 5. Rigid column; 6. Target ball cross; 7. Force device; 8. Rigid base; 9. Robot body; 10. Robot connecting flange; 11. Table body; 12. Six-dimensional torque sensor; 13. Sensor connecting flange; 14. Rope; 15. Target ball. DETAILED DESCRIPTION

[0033] It should be noted that the following detailed descriptions are exemplary only, are intended to describe specific embodiments, are intended to provide further explanation of the present invention, and are not intended to limit exemplary embodiments according to the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those of ordinary skill in the art to which the present invention belongs. In addition, it should also be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0034] Embodiment 1

[0035] This embodiment provides a robot stiffness identification method based on parallel rope-driven loading of multi-directional forces. By constructing an anti-disturbance identification external load evaluation index, the optimal load group is selected to lay a data foundation for subsequent accurate identification. The parallel rope-driven experimental measurement loading system is used to identify the stiffness characteristics of the robot in multiple degrees of freedom, multiple angles and multiple directions, to achieve accurate multi-directional stiffness measurement, and to ensure the adaptability of the robot operation.

[0036] Specifically, considering that the robot system has nonlinear stiffness characteristics under different external loads, and the load position is too close to the robot base or too far away from the robot end effector, the identified stiffness matrix may not accurately reflect the overall nonlinear stiffness characteristics of the robot, especially when the joint stiffness is highly dependent on the end effector position. Therefore, in this embodiment, before applying external loads to the robot and identifying stiffness, firstly, according to the robot structural characteristics and the identification requirements of the stiffness matrix, the relevant influencing factors and their relationships that lead to joint stiffness identification errors are analyzed, and the appropriate external load direction and position are selected based on the effects of these influencing factors, so as to stimulate each degree of freedom of the robot and improve the accuracy of stiffness identification.

[0037] First, the vector X is identified based on the robot joint flexibility i Compared with the actual joint flexibility vector X, the corresponding joint flexibility deviation vector is calculated for:

[0038]

[0039] Considering that the relationship between the stiffness and flexibility of the robot joint is known, the stiffness and flexibility are usually reciprocal to each other. Therefore, there is also an influence relationship between the stiffness deviation and the flexibility deviation, and there is usually a nonlinear influence change. According to the joint flexibility deviation vector obtained above, the joint stiffness identification error can be calculated and expressed as:

[0040]

[0041] Secondly, the observation matrix in the process of robot stiffness identification is defined. The observation matrix represents the linear relationship between torque and dynamic parameters. The definition process is:

[0042] Assume that the load force on the end of the robot is M = [f x ,f y ,f z ,w x ,w y ,w z ] T , the robot end deformation is δ=[σ x ,σ y ,σ z ,σα x ,σα y ,σα z ] T , where [f x ,f y ,f z ] T is the force on the end of the robot, [w x ,w y ,w z ] T is the torque on the end of the robot, [σ x ,σ y ,σ z ] T and [σα x ,σα y ,σα z ] T are the translation deformation and rotation deformation of the robot end, respectively. According to Hooke's law, there is a linear relationship between the load force M on the robot end and the deformation δ of the robot end, which can be expressed as: M = Kδ, where K is the 6×6 robot end stiffness matrix;

[0043] The robot joint stiffness matrix is in, is the stiffness of each joint. According to the traditional stiffness model defined by the virtual joint method, there is a quantitative relationship between the joint stiffness matrix and the end stiffness matrix, which can be expressed as: K = J -T K θ J -1 , where J represents the Jacobian matrix of the robot; similarly, the joint angular displacement Δ is related to the torque m acting on the joint: Δ = k θ m; Combining the above formula, we can get: δ=JK θ -1 J T M;

[0044] According to the relationship between stiffness and flexibility, the joint flexibility matrix is ​​defined as:

[0045]

[0046] By separating the joint flexibility matrix R, the relationship between the robot end deformation δ and the joint flexibility matrix R can be obtained: δ = GR, where G is a 6×6 matrix. This matrix is ​​defined as the observation matrix in the robot stiffness identification process, and this observation matrix is ​​only related to the robot Jacobian matrix and the loading force on the robot end.

[0047] Based on the observation matrix defined above, two matrices r and t are assumed to be related to the observation matrix G, where the matrix r is defined as r = G T G, the matrix t is defined as t = G T δ, let and are the error terms of the two matrices, namely the measurement and calculation errors in the stiffness identification process. According to the classical measurement error theory, the error relationship between the observation matrix and the end deformation is established. This relationship can be obtained through the two matrices r and t related to the observation matrix and the error terms of the two matrices. and As well as the joint flexibility vector X and the joint flexibility deviation vector Establish, which can be expressed as:

[0048]

[0049] Finally, according to the matrix norm theory, combined with the above formula (2), the above formula (3) is sorted out to solve the norm of the joint flexibility vector error, and the joint stiffness identification error can be obtained, which is:

[0050]

[0051] The above formula (4) gives the upper bound of the joint stiffness identification error, and also establishes the quantitative relationship between the joint stiffness identification error and the matrix r condition number cond(r). According to this formula, if the matrix condition number cond(r) is reduced, the upper bound of the joint stiffness identification error can be reduced, thereby achieving the purpose of improving the joint stiffness identification accuracy.

[0052] On the basis of the above analysis, considering that adaptability reflects the robot's ability to adapt and express stiffness characteristics under different load conditions, the robot stiffness usually exhibits nonlinear characteristics, and changes in load will lead to different stiffness responses. A load selection strategy with high adaptability can help reveal the law of stiffness characteristics changing with load, so that the identification model can accurately describe the nonlinear behavior of stiffness. In addition, by selecting load conditions with good adaptability, it is possible to avoid the situation where the load is too large, causing the robot to exceed the load range, or too small, causing the stiffness to change insignificantly, thereby effectively reducing experimental errors and improving identification accuracy. Therefore, in this embodiment, adaptability is used as the main evaluation index for external load selection. The adaptability of the robot to the applied external load can be expressed by the inverse of the normalized Jacobian matrix condition number. The adaptability index is defined as R, which can be expressed as:

[0053]

[0054] Among them, the calculation formula of the condition number of the normalized Jacobian matrix is:

[0055]

[0056] Among them, J N is the normalized Jacobian matrix, k(J N ) has a range of: k(J N )∈[1,+∞), then the range of the adaptability index R is: R∈(0,1].

[0057] Furthermore, when R=1, all singular values ​​corresponding to the normalized Jacobian matrix are equal. At this time, the multi-directional force-loaded parallel rope-driven robot joint stiffness identification test bench is far away from the singular posture, and the robot's motion flexibility is better. Therefore, its applied external load adaptability is better, and its corresponding normalized Jacobian matrix condition number is smaller.

[0058] According to the above analysis results, in the robot stiffness identification method proposed in this embodiment, the optimal load group is first selected, such as Figure 1 As shown, the following steps are included:

[0059] Step S1, select any multiple different applied external loads as the initial load group, calculate the adaptability evaluation index of the initial load group and the condition number of the observation matrix corresponding to the initial load group, and then establish an identification load evaluation index for system disturbance; wherein the size and direction of different external loads are different.

[0060] Specifically, 6 different applied external loads are randomly selected as the initial load group, and the normalized Jacobian matrix J corresponding to each external load in the initial load group is calculated respectively. Ni(i=1,2,…,6), where the normalization of the Jacobian matrix of the robot can be obtained by introducing the characteristic length and converting all physical quantities into dimensionless form; further, based on the normalized Jacobian matrix, the condition number k of the normalized Jacobian matrix of each external load can be calculated. i (J N )(i=1,2,…,6), the inverse of the normalized Jacobian matrix condition number is used as the adaptability of the robot to the applied external load, the maximum value of all normalized Jacobian matrix condition numbers of the initial load group is taken, and the adaptability evaluation index of the initial load group is determined, that is, the adaptability evaluation index of the initial load group is marked as n1=max[k i (J N )].

[0061] Afterwards, according to the above definition of the observation matrix, the observation matrix G corresponding to the initial load group is calculated and solved, and then the condition number of the observation matrix is ​​calculated, which is recorded as n2=cond(G).

[0062] Finally, according to the adaptability evaluation index of the initial load group and the condition number of the observation matrix, the identification load evaluation index n=n1×n2 for system disturbance is established. The identification load group for system disturbance has a smaller n value. Using this load group for joint stiffness identification can obtain higher accuracy.

[0063] Step S2, with the goal of minimizing the identification load evaluation index for system disturbance, and taking the load constraint force limit of the joint robot as a constraint condition, a variable load evaluation model for system disturbance is constructed, and the optimal load group is selected according to the model solution results.

[0064] Specifically, based on the above analysis, a variable load evaluation index model for system disturbance is constructed as follows:

[0065]

[0066] The above formula (7) is a variable load evaluation index model for system disturbance, which takes the constraint force limit of the robot to bear the load as the constraint condition. In this constraint condition, the upper limit F max is the maximum load that the robot can bear, and the lower limit F min is 0.

[0067] Based on the above variable load evaluation index model, the optimal load group is obtained by solving it with the help of the fmincon function in MATLAB.

[0068] Step S3: Based on the optimal load group, a parallel rope-driven measurement and loading system is used to perform a stiffness identification test on the robot, measure the coordinate value of the target ball at the end of the robot under different loads, calculate the deformation vector, and then calculate the joint stiffness of the robot.

[0069] Specifically, according to the optimal load group obtained by solving, a stiffness identification experiment is carried out on the robot, and the robot is set in a set of initial postures so that each of its joints has a certain flexibility to cope with the external load. The coordinate values ​​of the target ball of the robot under different loads are measured by a laser tester, and then the deformation vector is calculated (the vector value can be obtained by comparing the coordinate value of the end measurement point under load with the coordinate value of the end measurement point under no load state), so that the accurate robot joint stiffness can be calculated through data analysis and processing.

[0070] Furthermore, based on the obtained deformation vector, the robot joint stiffness is calculated as follows: Taking the optimal load group as the loading force M applied to the end of the robot, and the obtained deformation vector as the deformation of the end of the robot δ, according to the above formula δ = K -1 M can be used to calculate the robot end stiffness matrix K, and through the above formula K = J -T K θ J -1 Calculate the robot joint stiffness matrix K θ Finally, the least square method can be used to fit and calculate the final robot joint stiffness. Preferably, anisotropy comparison analysis can also be performed based on the deformation vector obtained by calculation, and a stiffness model can be constructed.

[0071] like Figure 2 and Figure 3 As shown, this embodiment provides a parallel rope-driven measurement and loading system, which includes a robot body and a parallel rope-driven force-applying loading platform, etc. The parallel rope-driven force-applying loading platform includes a rope-driven structure. Through the reasonable configuration of the parallel rope-driven force-applying loading platform and the corresponding sensing equipment in the system, accurate force loading, real-time force feedback and displacement monitoring of the robot end can be achieved. Specifically, the system includes a laser tracker 1, a drive device 2, a three-pulley tension sensor 3, a universal pulley 4, a rigid column 5, a target ball cross 6, a force device 7, a rigid base 8, a robot body 9, a robot connection flange 10, a table 11, a six-dimensional torque sensor 12, a sensor connection flange 13, a rope 14, and a target ball 15.

[0072] like Figure 2As shown, the robot body 9 to be tested is fixed at the center of the rigid base 8, and the rigid base 8 is stably installed on the ground through anchor bolts to reduce the interference of environmental vibration; the end of the robot body 9 is connected to the six-dimensional torque sensor 12 through the robot connection flange 10, and the six-dimensional torque sensor 12 is connected to the force-bearing device 7 through the sensor connection flange 13. A target ball 15 is provided on the outer surface of the force-bearing device 7, and a target ball cross 6 is sleeved on the target ball. The laser tracker 1 is used to record the three-dimensional coordinate information of each measuring point on the target ball and the target ball cross. Among them, the six-dimensional torque sensor 12 communicates with the central controller to collect the size and direction of the applied external load in real time and feedback it to the central controller. The feedback data can be used for tension and real-time measurement; the force-bearing device 7 applies external loads of different sizes and directions through a rope-driven structure, and the rope-driven structure is distributed in a three-dimensional polyhedron layout to cover different force directions.

[0073] Furthermore, the parallel rope-driven force-applying loading platform includes a table 11, on which a plurality of rigid columns 5 are erected, and a driving device 2 is provided on each rigid column. The driving device 2 may be a motor or a cylinder, etc. The rope-driven structure includes a plurality of independently controllable ropes 14, one end of each of which is respectively connected to a plurality of different positions of the force-bearing device 7, and the other ends of each of which are respectively wound on corresponding winding drums, which are driven by the driving device 2. In this embodiment, the force-bearing device 7 applies loads through eight independently controllable ropes 14, and the end of each rope 14 is connected to a corresponding driving device 2, which can apply tension in different directions. The driving device communicates with the central processor, and is used to drive the winding drum to rotate forward and reverse to control the retraction and release of the rope, thereby completing the change in the length of the rope 14, thereby controlling the independent ropes to apply external loads of different sizes to the force-bearing device.

[0074] Preferably, the other end of the rope 14 passes through the three-pulley tension sensor 3 and the universal pulley 4 in sequence and then is wound around the winding drum. The three-pulley tension sensor serves as a feedback device for collecting tension and feeding it back to the central processor to ensure the accuracy of the loading force.

[0075] Furthermore, if Figure 5 As shown in the figure, the parallel rope drive measurement and loading system is used to carry out the stiffness identification test of the robot. The process is as follows:

[0076] Step 1: calibrate the six-dimensional torque sensor 12, the three-pulley tension sensor 3 and the laser tracker 1 using a calibration instrument;

[0077] Step 2: Initialize the robot body 9 to be tested, ensure that each joint has no displacement or rotation in the initial state, and move it to the initial measurement posture;

[0078] Step 3: Use the laser tracker 1 to record the three-dimensional coordinate information of each measuring point on the target ball 15 and the target ball cross 6, control the drive device 2 to adjust the length of each rope 14, and the three-pulley tension sensor 3 records the tension of the corresponding rope 14, and the six-dimensional torque sensor 12 records the end force of the robot body 9;

[0079] Step 4, gradually adjust the driving device 2 to control and adjust the length of the rope 14, and manually adjust the universal pulley 4 to adjust the angle of the rope and change the torque direction of the applied load; through this adjustment method, the rope can apply each external load in the optimal load group obtained by the above method, and repeat steps 3 and 4 until all external load force information and position coordinate information are obtained; wherein the information obtained specifically includes the size and direction of the force load on the end of the robot, and the coordinate values ​​of each measurement point at the end under different loads;

[0080] Step 5: Repeat steps 3 and 4 until all external load groups are measured and the measurement is completed.

[0081] Embodiment 2

[0082] This embodiment provides a robot stiffness identification system based on parallel rope drive loading multi-directional forces, including:

[0083] The identification load evaluation index construction module is used to select multiple different applied external loads as the initial load group, calculate the adaptability evaluation index of the initial load group and the condition number of the observation matrix corresponding to the initial load group, and then establish the identification load evaluation index for the system disturbance; wherein the size and direction of different external loads are different;

[0084] The optimal load group selection module is used to minimize the identification load evaluation index facing the system disturbance, and the load constraint force limit of the joint robot is used as a constraint condition to build a variable load evaluation model facing the system disturbance, and select the optimal load group according to the model solution results;

[0085] The stiffness test and identification module is used to perform stiffness identification tests on the robot based on the optimal load group using a parallel rope-driven measurement and loading system, measure the coordinate values ​​of the target ball at the end of the robot under different loads, calculate the deformation vector, and then calculate the joint stiffness of the robot.

[0086] Embodiment 3

[0087] This embodiment provides an electronic device, including a memory and a processor, and computer instructions stored in the memory and running on the processor. When the computer instructions are run by the processor, the steps in the robot stiffness identification method based on parallel rope drive loading multi-directional forces as described above are completed.

[0088] Embodiment 4

[0089] This embodiment also provides a computer-readable storage medium for storing computer instructions. When the computer instructions are executed by a processor, the steps in the robot stiffness identification method based on parallel rope drive loading multi-directional forces as described above are completed.

[0090] The steps involved in the above embodiments 2 to 4 correspond to the method embodiment 1, and the specific implementation methods can refer to the relevant description part of embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood to include any medium that can store, encode or carry an instruction set for execution by a processor and enable the processor to execute any method in the present invention.

[0091] Those skilled in the art should understand that the modules or steps of the present invention described above can be implemented by a general-purpose computer device, or alternatively, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.

[0092] The above description is only a preferred embodiment of the present invention. Although the specific implementation mode of the present invention is described in conjunction with the accompanying drawings, it is not a limitation of the protection scope of the present invention. Those skilled in the art should understand that on the basis of the technical solution of the present invention, various modifications or variations that can be made by those skilled in the art without creative work are still within the protection scope of the present invention.

Claims

1. A robot stiffness identification method based on parallel rope drive loading multi-directional force, characterized in that: include: Select any multiple different applied external loads as the initial load group, calculate the adaptability evaluation index of the initial load group and the condition number of the observation matrix corresponding to the initial load group, and then establish the identification load evaluation index for system disturbance; wherein the size and direction of different external loads are different; With the goal of minimizing the identification load evaluation index for system disturbance, the load constraint force limit of the joint robot is used as a constraint condition, a variable load evaluation model for system disturbance is constructed, and the optimal load group is selected according to the model solution results. Based on the optimal load group, the parallel rope-driven measurement and loading system is used to carry out stiffness identification test on the robot. The coordinates of the target ball at the end of the robot under different loads are measured, and the deformation vector is calculated, and then the joint stiffness of the robot is calculated. The parallel rope-driven measurement and loading system comprises a robot body and a parallel rope-driven force-applying loading platform, and the parallel rope-driven force-applying loading platform comprises a rope-driven structure; The robot body to be tested is fixed at the center of a rigid base, and the rigid base is stably installed on the ground through anchor bolts; the end of the robot body is connected to a six-dimensional torque sensor through a robot connection flange, and the six-dimensional torque sensor is connected to a force-bearing device through a sensor connection flange, and a target ball is provided on the outer surface of the force-bearing device, and a target ball cross is sleeved on the target ball; The six-dimensional torque sensor communicates with the central controller to collect the size and direction of the applied external load in real time and feed it back to the central controller; the force-bearing device applies external loads of different sizes and directions through a rope-driven structure; the rope-driven structure is distributed in a three-dimensional polyhedron layout to cover different force directions.

2. A robot stiffness identification method based on parallel rope drive loading multi-directional force as claimed in claim 1, characterized in that: The process of constructing the identification load evaluation index for system disturbance includes: Select a plurality of different applied external loads as the initial load group, calculate the normalized Jacobian matrix corresponding to each external load in the initial load group, and then solve to obtain the normalized Jacobian matrix condition number; The inverse of the normalized Jacobian matrix condition number is used as the adaptability of the robot to the applied external load. The maximum value of all normalized Jacobian matrix condition numbers of the initial load group is taken to determine the adaptability evaluation index of the initial load group. According to the definition of the observation matrix, the observation matrix corresponding to the initial load group is solved, and then the condition number of the observation matrix is ​​calculated; Based on the adaptability evaluation index of the initial load group and the observation matrix condition number, an identification load evaluation index for system disturbance is established.

3. A robot stiffness identification method based on parallel rope drive loading multi-directional force as claimed in claim 1, characterized in that: The parallel rope-driven force-applying loading platform comprises a table body, on which a plurality of rigid columns are erected, and each rigid column is provided with a driving device; the rope-driven structure comprises a plurality of independently controllable ropes, one end of each of which is respectively connected to a plurality of different positions of the force-bearing device, and the other ends of each of which are respectively wound on corresponding winding drums, and the winding drums are driven by the driving device; The driving device communicates with the central processing unit and is used to drive the winding drum to rotate forward and reversely to control the retraction and release of the rope and the change of its length, thereby controlling the independent rope to apply external loads of different sizes to the force-bearing device.

4. A robot stiffness identification method based on parallel rope drive loading multi-directional force as claimed in claim 3, characterized in that: The other end of the rope passes through the three-pulley tension sensor and the universal pulley in sequence, and then is wound around the winding drum; the three-pulley tension sensor serves as a feedback device for collecting tension and feeding it back to the central processing unit.

5. A robot stiffness identification method based on parallel rope drive loading multi-directional force as claimed in claim 2, characterized in that: It also includes a laser tracker for recording the three-dimensional coordinate information of each measuring point on the target ball and the target ball cross.

6. A robot stiffness identification method based on parallel rope drive loading multi-directional force as claimed in claim 1, characterized in that: The robot stiffness identification test is carried out using a parallel rope drive measurement and loading system, including: Step 1: Calibrate the six-dimensional torque sensor, three-pulley tension sensor and laser tracker using a calibration instrument; Step 2: Initialize the robot to be tested, ensure that each joint has no displacement or rotation in the initial state, and move it to the initial measurement posture; Step 3: Use a laser tracker to record the three-dimensional coordinate information of each measuring point on the target ball and the target ball cross, control the drive device to adjust the length of each rope, and the three-pulley tension sensor records the tension of the corresponding rope, and the six-dimensional torque sensor records the end force of the robot; Step 4: gradually adjust the driving device to control and adjust the length of each rope, and adjust the universal pulley to adjust the angle of the rope, change the torque direction of the applied load, so that the rope applies each external load in the optimal load group, and obtain the external load force information and position coordinate information; Step 5: Repeat steps 3 and 4 until all external load groups are measured and the measurement is completed.

7. A robot stiffness identification system based on parallel rope drive loading multi-directional force, characterized in that: include: The identification load evaluation index construction module is used to select multiple different applied external loads as the initial load group, calculate the adaptability evaluation index of the initial load group and the condition number of the observation matrix corresponding to the initial load group, and then establish the identification load evaluation index for the system disturbance; wherein the size and direction of different external loads are different; The optimal load group selection module is used to minimize the identification load evaluation index facing the system disturbance, and the load constraint force limit of the joint robot is used as a constraint condition to build a variable load evaluation model facing the system disturbance, and select the optimal load group according to the model solution results; The stiffness test and identification module is used to perform stiffness identification tests on the robot based on the optimal load group using a parallel rope-driven measurement and loading system, measure the coordinate values ​​of the target ball at the end of the robot under different loads, calculate the deformation vector, and then calculate the joint stiffness of the robot; The parallel rope-driven measurement and loading system comprises a robot body and a parallel rope-driven force-applying loading platform, and the parallel rope-driven force-applying loading platform comprises a rope-driven structure; The robot body to be tested is fixed at the center of a rigid base, and the rigid base is stably installed on the ground through anchor bolts; the end of the robot body is connected to a six-dimensional torque sensor through a robot connection flange, and the six-dimensional torque sensor is connected to a force-bearing device through a sensor connection flange, and a target ball is provided on the outer surface of the force-bearing device, and a target ball cross is sleeved on the target ball; The six-dimensional torque sensor communicates with the central controller to collect the size and direction of the applied external load in real time and feed it back to the central controller; the force-bearing device applies external loads of different sizes and directions through a rope-driven structure; the rope-driven structure is distributed in a three-dimensional polyhedron layout to cover different force directions.

8. An electronic device, characterized in that: The invention comprises a memory and a processor and computer instructions stored in the memory and executed on the processor. When the computer instructions are executed by the processor, the steps of a robot stiffness identification method based on parallel rope drive loading multi-directional forces as claimed in any one of claims 1 to 6 are completed.

9. A computer-readable storage medium, characterized in that: Used to store computer instructions, which, when executed by a processor, complete the steps of a robot stiffness identification method based on parallel rope drive loading multi-directional forces as described in any one of claims 1-6.

Images