Robot kinetic parameter online identification method and system

Abstract

The invention belongs to the technical field of robot control, and provides a robot kinetic parameter online identification method and system. The online identification method for the kinetic parameters of the robot comprises the steps that the robot is controlled to move according to an ideal track, and joint information in the moving process of the robot is obtained online; the joint information comprises an angle vector, a joint angular velocity vector, an angular acceleration vector and a control torque vector; and according to an initial minimum parameter set of to-be-identified kinetic parameters of a robot kinetic model and the joint information, continuously iteratively reducing the minimum parameter set of the to-be-identified kinetic parameters by using an optimal bounded ellipsoid algorithm until the maximum error of a control torque vector predicted value of the robot completely operating for one cycle is smaller than a preset error value, and obtaining a kinetic parameter identification result.

Description

Technical Field

[0001] The present invention belongs to the technical field of robot control, and in particular relates to a method and system for online identification of robot dynamics parameters. Background Art

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

[0003] The dynamic model of a robot system usually contains some unknown parameters, such as the inertia matrix, Coriolis matrix, gravity vector, etc. These parameters have an important impact on the system's motion performance and control strategy design. Traditional parameter identification methods mainly rely on the least squares method and gradient descent method, which often require a large number of data samples, are sensitive to noise, and easily lead to large parameter estimation deviations. In addition, the parameter convergence speed is slow, and the online identification effect of time-varying dynamic parameters is poor. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides a method and system for online identification of robot dynamic parameters, which can improve the online identification effect of time-varying dynamic parameters.

[0005] In order to achieve the above object, the present invention adopts the following technical solution: A first aspect of the present invention provides a method for online identification of robot dynamic parameters.

[0006] In one or more embodiments, a method for online identification of robot dynamics parameters is provided, comprising: Control the robot to move along an ideal trajectory, and obtain online the joint information of the robot during movement; the joint information includes an angle vector, a joint angular velocity vector, an angular acceleration vector, and a control torque vector; According to the initial minimum parameter set of the dynamic parameters to be identified of the robot dynamics model and the joint information, the optimal bounded ellipsoid algorithm is used to iteratively reduce the minimum parameter set of the dynamic parameters to be identified until the maximum error of the control torque vector prediction value of a complete operation cycle of the robot is less than the preset error value, and the dynamic parameter identification result is obtained.

[0007] As an implementation method, the robot dynamics model is expressed as:

[0008] In the formula represents the joint angle vector, represents the inertia matrix, represents the centripetal-Coriolis matrix, and denote the gravity, friction and control torque vectors respectively; They are The first and second derivatives of ; , , and is the known error range, and is the kinetic parameter to be identified.

[0009] As an implementation method, the relationship between the control torque vector and the initial minimum parameter set of the to-be-identified dynamic parameters of the robot dynamics model is: The control torque vector is represented by the product of the transpose of the initial minimum parameter set of the to-be-identified dynamic parameters of the robot dynamics model and the regression matrix.

[0010] As an implementation method, in the process of continuously iteratively reducing the minimum parameter set of the dynamic parameters to be identified by using the optimal bounded ellipsoid algorithm, a preset filter is used to filter out the noise in the angular acceleration vector, and the expression of the robot dynamic model is updated to:

[0011] in, represents the control torque vector after denoising; represents the transpose of the optimal estimation matrix of the dynamic parameters to be identified corresponding to the minimization of the approximate error of the dynamic parameters to be identified; represents the regression matrix after denoising; Represents the approximate error of the dynamic parameters to be identified after denoising.

[0012] A second aspect of the present invention provides an online identification system for robot dynamics parameters.

[0013] In one or more embodiments, a robot dynamics parameter online identification system includes: A joint information online acquisition module, which is used to control the robot to move according to an ideal trajectory and to obtain online the joint information of the robot during movement; the joint information includes an angle vector, a joint angular velocity vector, an angular acceleration vector and a control torque vector; The dynamic parameter identification module is used to continuously iteratively reduce the minimum parameter set of the dynamic parameters to be identified based on the initial minimum parameter set of the dynamic parameters to be identified of the robot dynamic model and the joint information, using the optimal bounded ellipsoid algorithm, until the maximum error of the control torque vector prediction value of a complete operation cycle of the robot is less than the preset error value, thereby obtaining the dynamic parameter identification result.

[0014] As an implementation mode, in the dynamic parameter identification module, the expression of the robot dynamic model is:

[0015] In the formula represents the joint angle vector, represents the inertia matrix, represents the centripetal-Coriolis matrix, and denote the gravity, friction and control torque vectors respectively; They are The first and second derivatives of ; , , and is the known error range, and is the kinetic parameter to be identified.

[0016] As an implementation mode, in the dynamic parameter identification module, the relationship between the control torque vector and the initial minimum parameter set of the dynamic parameters to be identified of the robot dynamic model is: The control torque vector is represented by the product of the transpose of the initial minimum parameter set of the to-be-identified dynamic parameters of the robot dynamics model and the regression matrix.

[0017] As an implementation mode, in the dynamic parameter identification module, in the process of continuously iteratively reducing the minimum parameter set of the dynamic parameters to be identified by using the optimal bounded ellipsoid algorithm, a preset filter is used to filter out the noise in the angular acceleration vector, and the expression of the robot dynamic model is updated as follows:

[0018] in, represents the control torque vector after denoising; represents the transpose of the optimal estimation matrix of the dynamic parameters to be identified corresponding to the minimization of the approximate error of the dynamic parameters to be identified; represents the regression matrix after denoising; Represents the approximate error of the dynamic parameters to be identified after denoising.

[0019] A third aspect of the present invention provides a computer-readable storage medium.

[0020] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps in the above-mentioned method for online identification of robot dynamics parameters.

[0021] A fourth aspect of the present invention provides an electronic device.

[0022] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps in the above-mentioned method for online identification of robot dynamic parameters are implemented.

[0023] Compared with the prior art, the present invention has the following beneficial effects: The present invention uses the optimal bounded ellipsoid algorithm to continuously iteratively reduce the minimum parameter set of the dynamic parameters to be identified based on the initial minimum parameter set of the dynamic parameters to be identified of the robot dynamics model and the joint information of the robot during motion obtained online, thereby achieving accelerated convergence of unknown parameters in online identification, ensuring stability and convergence in the parameter identification process, and improving the stability and accuracy of the dynamic parameter identification results. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] 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.

[0025] Figure 1 1 is a flow chart of a method for online identification of robot dynamics parameters according to an embodiment of the present invention; FIG2 (a) is an evolution trajectory of a parameter to be identified according to an embodiment of the present invention; FIG2( b ) is the eigenvalue of the learning gain matrix of an embodiment of the present invention; Figure 3 1 is a schematic diagram of the structure of an online identification system for robot dynamics parameters according to an embodiment of the present invention; Figure 4 is a schematic diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0026] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0027] It should be noted that the following detailed descriptions are all illustrative and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0028] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should 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.

[0029] Figure 1 FIG. 1 is a flow chart of a method for online identification of robot dynamic parameters in an embodiment of the present invention. Figure 1 The robot dynamics parameter online identification method in the embodiment shown may include: S101, controlling the robot to move along an ideal trajectory, and obtaining online joint information of the robot during movement; the joint information includes an angle vector, a joint angular velocity vector, an angular acceleration vector, and a control torque vector; S102, based on the initial minimum parameter set of the dynamic parameters to be identified of the robot dynamics model and the joint information, the optimal bounded ellipsoid algorithm is used to continuously iteratively reduce the minimum parameter set of the dynamic parameters to be identified until the maximum error of the control torque vector prediction value of a complete operation cycle of the robot is less than the preset error value, and the dynamic parameter identification result is obtained.

[0030] Among them, the expression of the robot dynamics model is:

[0031] In the formula represents the joint angle vector, represents the inertia matrix, represents the centripetal-Coriolis matrix, and denote the gravity, friction and control torque vectors respectively; They are The first and second derivatives of ; , , and is the known error range, and is the kinetic parameter to be identified.

[0032] Assumptions and is measurable, is the expected output vector, are bounded. Define a tracking error ) and its filtered tracking error in represents a positive diagonal matrix, is an auxiliary signal. , , col represents column vector concatenation; In this embodiment, M(p) is symmetric and positive definite, satisfying in are some constants.

[0033] In this embodiment, .

[0034] In some optional embodiments, the minimum parameter set and regression matrix of the robot's dynamic parameters can be obtained through existing methods, and the relationship between the control torque vector and the initial minimum parameter set of the dynamic parameters to be identified of the robot's dynamic model is as follows: the control torque vector is represented by the transpose of the initial minimum parameter set of the dynamic parameters to be identified of the robot's dynamic model and the multiplication of the regression matrix.

[0035] Formula (1) can be modified as follows:

[0036] in, Transpose of the initial minimal parameter set of the kinetic parameters to be identified; represents the regression matrix; express The first derivative of is the joint angular velocity.

[0037] Define the excitation trajectory, ideal trajectory, tracking error and parameters to be identified as bounded and define corresponding compact sets for them respectively , .

[0038]

[0039]

[0040] in, is the initial minimum parameter set of the kinetic parameters to be identified In Number of columns; , , Indicates the upper limit of the corresponding parameter.

[0041] Define the approximate error of the parameter to be identified :

[0042]

[0043]

[0044]

[0045]

[0046] in and The corresponding approximate errors are given respectively. The optimal estimate of the parameter to be identified that minimizes the absolute value, . , , , and are the regression matrices in the corresponding approximate errors. The meaning of is: the representation of the corresponding parameter, that is, equal to.

[0047] Using variable substitution , the robot dynamics model can be rewritten as follows

[0048] in, is the approximate error of the dynamic parameters to be identified; are all regression matrices.

[0049] To avoid using the noisy acceleration signal in equation (2) , using a stable filter for equation (2) ,in is the filter parameter, and c is a complex variable. Then, we can get

[0050] in, represents the control torque vector after denoising; represents the transpose of the optimal estimation matrix of the dynamic parameters to be identified corresponding to the minimization of the approximate error of the dynamic parameters to be identified; represents the regression matrix after denoising; Represents the approximate error of the dynamic parameters to be identified after denoising.

[0051] That It can be calculated as follows:

[0052] The left side of the equal sign in formula (4) is The first derivative of Approximate error of optimal parameters to be identified in is a constant.

[0053] According to the above theorem, the approximate error of the optimal parameter value to be identified is by Constraints, so the corresponding items after filtering There is also a limit, that is ,in yes The i-th element of is a constant, i = 1~n.

[0054] From formula (4), we can see that the control torque of the i-th joint can be written as follows:

[0055] in express The i-th column of , i = 1~n. Then, The ellipsoid set centered It can be defined as follows:

[0056] in i = 1~n. is chosen to be a value large enough so that Belongs to the ellipsoid set The approximate value of the control torque of the ith joint is:

[0057] in for The estimated value of . The filtered recognition error of the i-th joint is defined as

[0058] Identification algorithm with fast parameter convergence:

[0059]

[0060]

[0061]

[0062]

[0063]

[0064] in, represents the projection operator; Represents an intermediate physical quantity; is the learning gain matrix; is the regression matrix after denoising; is the identification error after filtering; is a constant coefficient; For the cause of forgetfulness; is the intermediate parameter; is the unit diagonal matrix; , is the intermediate parameter; It is the recognition error value after preset filtering.

[0065] is satisfied with Gain > 1, Satisfied with 0< A fixed forgetting factor < 1, are two freely chosen parameters that satisfy . , they represent The upper and lower limits of . i = 1~n.

[0066] (15) and Is (16) The learning gain matrix of this embodiment Bounded, that is , The estimated parameters to be identified are bounded by the preset ellipse set, namely, formula (6), that is, .

[0067] In this embodiment, the optimal bounded ellipsoid algorithm is used to iteratively reduce the minimum parameter set of the dynamic parameters to be identified. The process of online identification of the dynamic parameters is: Define the dynamic parameters to be identified as bounded, and then let the robot move along the ideal trajectory. During the robot movement, collect the joint angle, joint angular velocity, angular acceleration and control torque, and calculate them according to formula (4): According to formulas (5) to (8), the recognition error after filtering is obtained: , and then according to formulas (10) to (16) we get get , the integral is ,Then repeat the above process and iterate continuously until the maximum error of the control torque vector prediction value of the robot's complete operation cycle is less than the preset error value, and the dynamic parameter identification result is obtained.

[0068] As shown in Figure 2 (a), in order to study the performance under time-varying parameters, a load of 0.8 kg is added to the end of the robot manipulator. The load of 0.8 kg is added to the end of the robot manipulator through the gripper for t = 20 seconds. The evolution trajectory of the parameters to be identified is shown in Figure 2 (a), and the eigenvalues ​​of the learning gain matrix are shown in Figure 2 (b). It can be seen from the figure that the parameters have good convergence after using the improved optimal bounded ellipsoid algorithm.

[0069] This embodiment uses the optimal bounded ellipsoid algorithm to continuously iteratively reduce the minimum parameter set of the dynamic parameters to be identified based on the initial minimum parameter set of the dynamic parameters to be identified of the robot dynamics model and the joint information of the robot during motion obtained online, thereby achieving accelerated convergence of unknown parameters in online identification, and ensuring stability and convergence in the parameter identification process, thereby improving the stability and accuracy of the dynamic parameter identification results.

[0070] Figure 3 is a schematic diagram of the structure of a robot dynamics parameter online identification system in an embodiment of the present invention. Figure 1 The robot dynamics parameters online identification method corresponds to Figure 3 As shown, the robot dynamics parameter online identification system in this embodiment may include: A joint information online acquisition module 301 is used to control the robot to move according to an ideal trajectory and to acquire the joint information of the robot during its movement online; the joint information includes an angle vector, a joint angular velocity vector, an angular acceleration vector and a control torque vector; The dynamic parameter identification module 302 is used to continuously iteratively reduce the minimum parameter set of the dynamic parameters to be identified based on the initial minimum parameter set of the dynamic parameters to be identified of the robot dynamic model and the joint information using the optimal bounded ellipsoid algorithm until the maximum error of the control torque vector prediction value of a complete operation cycle of the robot is less than the preset error value, thereby obtaining the dynamic parameter identification result.

[0071] In the dynamic parameter identification module 302, the expression of the robot dynamic model is:

[0072] In the formula represents the joint angle vector, represents the inertia matrix, represents the centripetal-Coriolis matrix, and denote the gravity, friction and control torque vectors respectively; They are The first and second derivatives of ; , , and is the known error range, and is the kinetic parameter to be identified.

[0073] In the dynamic parameter identification module 302, the relationship between the control torque vector and the initial minimum parameter set of the dynamic parameters to be identified of the robot dynamic model is: The control torque vector is represented by the product of the transpose of the initial minimum parameter set of the to-be-identified dynamic parameters of the robot dynamics model and the regression matrix.

[0074] In the dynamic parameter identification module 302, in the process of continuously iteratively reducing the minimum parameter set of the dynamic parameters to be identified by using the optimal bounded ellipsoid algorithm, a preset filter is used to filter out the noise in the angular acceleration vector, and the expression of the robot dynamic model is updated to:

[0075] in, represents the control torque vector after denoising; represents the transpose of the optimal estimation matrix of the dynamic parameters to be identified corresponding to the minimization of the approximate error of the dynamic parameters to be identified; represents the regression matrix after denoising; Represents the approximate error of the dynamic parameters to be identified after denoising.

[0076] It should be noted here that Figure 3 The various modules in the robot dynamics parameter online identification system are Figure 1 The various steps in the online identification method of robot dynamic parameters in correspond to each other, and the specific implementation process is the same, which will not be repeated here.

[0077] Reference Figure 4 , a schematic diagram of an electronic device is given. It should be noted that, Figure 4 The electronic device 400 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0078] like Figure 4 As shown, the electronic device 400 includes a central processing unit (CPU) 401, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 402 or the program loaded from the storage part 408 to the random access memory (RAM) 403. In the RAM 403, various programs and data required for system operation are also stored. The central processing unit 401, the ROM 402 and the RAM 403 are connected to each other through a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0079] The following components are connected to the I / O interface 405: an input section 406 including a keyboard, a mouse, etc.; an output section 407 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 408 including a hard disk, etc.; and a communication section 409 including a network interface card such as a local area network (LAN) card, a modem, etc. The communication section 409 performs communication processing via a network such as the Internet. A drive 410 is also connected to the I / O interface 405 as needed. A removable medium 411, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 410 as needed, so that a computer program read therefrom is installed into the storage section 408 as needed.

[0080] When the central processing unit 401 in the electronic device of this embodiment executes the program, the following is achieved: Figure 1 The steps in the online identification method of robot dynamic parameters are shown.

[0081] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer readable medium, the computer program including a computer program for executing Figure 1 In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 409, and / or installed from the removable medium 411. When the computer program is executed by the central processing unit 401, various functions defined in the apparatus of the present application are executed.

[0082] in, Figure 1 The computer program instructions corresponding to the method shown may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0083] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0084] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for online identification of robot dynamic parameters, characterized in that: include: Control the robot to move along the ideal trajectory and obtain the joint information of the robot during movement online; The joint information includes an angle vector, a joint angular velocity vector, an angular acceleration vector and a control torque vector; According to the initial minimum parameter set of the dynamic parameters to be identified of the robot dynamics model and the joint information, the optimal bounded ellipsoid algorithm is used to iteratively reduce the minimum parameter set of the dynamic parameters to be identified until the maximum error of the control torque vector prediction value of a complete operation cycle of the robot is less than the preset error value, and the dynamic parameter identification result is obtained.

2. The method for online identification of robot dynamics parameters according to claim 1, characterized in that: The expression of the robot dynamics model is: In the formula represents the joint angle vector, represents the inertia matrix, represents the centripetal-Coriolis matrix, and denote the gravity, friction and control torque vectors respectively; They are The first and second derivatives of ; , , and is the known error range, and is the kinetic parameter to be identified.

3. The method for online identification of robot dynamic parameters according to claim 1, characterized in that: The relationship between the control torque vector and the initial minimum parameter set of the dynamic parameters to be identified in the robot dynamics model is: The control torque vector is represented by the product of the transpose of the initial minimum parameter set of the to-be-identified dynamic parameters of the robot dynamics model and the regression matrix.

4. The method for online identification of robot dynamics parameters according to claim 1, characterized in that: In the process of iteratively reducing the minimum parameter set of the dynamic parameters to be identified by using the optimal bounded ellipsoid algorithm, the noise in the angular acceleration vector is filtered out by a preset filter, and the expression of the robot dynamic model is updated as follows: in, represents the control torque vector after denoising; represents the transpose of the optimal estimation matrix of the dynamic parameters to be identified corresponding to the minimization of the approximate error of the dynamic parameters to be identified; represents the regression matrix after denoising; Represents the approximate error of the dynamic parameters to be identified after denoising.

5. A robot dynamics parameter online identification system, characterized in that: include: The joint information online acquisition module is used to control the robot to move according to the ideal trajectory and obtain the joint information of the robot during movement online; The joint information includes an angle vector, a joint angular velocity vector, an angular acceleration vector and a control torque vector; The dynamic parameter identification module is used to continuously iteratively reduce the minimum parameter set of the dynamic parameters to be identified based on the initial minimum parameter set of the dynamic parameters to be identified of the robot dynamic model and the joint information, using the optimal bounded ellipsoid algorithm, until the maximum error of the control torque vector prediction value of a complete operation cycle of the robot is less than the preset error value, thereby obtaining the dynamic parameter identification result.

6. The robot dynamics parameter online identification system according to claim 5, characterized in that: In the dynamic parameter identification module, the expression of the robot dynamic model is: In the formula represents the joint angle vector, represents the inertia matrix, represents the centripetal-Coriolis matrix, and denote the gravity, friction and control torque vectors respectively; They are The first and second derivatives of ; , , and is the known error range, and is the kinetic parameter to be identified.

7. The robot dynamics parameter online identification system according to claim 5, characterized in that: In the dynamic parameter identification module, the relationship between the control torque vector and the initial minimum parameter set of the dynamic parameters to be identified of the robot dynamic model is: The control torque vector is represented by the product of the transpose of the initial minimum parameter set of the to-be-identified dynamic parameters of the robot dynamics model and the regression matrix.

8. The robot dynamics parameter online identification system according to claim 5, characterized in that: In the dynamic parameter identification module, in the process of continuously iteratively reducing the minimum parameter set of the dynamic parameters to be identified by using the optimal bounded ellipsoid algorithm, the noise in the angular acceleration vector is filtered out by using a preset filter, and the expression of the robot dynamic model is updated to: in, represents the control torque vector after denoising; represents the transpose of the optimal estimation matrix of the dynamic parameters to be identified corresponding to the minimization of the approximate error of the dynamic parameters to be identified; represents the regression matrix after denoising; Represents the approximate error of the dynamic parameters to be identified after denoising.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps in the method for online identification of robot dynamics parameters as described in any one of claims 1 to 4 are implemented.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps in the method for online identification of robot dynamics parameters as described in any one of claims 1 to 4 are implemented.

Images