CN110110469A - Parallel robot dynamic parameters identification method based on singular value decomposition - Google Patents
Parallel robot dynamic parameters identification method based on singular value decomposition Download PDFInfo
- Publication number
- CN110110469A CN110110469A CN201910409939.6A CN201910409939A CN110110469A CN 110110469 A CN110110469 A CN 110110469A CN 201910409939 A CN201910409939 A CN 201910409939A CN 110110469 A CN110110469 A CN 110110469A
- Authority
- CN
- China
- Prior art keywords
- matrix
- singular value
- observing matrix
- parameter
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/06—Power analysis or power optimisation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Numerical Control (AREA)
Abstract
The invention discloses the work Identification of Dynamic Parameters of Amanipulator methods in parallel based on singular value decomposition, it is more for the kinetic parameter of parallel robot, cognizable observing matrix is difficult to through Analytic Method, and observing matrix is reduced to a cognizable matrix using singular value method.Data needed for parameter identification process are obtained by design excitation track, observing matrix are estimated using least square method, to estimate the kinetic parameter of robot.This method is versatile, convenience of calculation, can be very good the dynamic parameters identification for realizing parallel robot.
Description
Technical field
The present invention relates to robotic technology fields, more particularly to the parallel robot kinetic parameter based on singular value decomposition
Discrimination method.
Background technique
In recent years, industrial robot has been widely used for the fields such as spraying, welding, assembly and stacking.With robot
Develop to high speed, high-precision direction, to the control precision of industrial robot, more stringent requirements are proposed.Traditional industrial machine
People's controller has ignored the non-linear factor of robot dynamics, reduces the tracking accuracy of track.Based on kinetic model
Control technology can compensate these non-linear factors, to improve robot control precision.It is therefore desirable to accurately pick out
The kinetic parameter of robot.In general, Identification of Dynamic Parameters of Amanipulator process includes Manipulator Dynamics modeling, excitation track
Design and optimization, data acquisition and processing (DAP), parameter Estimation and model verifying
It is found after carrying out literature search to existing the relevant technologies, China Patent No.: CN109062051A, title: Yi Zhongti
The method of high Identification of Dynamic Parameters of Amanipulator precision, the patent are carried out mainly for the data processing method in identification process
It improves.China Patent No.: CN106125548A, title: industrial robot dynamic parameters identification method, the patent mainly describe
The dynamic parameters identification process of serial manipulator.Different with serial manipulator, the dynamic (dynamical) complexity of parallel robot is to ginseng
Number identification bring very big difficulty, currently, most research be for specific structure parallel robot carried out identification test
Card, in general, having to the part inertial parameter of parallel robot.
Therefore, for parallel robot dynamic parameters identification, the new method of one kind is proposed to solve the above problems.
Summary of the invention
The technical problems to be solved by the invention are to overcome lacking for existing parallel robot dynamic parameters identification method
It falls into.It is proposed to this end that the parallel robot dynamic parameters identification method based on singular value decomposition.
In order to achieve the above objectives, the present invention adopts the following technical scheme:
Parallel robot dynamic parameters identification method based on singular value decomposition, comprising:
S1, parallel manipulator human occupant dynamic model is established, the cognizable observing matrix of full rank is obtained by singular value decomposition method.
Track is motivated in S2, design joint, to data needed for setting up observing matrix.
S3, the corresponding joint angle in acquisition excitation track and joint moment related data, and data are filtered.
S4, kinetic parameter is estimated using least square method, obtains the estimated value of parameter.
S5, parallel manipulator human occupant dynamic model is verified.
Further, the kinetic model are as follows:
Wherein, q,It is joint angle, velocity and acceleration；τ is driving moment；ΘAllIt is complete observing matrix, βAllIt is
Whole kinetic parameters.
Further, the singular value value decomposition method can be ΘAllIt is write as:
ΘAll=U ∑AllVT
Wherein, U is unitary matrice, ∑AllIt is positive semi-definite diagonal matrix, and VTIt is unitary matrice, i.e. the conjugate transposition of V.∑ pair
Element on linea angulata is exactly matrix ΘAllSingular value can be ∑ by judging the size of singular valueAllIt is write as:
Wherein, ∑ is the matrix of linear independence.
It is assured that matrix Θ in this wayAllOrder obtained so as to find out one group of cognizable minimum inertial parameter:
τ=Θ β
Wherein, Θ is the observing matrix of full rank, and β is cognizable kinetic parameter.
Further, the excitation Trajectory Design uses seven order polynomial tracks.
Further, the filtering method is filtered using Butterworth, to improve the signal-to-noise ratio of signal.
Further, the method will treated in K time point tkThe data of acquisition substitute into observing matrix Θ, group
Conjunction obtains global observing matrix,
Corresponding measurement torque are as follows:Equation is solved by least square method, is obtained:
In formula,It is the estimated value of kinetic parameter.
Further, kinetic model is assessed using the residual mean square (RMS) root of torque prediction error.
Detailed description of the invention
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to needed in the embodiment
Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for ability
For the those of ordinary skill of domain, without creative efforts, it can also be obtained according to these attached drawings other attached
Figure.
Fig. 1 is the flow chart of the method for the present invention；
Fig. 2 is the connecting rod schematic diagram of parallel robot.
Fig. 3 is that track schematic diagram is motivated in joint；
Fig. 4 is the curve that robot joint moment is filtered；
Fig. 5 is the torque prediction curve in robot joint；
Specific embodiment
Below with reference to specific example, the method for the present invention is described in further details, the flow chart of the method for the present invention such as Fig. 1
It is shown, specific implementation the following steps are included:
(1) by taking Stewart parallel robot as an example, schematic diagram is as shown in Fig. 2, establish six degree of freedom Stewart parallel machine
Device human occupant dynamic model obtains the cognizable observing matrix of full rank by singular value decomposition.
The kinetic model of Stewart parallel robot is established using Lagrangian method and is linearized, and is obtained:
Wherein, q,It is joint angle, velocity and acceleration；τ is driving moment；ΘAll∈R6×142It is observing matrix, βAll
∈R142×1It is kinetic parameter, the kinetic parameter of Stewart parallel robot has 142, is difficult to obtain line by analytic method
The unrelated kinetic parameter of property.
Observing matrix ΘAllIt is not full rank, it can be Θ by singular value value decomposition methodAllIt is write as
ΘAll=U ∑AllVT
Wherein U is unitary matrice；∑AllIt is positive semi-definite diagonal matrix；And VTIt is unitary matrice, i.e. the conjugate transposition of V.∑ pair
Element on linea angulata is exactly matrix ΘAllSingular value can write ∑ as by judging the size of singular value:
Wherein, ∑ ∈ R6×88It is the matrix of linear independence.
It is assured that matrix Θ in this wayAllOrder obtained so as to find out one group of cognizable minimum inertial parameter:
τ=Θ β
Wherein, Θ ∈ R6×88It is the observing matrix of full rank, β ∈ R88×1It is cognizable 88 kinetic parameters.
(2) design excitation track, to data needed for setting up observing matrix.
Excitation Trajectory Design connects the starting point of setting using seven order polynomial tracksAnd terminalInterpolation among every axis
Track consists of two parts, and is write as:
In formula, coefficient aiAnd biFor meeting boundary condition, by adjusting d1And d2Track needed for generating, Fig. 3 is joint
Motivate track schematic diagram.
(3) the corresponding joint angles q in acquisition excitation track and joint moment τ data, and data are filtered.
The processing of Butterworth low pass wave is carried out to the data of acquisition, to improve the signal-to-noise ratio of signal, Butterworth filtering
The transmission function of device may be expressed as:
In formula, a, b are real constant, and m is the number of filter, and n is the order of filter, m≤n., Fig. 4 is robot
The curve that joint moment is filtered.
(4) according to treated, data form observing matrix, using weighted least-squares method to industrial robot dynamics
Parameter Estimation obtains the estimated value of parameter.
By treated in K time point tkThe data of acquisition substitute into observing matrix Θ, and combination obtains global observation square
Battle array,
Corresponding measurement torque are as follows:Equation is solved by least square method, is obtained:
In formula,It is the estimated value of kinetic parameter, Fig. 5 is the torque prediction curve in robot joint；
(5) Stewart parallel manipulator human occupant dynamic model is verified.
Kinetic model is assessed using the residual mean square (RMS) root RMS of torque prediction error.
The present invention obtains cognizable observing matrix by singular value decomposition, realizes parallel robot based on this
Dynamic parameters identification, this method is versatile, and convenience of calculation can be widely applied to robot field.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
In the technical scope disclosed by the present invention, any changes or substitutions that can be easily thought of by those familiar with the art, all answers
It is included within the scope of the present invention.Therefore, protection scope of the present invention should be subject to the protection scope in claims.
Claims (6)
1. the parallel robot dynamic parameters identification method based on singular value decomposition characterized by comprising
S1, parallel manipulator human occupant dynamic model is established, the cognizable observing matrix of full rank is obtained by singular value decomposition method.
Track is motivated in S2, design joint, to data needed for setting up observing matrix.
S3, the corresponding joint angle in acquisition excitation track and joint moment related data, and data are filtered.
S4, kinetic parameter is estimated using least square method, obtains the estimated value of parameter.
S5, parallel manipulator human occupant dynamic model is verified.
2. the method according to claim 1, wherein in the step S1, being established simultaneously using Lagrangian method
Connection Dynamic Models of Robot Manipulators is simultaneously linearized, and is obtained:
Wherein,It is joint angle, velocity and acceleration；τ is driving moment；ΘAllIt is complete observing matrix, βAllIt is all
Kinetic parameter.
In general, observing matrix ΘAllIt is non-full rank, it can be Θ by singular value value decompositionAllIt is write as:
ΘAll=U ΣAllVT
Wherein, U is unitary matrice, ∑AllIt is positive semi-definite diagonal matrix, and VTIt is unitary matrice, i.e. the conjugate transposition of V.∑ diagonal line
On element be exactly matrix ΘAllSingular value can be ∑ by judging the size of singular valueAllIt is write as:
Wherein, ∑ is the matrix of linear independence.
It is assured that matrix Θ in this wayAllOrder obtained so as to find out one group of cognizable minimum inertial parameter:
τ=Θ β
Wherein, Θ is the observing matrix of full rank, and β is cognizable kinetic parameter.
3. the method according to claim 1, wherein excitation Trajectory Design uses seven times in the step S2
Multinomial track.
4. according to the method described in claim 3, it is characterized in that, in the step S3, carrying out Bart to the data of acquisition
Butterworth filtering processing, to improve the signal-to-noise ratio of signal.
5. according to the method described in claim 4, it is characterized in that, treated K time point by general in the step S4
tkThe data of acquisition substitute into observing matrix Θ, and combination obtains global observing matrix,
Corresponding measurement torque are as follows:Equation is solved by least square method, is obtained:
In formula,It is the estimated value of kinetic parameter.
6. according to the method described in claim 5, it is characterized in that, using the residual of torque prediction error in the step S5
Poor root mean square assesses kinetic model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910409939.6A CN110110469A (en) | 2019-05-08 | 2019-05-08 | Parallel robot dynamic parameters identification method based on singular value decomposition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910409939.6A CN110110469A (en) | 2019-05-08 | 2019-05-08 | Parallel robot dynamic parameters identification method based on singular value decomposition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110110469A true CN110110469A (en) | 2019-08-09 |
Family
ID=67490716
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910409939.6A Pending CN110110469A (en) | 2019-05-08 | 2019-05-08 | Parallel robot dynamic parameters identification method based on singular value decomposition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110110469A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111496791A (en) * | 2020-04-27 | 2020-08-07 | 无锡信捷电气股份有限公司 | Overall dynamics parameter identification method based on series robot |
| CN113885493A (en) * | 2021-09-17 | 2022-01-04 | 华南理工大学 | Parallel robot trajectory tracking control method based on PD + robust controller |
-
2019
- 2019-05-08 CN CN201910409939.6A patent/CN110110469A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111496791A (en) * | 2020-04-27 | 2020-08-07 | 无锡信捷电气股份有限公司 | Overall dynamics parameter identification method based on series robot |
| CN113885493A (en) * | 2021-09-17 | 2022-01-04 | 华南理工大学 | Parallel robot trajectory tracking control method based on PD + robust controller |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220314445A1 (en) | Kinematics model-free trajectory tracking method for robotic arms and robotic arm system | |
| CN109159151A (en) | A kind of mechanical arm space tracking tracking dynamic compensation method and system | |
| CN105772917B (en) | A kind of three joint spot welding robot's Trajectory Tracking Control methods | |
| CN101574586B (en) | Shuttlecock robot and control method thereof | |
| CN105549383B (en) | Linear motor precision track tracing device and method | |
| CN101972170B (en) | Self-adapting filter for least square support vector machine and filtering method thereof | |
| CN110110469A (en) | Parallel robot dynamic parameters identification method based on singular value decomposition | |
| CN103878770A (en) | Space robot visual delay error compensation method based on speed estimation | |
| CN108267952B (en) | Self-adaptive finite time control method for underwater robot | |
| CN109828467B (en) | Data-driven unmanned ship reinforcement learning controller structure and design method | |
| CN107702709A (en) | A kind of noncooperative target moves the time-frequency domain mixing discrimination method with inertial parameter | |
| CN113051673A (en) | Robot improved Stribeck friction model identification method | |
| CN113910218B (en) | Robot calibration method and device based on kinematic and deep neural network fusion | |
| CN111624878A (en) | Integral sliding mode obtaining method and system for autonomous water surface robot trajectory tracking | |
| CN109514041A (en) | Car body tracking and welding robot of creeping | |
| CN107632559A (en) | A kind of copline cricket experimental system and its method of work based on image procossing | |
| CN108717262A (en) | A kind of abnormal curved surface tracking and system based on moment characteristics learning neural network | |
| CN109764876B (en) | Multi-mode fusion positioning method of unmanned platform | |
| CN108664043A (en) | Characteristics of Drivers ' Behavior point measuring device based on binocular vision and its control method | |
| CN107203271A (en) | Both hands recognition methods based on multi-sensor fusion technology | |
| CN111872933B (en) | SCARA robot trajectory tracking control method based on improved quadratic iterative learning control | |
| CN106444382A (en) | Series robot kinetic parameter identification method capable of ensuring physical feasibility | |
| CN1672881A (en) | On-line robot hand and eye calibrating method based on motion selection | |
| CN114589702B (en) | Collaborative robot dragging method based on dynamic parameter identification and admittance control | |
| CN114516050B (en) | Robot external force estimation method and system based on position jacobian condition number optimization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication |