CN110154023A - A kind of multi-arm collaboration welding robot control method based on kinematics analysis - Google Patents
A kind of multi-arm collaboration welding robot control method based on kinematics analysis Download PDFInfo
- Publication number
- CN110154023A CN110154023A CN201910428896.6A CN201910428896A CN110154023A CN 110154023 A CN110154023 A CN 110154023A CN 201910428896 A CN201910428896 A CN 201910428896A CN 110154023 A CN110154023 A CN 110154023A
- Authority
- CN
- China
- Prior art keywords
- arm
- welding robot
- robot
- kinematics
- cooperates
- 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.)
- Granted
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J11/00—Manipulators not otherwise provided for
- B25J11/005—Manipulators for mechanical processing tasks
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1628—Programme controls characterised by the control loop
- B25J9/1653—Programme controls characterised by the control loop parameters identification, estimation, stiffness, accuracy, error analysis
Abstract
The present invention relates to a kind of, and the multi-arm based on kinematics analysis cooperates with welding robot control method, to realize the accurate welding of T-type structure connector bilateral weld seam between siding and stringer, comprising the following steps: 1) establish modularization multi-arm welding robot kinematics model；2) it establishes collaboration welding robot mechanical arm tail end and exports position auto―control, carry out robot Forward Kinematics Analysis, the correctness of verifier people kinematics parameters and forward kinematics solution, and obtain the data sample of each amount of articulation of robot；3) GRNN neural network is constructed and trained, and cooperates with welding robot linear joint and cradle head amount of exercise to predict multi-arm, is finally controlled according to the welding that predicted value completes multi-arm collaboration welding robot.Compared with prior art, the present invention has many advantages, such as that strong applicability, prediction is accurate, error is small, application is good.
Description
Technical field
The present invention relates to aviation welding robot motion control fields, more particularly, to a kind of based on the more of kinematics analysis
Arm cooperates with welding robot control method.
Background technique
In aviation, the muscle siding of large-sized structural parts generallys use T connector structure, and this structure is suitble to use
Dual-beam laser welding technique.This welding procedure is suitble to robot manipulating task, but structural member medium-and-large-sized for aircraft industry is complicated
The welding of Welded Joint Curve, single robot can not be competent at, and need to complete welding using the collaboration of a plurality of mechanical arm welding robot to make
Industry.It solves that there is multi-arm to cooperate with welding robot kinematics problem, can lay the foundation for its Analysis of The Working, facilitate this
One kind has the follow-up studies such as speed, acceleration, the dynamics of the robot of practical application value configuration in field of aerospace.
The kinematics solution method of robot mainly has analytic method and numerical method.Although analytic method is because acquiring robot whole
Inverse Kinematics Solution and the kinematics characteristic of robot can be completely described, but be directed to different topology structure robot,
Its forward steps is not quite similar, and lacks versatility.Numerical method calculating process is clear, succinct, to robot location's forward and reverse solution problem
Solution can be effectively performed, but generally require constraint condition, be unable to get all solutions of forward position analysis.
Welding robot inverse kinematics are cooperateed with for multi-arm, there has been no the elder generations that Efficient Solution is carried out to such robot
Example.Neural network can Nonlinear Function Approximation, generalized regression nerve networks (GRNN) be efficiently that a kind of highly-parallel is radial
Base net network, for error backward propagation method (BPNN), GRNN is based on radial base neural net, stability
Preferably, it is suitable for nonlinear prediction, compared with conventional radial base neural net, GRNN has stronger Approximation effect and higher instruction
Practice efficiency.
Summary of the invention
It is an object of the present invention to overcome the above-mentioned drawbacks of the prior art and provide one kind based on movement credit
The multi-arm of analysis cooperates with welding robot control method.
The purpose of the present invention can be achieved through the following technical solutions:
A kind of multi-arm collaboration welding robot control method based on kinematics analysis, to realize between siding and stringer
The accurate welding of T-type structure connector bilateral weld seam, comprising the following steps:
1) modularization multi-arm collaboration welding robot kinematics model is established；
2) it establishes collaboration welding robot mechanical arm tail end and exports position auto―control, carry out robot Forward Kinematics Analysis, test
The correctness of robot kinematics' parameter and forward kinematics solution is demonstrate,proved, and obtains the data sample of each amount of articulation of robot；
3) GRNN neural network is constructed and trained, and cooperates with welding robot linear joint and cradle head to move multi-arm
Amount is predicted, is finally controlled according to the welding that predicted value completes multi-arm collaboration welding robot.
In the step 1), multi-arm collaboration welding robot includes that Three Degree Of Freedom moves truss and is arranged in a truss left side
Two 6DOF revolute pair electroplating equipment wielding machine arms on right both sides, a 3DOF revolute pair being arranged among truss pressing are mechanical
Arm cooperates with welding robot topological structure for multi-arm, according to the geometrical condition of robot each branch assembly, by its modularization,
The DH parameter list for cooperateing with welding robot topological structure mutually unified with multi-arm is established, and determines robot scale parameter, constructs machine
Device people's kinematics model.
The multi-arm collaboration welding robot meets the following conditions:
(1) there is overhead viaduct type structure, multiple degrees of freedom, existing linear joint has cradle head again；
(2) there is two electroplating equipment wielding machine arms and a pressing mechanical arm；
(3) two electroplating equipment wielding machine arms are installed on mobile truss two sides at a certain angle；
(4) etc. dimensions input, etc. dimensions output.
In the step 2), collaboration i-th mechanical arm tail end of welding robot exports position auto―control TiAre as follows:
Wherein, piFor basis coordinates lower end actuator position matrix, [ni oi ai] it is basis coordinates lower end actuator posture
Matrix.
In the step 3), the input for cooperateing with welding robot end pose x as GRNN neural network using multi-arm becomes
Amount cooperates with welding robot linear joint and cradle head amount of exercise y as the output variable of GRNN neural network using multi-arm, right
GRNN neural network is trained, then is had:
Wherein, k is the number of data sample, and input variable dimension is 18, and output variable dimension is 18,For base
The h group data samples values of m-th of terminal position of robot under coordinate,For m-th of end of robot under basis coordinates
The h group data samples values of posture, n are robot linear joint number, and s is revolute joint number,Respectively
The amount of exercise of robot linear joint and cradle head.
In the GRNN neural network, the number of input layer and the dimension of input variable are corresponding, and input layer will
Input variable passes to mode layer, and the number of mode layer neuron is corresponding with number of samples k, transmission function are as follows:
Wherein, σ is smoothing factor, and x is input variable, xhFor the corresponding learning sample of h-th of neuron, h=1,2 ...,
k。
In the GRNN neural network, summation layer is summed using two types neuron, is specifically included:
Wherein, yhlFor h-th of output sample y of mode layerhIn first of element, i.e., with this element as in mode layer
Connection weight in h-th of neuron and summation layer between first of neuron.
In the GRNN neural network, the number of output layer neuron and the dimension of output variable are corresponding, each neuron
Output be being divided by for two kinds of summed results, then have:
In the step 3), in order to verify the forecasting accuracy of GRNN neural network, N group test data is inputted, is made more
Arm collaboration welding robot end is positioned in the space different location, and the absolute mistake of each amount of articulation is solved to every group of tracing point
Difference, using anticipation error ewEvaluation index judges prediction effect, then has:
Wherein, qapFor the amount of articulation of prediction, qasFor theoretical amount of articulation, a is the number in variety classes joint,
When calculating linear joint amount of exercise error, Nu=3, when calculating cradle head amount of exercise error, Nu=15.
Compared with prior art, the invention has the following advantages that
One, strong applicability: welding is cooperateed with for the multi-arm for welding Double curve degree siding T junction bilateral weld seam in aviation field
Robot establishes the kinematics model for the welding robot that there is a plurality of mechanical arm to cooperate, and can have overpass to meet
Formula structure, multiple degrees of freedom, existing linear joint has cradle head again, have two electroplating equipment wielding machine arms and a pressing mechanical arm and
Same type robot kinematics' model foundation offer that two electroplating equipment wielding machine arms are installed on mobile truss two sides at a certain angle is borrowed
Mirror is of great significance for improving robot trajectory planning and precision of real time control.
Two, prediction is accurate: analysis multi-arm cooperates with welding robot positive kinematics, provides its kinematical equation, and pass through number
It is worth sample calculation analysis, verifier people's kinematics model is correctly effective, to obtain robot effectively mobile and cradle head data
Sample provides practical foundation, ensure that the accuracy of prediction data sample.
Three, error is small, application is good: it is based on GRNN neural network, Robotic inverse kinematics are predicted and are calculated,
GRNN neural network fast convergence rate and being not easy in training falls into local minimum, and multi-arm cooperates with the inverse kinematics of welding robot
The error of method for solving is low, has good prediction effect, and application is good.
Detailed description of the invention
Fig. 1 is that multi-arm cooperates with welding robot Analytical Methods of Kinematics flow chart.
Fig. 2 is that multi-arm cooperates with welding robot threedimensional model.
Fig. 3 is electroplating equipment wielding machine arm setting angle schematic diagram.
Fig. 4 is that multi-arm cooperates with welding robot DH coordinate system.
Fig. 5 is robot architecture's parameter schematic diagram, wherein figure (5a) is electroplating equipment wielding machine arm, and figure (5b) is pressing mechanical arm.
Fig. 6 is that multi-arm cooperates with welding robot inverse kinematics linear joint predicted value and theoretical value absolute error figure.
Fig. 7 is that multi-arm cooperates with welding robot inverse kinematics cradle head predicted value and theoretical value absolute error figure.
Specific embodiment
The present invention is described in detail with specific embodiment below in conjunction with the accompanying drawings.
The basic idea of the invention is that cooperateing with electroplating equipment wielding machine arm based on multi-arm, modularization kinematics model is established, using mind
Through network can efficiently Nonlinear Function Approximation, for GRNN based on radial base neural net, stability is preferable, is suitable for non-
Linear prediction, compared with conventional radial base neural net, GRNN has stronger Approximation effect and higher training effectiveness.It is based on
The multi-arm of GRNN cooperates with electroplating equipment wielding machine arm Analysis of Inverse Kinematics method, helps to improve the precision, efficiency and stability of solution.
(1) multi-arm cooperates with welding robot kinematics model
Multi-arm cooperates with welding robot threedimensional model as shown in Fig. 2, its main machine frame is moved by three mechanical arms and Three Degree Of Freedom
Dynamic truss is constituted, and the mechanical arm of the right and left arrangement is 6DOF revolute pair electroplating equipment wielding machine arm, for Double curve degree covering T shape
The welding of joint bilateral weld seam；Intermediate mechanical arm is that 3DOF revolute pair presses mechanical arm, for during weld job
Press the stringer on aircraft Double curve degree covering；Three mechanical arms are hung upside down on the mobile truss moved along the Z direction.
According to the geometrical condition of each branch assembly of robot, its module is turned to three branches: branch 1 is moved by 3DOF
Dynamic truss and intermediate 3DOF pressing mechanical arm composition, left and right 6DOF electroplating equipment wielding machine arm is respectively branch 2,3, three machines
The structural parameters of tool arm indicate as shown in Figure 3.According to the relationship between robot architecture's parameter and joint, foundation is opened up with robot
The unified DH parameter list of structure and robot DH coordinate system diagram are flutterred, to express the kinematics in i-th branch of robot, j-th of joint
Parameter, as shown in table 1 and Fig. 4.Multi-arm collaboration two electroplating equipment wielding machine arms of welding robot are installed on zwThe mobile truss in direction is left
Right two sides, installation position angle are different from traditional DH table as shown in figure 3, robot in order to indicate mounting shift angle, need to be in table
Middle addition joint 1-0,2-0,3-0, are shown in Table 1.
1 welding robot DH parameter of table
(2) it establishes collaboration welding robot electroplating equipment wielding machine arm end and pressing mechanical arm tail end exports position auto―control
With joint rotation angle and translation for variable, robot homogeneous transform matrix are as follows:
The comprehensive transformation matrix in robotic weld arm end can be obtained as a result, are as follows:
According to the transformation matrix in each joint, the expression of robot arm end pose output matrix is obtained are as follows:
In formula, piFor basis coordinates lower end actuator position matrix, [ni oi ai] it is basis coordinates lower end actuator posture
Matrix.
Setting robot end's coordinate system be overlapped with basis coordinates system, then by ending coordinates system relative to basis coordinates system with
zw-yw-xwSequence rotation, i.e., first by ending coordinates system around zwAxis rotation alphaiAngle, further around ywAxis rotation βiAngle, finally around xwAxis rotation
Turn γiAngle, robot arm terminal angle can be by αi、βi、γiIt indicates, correspondingly, robot arm end is obtained by spin matrix
The posture at end are as follows:
(3) correctness of robot normal solution verifying kinematics model
Bring the structural parameters of robot and driving parameter into above formula, the kinematics for obtaining robotic manipulator is real
Number normal solution.It is emulated in ADAMS, obtains the end output of three mechanical arms of robot, comparing result, verifier people fortune
The dynamic correctness for learning model.
(4) robot movement and cradle head data sample are obtained
When carrying out inverse kinematics to multi-arm collaboration welding robot, training sample is inputted, the output and reality of network are made
The error of border output reaches minimum, reasonably selects sample by the study of strength neural network and generalization ability, to improve most
Whole solving precision.And training sample needs are selected according to the motion range in each joint of robot.According to robot
The motion range in joint, being evenly dividing each joint motions space is k group.For the value of first of joint h group kinematic variables
Are as follows:
In formula,For the minimum value of first of joint variable,For the maximum value of first of joint variable.
(5) it is based on GRNN neural network algorithm, completes to control the calculating of multi-arm collaboration welding robot inverse kinematics
The accurate numerical value of amount of exercise when each joint coordination of Zhi Qizhong robot moves.
The input variable of network is that multi-arm cooperates with welding robot end pose K is the number of data sample, and input variable dimension is 18,For under basis coordinates
The h group data samples values of m-th of terminal position of robot,For m-th of terminal angle of robot under basis coordinates
H group data samples values.Output variable is that multi-arm cooperates with welding robot linear joint and cradle head amount of exerciseOutput variable dimension is that 18, n is robot linear joint number
Mesh, s are revolute joint number,The respectively amount of exercise of robot linear joint and cradle head.Pass through data
Sample, training GRNN neural network.
The number of input layer corresponds to the dimension of input variable, and input variable is passed to mode layer by input layer.
The number of mode layer neuron corresponds to number of samples k, transmission function are as follows:
In formula, σ is smoothing factor, and x is network inputs variable, xhFor the corresponding learning sample of h-th of neuron, h=1,
2,…,k.Summation layer is summed using two types neuron, is respectively as follows:
In formula, yhlIt is h-th of output sample y of mode layerhIn first of element, i.e., with this element as in mode layer
Connection weight in h-th of neuron and summation layer between first of neuron.
The number of output layer neuron corresponds to the dimension of output variable, and the output of each neuron is two kinds of summed result phases
It removes, it may be assumed that
For remaining data sample as test data, the network being input to after training obtains machine after training data will be removed
The predicted value in each joint of device people, i.e., the input of each amount of articulation in robot controller, and by predicted value to evaluate mind
Through network training effect.
(6) error analysis
For the accuracy of accurate valuation prediction models, N group test data is inputted, makes multi-arm collaboration welding robot end
End is positioned in the space different location, and the absolute error of each amount of articulation is solved to every group of tracing point, is commented using anticipation error
The prediction effect of valence index scoring model, it may be assumed that
Wherein, qapFor the amount of articulation of prediction, qasFor theoretical amount of articulation, a is the number in variety classes joint,
When calculating linear joint amount of exercise error, Nu=3；When calculating cradle head amount of exercise error, Nu=15.
Embodiment:
The present invention is calculated using MATLAB Neural Network Toolbox according to summary of the invention.Its effect passes through specific multi-arm
Collaboration welding robot example is illustrated.Multi-arm cooperates with the structural parameters numerical value of welding robot as follows: L11=41mm, L12
=1000mm, L13=215mm, L21=L31=675mm, L22=L32=260mm, L23=L33=680mm, L24=L34=35mm,
L25=L35=670mm, L26=L36=158mm.It is any to choose one group of joint of robot driving parameter: θ11=0 °, θ12=0 °, θ13
=45 °, θ21=-30 °, θ22=90 °, θ23=-30 °, θ24=-15 °, θ25=-60 °, θ26=60 °, θ31=30 °, θ32=90 °,
θ33=-30 °, θ34=15 °, θ35=-60 °, θ36=-60 °, x=0mm, y=0mm, z=0mm.It substitutes into multi-arm and cooperates with welding robot
People's kinematics model, the kinematics real number normal solution that robotic manipulator is calculated are as shown in table 2.
The forward kinematics solution of 2 robot of table
It is emulated in ADAMS, the end output for obtaining three mechanical arms of robot is as shown in table 3.Contrast table 2, this
The robot end's pose solved under the multi-arm cooperating robot's kinematics model established is invented, is emulated under ADAMS environment
Obtained robot end's pose is very close, it was demonstrated that robot kinematics' model is correctly effective.
Robot pose in table 3ADAMS
According to the motion range of joint of robot in table 4, it is evenly dividing each joint motions space, by each joint motions
It is 10000 groups that range, which is evenly dividing,.The data sample of corresponding 10000 groups of robot poses is calculated, as shown in table 5.
4 range of motion of table
| Joint | Branch 1 | Branch 2 | Branch 3 |
| 1 | The 502mm of 0mm~5 | ±185° | ±185° |
| 2 | The 006mm of 0mm~3 | - 65 °~125 ° | - 65 °~125 ° |
| 3 | The 003mm of 0mm~1 | - 220 °~64 ° | - 220 °~64 ° |
| 4 | ±350° | ±350° | ±350° |
| 5 | ±130° | ±130° | ±130° |
| 6 | ±350° | ±350° | ±350° |
5 robot end's pose of table
Using 9960 groups of robot poses and robot corner as training sample, wherein xi、yi、zi、αi、βi、γi(i=1,
2,3) the end pose of branch 1,2,3 is indicated, the input variable of network is that multi-arm cooperates with welding robot end poseOutput variable is
Multi-arm cooperates with welding robot linear joint and cradle head amount of exercise Training GRNN neural network is completed to cooperate with welding robot to multi-arm
The calculating of inverse kinematics.
By remaining 40 groups more mechanical arm tail end pose in table 5, as test data, the network being input to after training is obtained
To the predicted value in each joint of robot, to evaluate neural metwork training effect.
The desired value of the desired value of 15 cradle head kinematic errors and 3 translation joint motions errors, such as Fig. 6 and Fig. 7
It is shown.Robot linear joint amount of exercise average error based on GRNN neural network prediction is 1.71 × 10-7Mm, rotation are closed
Saving amount of exercise average error is 1.34 × 10-7°, the calculating time: 1.276959 seconds；Above data is in processor are as follows: Intel
(R) Xeon (R) W-2135CPU 3.70GHz is inside saved as and is calculated gained on the computer of 16.0GB.
The result shows that the prediction essence of the inverse solution prediction model of multi-arm collaboration welding robot based on GRNN neural network
It spends very high, fast convergence rate and is not easy to fall into local minimum, error change is little, and stability is good, has good prediction effect.
A specific embodiment of the invention is described in conjunction with attached drawing above, but these explanations cannot be understood to limit
The scope of the present invention, protection scope of the present invention are limited by appended claims, any in the claims in the present invention base
Change on plinth is all protection scope of the present invention.
Claims (9)
1. a kind of multi-arm based on kinematics analysis cooperates with welding robot control method, to realize T between siding and stringer
The accurate welding of type structural joint bilateral weld seam, which comprises the following steps:
1) modularization multi-arm collaboration welding robot kinematics model is established；
2) it establishes collaboration welding robot mechanical arm tail end and exports position auto―control, carry out robot Forward Kinematics Analysis, verification machine
The correctness of device people kinematics parameters and forward kinematics solution, and obtain the data sample of each amount of articulation of robot；
3) construct and training GRNN neural network, and to multi-arm cooperate with welding robot linear joint and cradle head amount of exercise into
Row prediction is finally controlled according to the welding that predicted value completes multi-arm collaboration welding robot.
2. a kind of multi-arm based on kinematics analysis according to claim 1 cooperates with welding robot control method, special
Sign is, in the step 1), multi-arm collaboration welding robot includes that Three Degree Of Freedom moves truss and is arranged in a truss left side
Two 6DOF revolute pair electroplating equipment wielding machine arms on right both sides, a 3DOF revolute pair being arranged among truss pressing are mechanical
Arm cooperates with welding robot topological structure for multi-arm, according to the geometrical condition of robot each branch assembly, by its modularization,
The DH parameter list for cooperateing with welding robot topological structure mutually unified with multi-arm is established, and determines robot scale parameter, constructs machine
Device people's kinematics model.
3. a kind of multi-arm based on kinematics analysis according to claim 1 cooperates with welding robot control method, special
Sign is that the multi-arm welding robot meets the following conditions:
(1) there is overhead viaduct type structure, multiple degrees of freedom, existing linear joint has cradle head again；
(2) there is two electroplating equipment wielding machine arms and a pressing mechanical arm；
(3) two electroplating equipment wielding machine arms are installed on mobile truss two sides at a certain angle；
(4) etc. dimensions input, etc. dimensions output.
4. a kind of multi-arm based on kinematics analysis according to claim 1 cooperates with welding robot control method, special
Sign is, in the step 2), collaboration i-th mechanical arm tail end of welding robot exports position auto―control TiAre as follows:
Wherein, piFor basis coordinates lower end actuator position matrix, [ni oi ai] it is basis coordinates lower end actuator attitude matrix.
5. a kind of multi-arm based on kinematics analysis according to claim 4 cooperates with welding robot control method, special
Sign is, in the step 3), the input for cooperateing with welding robot end pose x as GRNN neural network using multi-arm becomes
Amount cooperates with welding robot linear joint and cradle head amount of exercise y as the output variable of GRNN neural network using multi-arm, right
GRNN neural network is trained, then is had:
Wherein, k is the number of data sample,For the h group data sample of m-th of terminal position of robot under basis coordinates
This value,For the h group data samples values of m-th of terminal angle of robot under basis coordinates, n, which is that robot is mobile, to be closed
Joint number mesh, s are revolute joint number,The respectively amount of exercise of robot linear joint and cradle head.
6. a kind of multi-arm based on kinematics analysis according to claim 1 cooperates with welding robot control method, special
Sign is, in the GRNN neural network, the number of input layer and the dimension of input variable are corresponding, and input layer will be defeated
Enter variable transferring to mode layer, the number of mode layer neuron is corresponding with number of samples k, transmission function are as follows:
Wherein, σ is smoothing factor, and x is input variable, xhFor the corresponding learning sample of h-th of neuron, h=1,2 ..., k.
7. a kind of multi-arm based on kinematics analysis according to claim 6 cooperates with welding robot control method, special
Sign is, in the GRNN neural network, summation layer is summed using two types neuron, specifically includes:
Wherein, yhlFor h-th of output sample y of mode layerhIn first of element, i.e., with this element as h-th in mode layer
Connection weight in neuron and summation layer between first of neuron.
8. a kind of multi-arm based on kinematics analysis according to claim 7 cooperates with welding robot control method, special
Sign is, in the GRNN neural network, the number of output layer neuron and the dimension of output variable are corresponding, each neuron
Output is being divided by for two kinds of summed results, then has:
9. a kind of multi-arm based on kinematics analysis according to claim 7 cooperates with welding robot control method, special
Sign is, in the step 3), in order to verify the forecasting accuracy of GRNN neural network, inputs N group test data, makes multi-arm
Collaboration welding robot end is positioned in the space different location, and the absolute mistake of each amount of articulation is solved to every group of tracing point
Difference, using anticipation error ewEvaluation index judges prediction effect, then has:
Wherein, qapFor the amount of articulation of prediction, qasFor theoretical amount of articulation, a is the number in variety classes joint, works as calculating
When linear joint amount of exercise error, Nu=3, when calculating cradle head amount of exercise error, Nu=15.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910428896.6A CN110154023B (en) | 2019-05-22 | 2019-05-22 | Multi-arm cooperative welding robot control method based on kinematic analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910428896.6A CN110154023B (en) | 2019-05-22 | 2019-05-22 | Multi-arm cooperative welding robot control method based on kinematic analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110154023A true CN110154023A (en) | 2019-08-23 |
| CN110154023B CN110154023B (en) | 2021-06-04 |
Family
ID=67631998
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910428896.6A Active CN110154023B (en) | 2019-05-22 | 2019-05-22 | Multi-arm cooperative welding robot control method based on kinematic analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110154023B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111027674A (en) * | 2019-11-21 | 2020-04-17 | 腾讯科技（深圳）有限公司 | Test method and device applied to multi-agent system |
| CN112496582A (en) * | 2020-11-23 | 2021-03-16 | 博迈科海洋工程股份有限公司 | Ocean engineering complex node multi-robot welding cooperative control method |
| CN113552807A (en) * | 2021-09-22 | 2021-10-26 | 中国科学院自动化研究所 | Data set generation method and device, electronic equipment and storage medium |
| CN114734440A (en) * | 2022-04-15 | 2022-07-12 | 同济大学 | UPF-RBF combined model-based kinematic parameter accurate calibration method for parallel-series double-arm transfer robot |
| CN114888793A (en) * | 2022-04-21 | 2022-08-12 | 同济大学 | Double-layer cooperative control method for multi-arm double-beam laser welding robot |
| CN115026418A (en) * | 2022-07-14 | 2022-09-09 | 同济大学 | Method for decomposing movement branched chain and expressing topological structure of truss multi-arm robot |
| CN116476080A (en) * | 2023-06-20 | 2023-07-25 | 西湖大学 | Aerial automatic grabbing operation planning method based on geometric feasibility |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102010023663A1 (en) * | 2010-06-12 | 2011-12-15 | Daimler Ag | Tester for online inspection of welding and/or soldering connection of components, has independent measuring modules for detecting process parameters of welding and/or soldering process, and produced welding/soldering connection |
| CN102581490A (en) * | 2012-03-21 | 2012-07-18 | 中国航空工业集团公司北京航空制造工程研究所 | Double-beam laser welding head in three-directional rotation |
| CN202506943U (en) * | 2011-12-21 | 2012-10-31 | 中国一冶集团有限公司 | Double-head riding welder |
| CN205218362U (en) * | 2015-12-24 | 2016-05-11 | 于波涛 | Two weldment work of robot stations of gate -type |
| CN108081266A (en) * | 2017-11-21 | 2018-05-29 | 山东科技大学 | A kind of method of the mechanical arm hand crawl object based on deep learning |
-
2019
- 2019-05-22 CN CN201910428896.6A patent/CN110154023B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102010023663A1 (en) * | 2010-06-12 | 2011-12-15 | Daimler Ag | Tester for online inspection of welding and/or soldering connection of components, has independent measuring modules for detecting process parameters of welding and/or soldering process, and produced welding/soldering connection |
| CN202506943U (en) * | 2011-12-21 | 2012-10-31 | 中国一冶集团有限公司 | Double-head riding welder |
| CN102581490A (en) * | 2012-03-21 | 2012-07-18 | 中国航空工业集团公司北京航空制造工程研究所 | Double-beam laser welding head in three-directional rotation |
| CN205218362U (en) * | 2015-12-24 | 2016-05-11 | 于波涛 | Two weldment work of robot stations of gate -type |
| CN108081266A (en) * | 2017-11-21 | 2018-05-29 | 山东科技大学 | A kind of method of the mechanical arm hand crawl object based on deep learning |
Non-Patent Citations (3)
| Title |
|---|
| 朴松昊等: "《工业机器人技术基础》", 30 June 2018 * |
| 梁斌焱: "三维T型接头激光焊接轨迹控制关键技术", 《中国优秀硕士论文学位论文全文数据库，工程科技Ⅰ辑》 * |
| 王鑫: "助老助残机械手控制技术研究", 《中国优秀硕士论文学位论文全文数据库信息科技辑》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111027674A (en) * | 2019-11-21 | 2020-04-17 | 腾讯科技（深圳）有限公司 | Test method and device applied to multi-agent system |
| CN111027674B (en) * | 2019-11-21 | 2023-02-07 | 腾讯科技（深圳）有限公司 | Test method and device applied to multi-agent system |
| CN112496582A (en) * | 2020-11-23 | 2021-03-16 | 博迈科海洋工程股份有限公司 | Ocean engineering complex node multi-robot welding cooperative control method |
| CN113552807A (en) * | 2021-09-22 | 2021-10-26 | 中国科学院自动化研究所 | Data set generation method and device, electronic equipment and storage medium |
| CN114734440A (en) * | 2022-04-15 | 2022-07-12 | 同济大学 | UPF-RBF combined model-based kinematic parameter accurate calibration method for parallel-series double-arm transfer robot |
| CN114734440B (en) * | 2022-04-15 | 2023-09-05 | 同济大学 | Precise calibration method for kinematic parameters of hybrid double-arm transfer robot |
| CN114888793A (en) * | 2022-04-21 | 2022-08-12 | 同济大学 | Double-layer cooperative control method for multi-arm double-beam laser welding robot |
| CN115026418A (en) * | 2022-07-14 | 2022-09-09 | 同济大学 | Method for decomposing movement branched chain and expressing topological structure of truss multi-arm robot |
| CN116476080A (en) * | 2023-06-20 | 2023-07-25 | 西湖大学 | Aerial automatic grabbing operation planning method based on geometric feasibility |
| CN116476080B (en) * | 2023-06-20 | 2023-08-29 | 西湖大学 | Aerial automatic grabbing operation planning method based on geometric feasibility |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110154023B (en) | 2021-06-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110154023A (en) | A kind of multi-arm collaboration welding robot control method based on kinematics analysis | |
| Liang et al. | Teaching robots to perform quasi-repetitive construction tasks through human demonstration | |
| CN106426164B (en) | A kind of multi objective coordinated movement of various economic factors planing method of redundancy double mechanical arms | |
| CN106945020A (en) | A kind of space double mechanical arms system motion control method for coordinating | |
| CN104933232B (en) | A kind of kinematics solution method of rod end float type six-degree-of-freedom parallel robot with angle compensation | |
| CN106914897A (en) | Inverse Solution For Manipulator Kinematics method based on RBF neural | |
| CN113878581B (en) | Error prediction and real-time compensation method for five-degree-of-freedom hybrid robot | |
| CN113807488B (en) | Space position optimization method for robot system component of labeling machine | |
| CN100498600C (en) | Large condenser underwater operation environment two-joint robot control method | |
| CN108717262B (en) | Special-shaped curved surface tracking method and system based on moment feature learning neural network | |
| CN108638058A (en) | A kind of posture decision dynamic programming method | |
| CN106041932A (en) | Movement control method for UR robot | |
| CN108153957A (en) | Space manipulator kinetics simulation analysis method, system and storage medium | |
| Liu et al. | Robot path planning with two-axis positioner for non-ideal sphere-pipe joint welding based on laser scanning | |
| Liang et al. | Teaching robots to perform construction tasks via learning from demonstration | |
| CN104678766A (en) | Method for solving optimal batting configuration for flight ball body operation of humanoid manipulator | |
| CN113799137A (en) | Mobile processing robot rapid visual positioning method based on neural network | |
| CN112157654A (en) | Optimization method for positioning error of robot machining system | |
| CN106875008A (en) | A kind of Inverse Solution For Manipulator Kinematics method and system | |
| Jiang et al. | Autonomous location control of a robot manipulator for live maintenance of high-voltage transmission lines | |
| Yuan et al. | A new algorithm for solving inverse kinematics of robot based on bp and rbf neural network | |
| Abderrahim et al. | Intelligent ANFIS controller of two cooperative 3-DOF manipulators: The case of manipulation under non-slip constraints | |
| CN114012733A (en) | Mechanical arm control method for scribing PC (personal computer) component mold | |
| CN112894819A (en) | Robot dynamic motion control method and device based on double neural networks | |
| Abaas et al. | Inverse kinematic analysis of lab-volt R5150 robot system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |