CN106425181A - Curve welding seam welding technology based on line structured light - Google Patents
Curve welding seam welding technology based on line structured light Download PDFInfo
- Publication number
- CN106425181A CN106425181A CN201610922507.1A CN201610922507A CN106425181A CN 106425181 A CN106425181 A CN 106425181A CN 201610922507 A CN201610922507 A CN 201610922507A CN 106425181 A CN106425181 A CN 106425181A
- Authority
- CN
- China
- Prior art keywords
- point
- curve
- welding
- image
- robot
- 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
- 238000003466 welding Methods 0.000 title claims abstract description 92
- 238000005516 engineering process Methods 0.000 title claims abstract description 8
- 238000000034 method Methods 0.000 claims abstract description 32
- 238000000605 extraction Methods 0.000 claims abstract description 7
- 239000011159 matrix material Substances 0.000 claims description 24
- 239000013598 vector Substances 0.000 claims description 17
- 238000004422 calculation algorithm Methods 0.000 claims description 16
- 239000011324 bead Substances 0.000 claims description 8
- 238000001914 filtration Methods 0.000 claims description 7
- 238000009795 derivation Methods 0.000 claims description 3
- 230000003628 erosive effect Effects 0.000 claims description 3
- 229910000679 solder Inorganic materials 0.000 claims description 3
- 230000007704 transition Effects 0.000 claims description 3
- 210000000707 wrist Anatomy 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims 1
- 238000002203 pretreatment Methods 0.000 abstract 1
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 230000000877 morphologic effect Effects 0.000 description 2
- 238000011946 reduction process Methods 0.000 description 2
- 239000006002 Pepper Substances 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K37/00—Auxiliary devices or processes, not specially adapted to a procedure covered by only one of the preceding main groups
Landscapes
- Physics & Mathematics (AREA)
- Optics & Photonics (AREA)
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Numerical Control (AREA)
- Manipulator (AREA)
Abstract
The invention discloses a curve welding seam welding technology based on line structured light and oriented to an oblique crossed tube. According to the method, machine version and robot kinematics are combined very well, so that the problems that the process of obtaining curve welding seam feature points is not flexible, and a curve track model are difficult to establish are solved. The curve welding seam welding technology comprises the following steps of: firstly, controlling a six-degree-of-freedom robot, enabling a robot end to drive an industrial camera to shoot a curve welding seam image with the line structured light; and then, carrying out a series of image pre-treatment and feature point extraction work to obtain space coordinates of all feature points of a curve welding seam, and fitting through a B sample curve, and finally obtaining a smooth space curve. Besides, the gesture of a welding gun can be changed in real time according to included angles at the two sides of a workpiece when the welding gun is used for welding a curve, so that the welding process and welding requirements are improved better.
Description
Art
The invention belongs to industrial automation welding technology field, it is related to the curved welding seam detection technique of line-structured light guiding,
Particularly can in real time the angle according to side surface of workpiece changing the method for planning track of posture of welding torch.
Background technology
With the continuous development of industrial automation, automatic welding has become as the development of the manufacturing enterprises such as automobile, ship
Main flow.Thus attract widespread attention and pay close attention to.It is currently based on realizing with machine vision of line-structured light for guiding
The major part of automatic welding with straight bead detect based on, and with regard to curved welding seam welding for, the determination one of weld bead feature points
As be to be determined by the method for teaching, this kind of method acquisition process machinery is it is impossible to meet the demand of automated production.
In actual welding, it is frequently encountered continuous line welding situation during oblique pipe grafting.Weldering is passed through for Guan Guanxiang
For connecing, its model is typical space curve.For curved welding seam, calculated by simple pose interpolation trajectory planning
Method can not meet welding quality and the requirement of strict demand, need in real time the angle according to side surface of workpiece changing welding gun
Position and attitude.
How robot precisely quickly finds curved welding seam the folder according to side surface of workpiece during welding
Angle, to control the attitude of welding gun, is a great problem perplexing field of automatic welding always.By Six-DOF industrial robot
Robot end's industrial camera is driven to carry out IMAQ to curved welding seam, through certain Image semantic classification and characteristics of weld seam
Point detection work, using three-dimensional reconstruction, obtains the robot coordinate of each curved welding seam characteristic point, finally utilizes B-spline curves to insert
Mend and posture of welding torch discrete logarithm, arc welding robot not only can be made to carry out welding according to optimal path, can also be greatly
Improve welding quality, improve production efficiency, significant in actual production
Content of the invention
The present invention can effectively realize industrial robot to the automatic identification of curved welding seam and welding, and being capable of basis
The angle in oblique tube side face adjusts the attitude of welding gun in real time, through real example show this algorithm not only accurate positioning and also can expire
The requirement of the various welding procedure of foot.
The technical scheme that the present invention solves above-mentioned technical problem is to propose a kind of curved welding seam based on line-structured light automatic
Change solder technology.Its detailed process is as follows：
Step one：Adjusting Six-DOF industrial robot makes the industrial camera on its driving mechanical end and line-structured light
Sensor, makes structure light light belt irradiation position s1Place, is shot to the light belt on curved welding seam using industrial camera, a pen-hold grip
Take the photograph snPlace.Shoot 6 width images in real example, and the Digital Image Transmission obtaining will be shot to image processor.
Step 2：Image processing work is carried out through feature point extraction algorithm to captured curved welding seam image, obtains
The each characteristic point of weld seam and the image coordinate of reference point, its concrete steps includes：
Step 2-1, noise reduction process, using medium filtering and LOG filtering, are carried out necessarily to the weld image collecting
Noise reduction process, make the light striped of structure light distincter；
Step 2-2, the intercepting of ROI region, by the gray value scanning to image Y direction, find a certain row vertically side
To the pixel position that gray value is maximum, according to practical work piece size, being put with this is to translate n with reference to the Y-axis negative direction of this point
The position of individual pixel, and this is set to the Y-axis coordinate of clip image starting point, X-axis coordinate in the position numerical value of Y direction
If set.
Step 2-3, image binaryzation are processed, and select varimax herein.This algorithm be not required to setting that very important person is other
Parameter, is a kind of method automatically selecting threshold value；
Step 2-4, morphological erosion, used herein be 3*3 template processed so that the burr at striation edge and
Some isolate scattered point and are eliminated；
Step 2-5, image thinning, by horizontally and vertically scanning direction, find out the point that all pixels value is 1, find out light
Band, in the ordinate of each row up-and-down boundary, is designated as y respectively1iAnd y2i.The width of light belt is designated as Δ yi, the ordinate discipline of center line
For ymidi, so center point coordinate can be expressed as (xi, ymidi), wherein
Step 2-6, feature point extraction, constantly calculate the slope of the line segment with N number of pixel for unit length from left to right,
The slope of whole line segment is represented with the type of curve.It is clear that the slope variation of line segment from slope variation curve,
Learn that by analysis the point corresponding to gradient maxima is the position of the required weld bead feature points asked on curve.
Step 3：Three-dimensional reconstruction, is transformed into robot seat by feature point coordinates and with reference to point coordinates by image coordinate system
Mark system, by demarcating the transition matrix T understanding video camera to wrist.The six degree of freedom joint angle of robot can be from teaching machine
Learn, can learn, through the normal solution of robot, the position auto-control B that robot end is with respect to robot base.Finally, pass through
Formula pb=B*T*pc, p can be tried to achievecCoordinate pb under basis coordinates system for the point.
Step 4：The curve interpolation of each characteristic point, this problem to realize inserting of space weld curve using the theory of B-spline
Value.In welding processing, we, first according to the image processing algorithm in step 2 and step 3 and three-dimensional reconstruction, get
The three-dimensional coordinate of each characteristic point of curved welding seam, that is, offset point coordinates, then calculate control vertex.
Step 5：Determine the attitude in each characteristic point position for the welding gun, a reference Point C is respectively chosen on straight line and curve
And B, on the tangent extending line of curved welding seam characteristic point starting point, C point is on straight line striation for wherein B point.Angle BAC approximate representation is
The side subtended angle of oblique pipe.According to welding procedure and requirement, the position that welding gun is located should be oblique tube side face subtended angle angular bisector
Position.Welded Joint Curve expression formula U is obtained according to B-spline curves interpolation, by U derivation, thus obtaining along curve tangent line side
Direction vector X to (i.e. X-direction)x.VectorWithUnitization：Primary Calculation welding gun point
The direction vector of point Z axis：Ask for the direction vector of characteristic point Y direction further：Xy=Xz′×Xx；Obviously may be used
To draw, XxAnd XyIt is orthogonal, and XxAnd Xz' it cannot be guaranteed that vertical, so needing to Xz' direction vector travel direction
Revise：Xz=Xx×Xy；Determination according to this feature three direction vectors of point and then welding gun position corresponding to this feature point can be obtained
The attitude matrix put：
Step 6：Posture of welding torch discrete, according to the attitude matrix of each characteristic point obtaining in step 5, so solve every phase
The attitude of each discrete point between adjacent two characteristic points.Assume that the attitude matrix of the first and second characteristic points is respectively R1And R2, R1With
R2Between need discrete N number of point.We can obtain here：R1To R2Posture changing matrix：Rot1=R1\R2；Using machine
People's kinematics solution obtains R1And R2Between Eulerian angles gradient E between each discrete point1, and then obtain each two discrete point it
Between attitude gradient R01；Finally give the attitude matrix R of first discrete point11=R1*R01；Can obtain in the same manner second from
The attitude matrix of scatterplot is R12=R1*R01*R01, ask for the attitude matrix of next each discrete point by that analogy.
The curved welding seam automatic welding technique and the tradition that are intended for oblique pipe based on line-structured light that the present invention provides
Trajectory planning techniques be essentially different, the present invention combines machine vision well with robot kinematics,
Embody the nowadays developing direction of Automation Industry and demand for development, the teaching method having abandoned complicated backwardness is special to determine weld seam
Levy process a little.Show through real example, this algorithm has higher robustness and real-time, actual welding can meet welding
Various requirement and there is higher welding precision.
Brief description
Fig. 1 is the steps flow chart block diagram of algorithm；
Fig. 2 is image processing section FB(flow block)；
Fig. 3 is welding robot each Coordinate Conversion schematic diagram；
Fig. 4 is oblique tube model schematic diagram；
Specific embodiment
It is illustrated in figure 1 the curved welding seam automatic welding technique flow chart element being intended for oblique pipe based on line-structured light
Figure, is described further to the enforcement of the present invention below according to accompanying drawing and instantiation：
Step one：Adjusting Six-DOF industrial robot makes the industrial camera on its driving mechanical end and line-structured light
Sensor, makes structure light light belt irradiation position S1Place, is shot to the light belt on curved welding seam using industrial camera, and always
Photograph SnPlace, shoots 6 width images in real example, that is, photographs S6Place.And by shoot the Digital Image Transmission that obtains to image at
Reason device.
Step 2：Carry out image procossing for captured curved welding seam image, its detailed process is as follows：
Step 2-1, image is filtered：
Step 2-1-1, medium filtering：The Filtering Template selecting 3*3 carries out image procossing to weld seam.This template can be directed to
Dust, the salt-pepper noise bringing that splashes significantly are eliminated.
Step 2-1-2, LOG filters：The present invention selects to arrange to template using Gauss-Laplace filtering is one-dimensional.Its template
For s=[- 2；-2；-1；0；1；2；4；2；1；0；-1；-2；- 2], this operation can be very good to process these noises and empty noise
Internal so as to marginalisation.
Step 2-2, ROI region are extracted：A kind of extraction algorithm of new ROI region is proposed, this algorithm can make non-interested
The interference that region is brought is eliminated, and reduces data operation scale, improves the operation efficiency of image procossing.ROI region carries
Take algorithm as follows：
Step 2-2-1, reading one width gray level image, and the gray value of all pixels point on this width image is stored in one
In the middle of defined one-dimension array；
Step 2-2-2, determine ROI region shear parameters：Xmin(the x-axis coordinate of the starting point of clip image)：It is set to " 1 "；
Ymin：(the Y-axis coordinate of the starting point of clip image)：By the gray value scanning of Y direction, find a certain row vertical direction gray scale
The maximum pixel position of value, according to workpiece size, being put with this is to translate n pixel with reference to the Y-axis negative direction of this point
Position, and this is set to " Y in the position numerical value of Y directionmin”；Weight：Choose the width conduct of original image
" Weight " value；Height：Workpiece size, " Height " numerical value is set to h；I.e.：
Step 2-3, image binaryzation are processed, and select varimax herein.This algorithm be not required to setting that very important person is other
Parameter, is a kind of method automatically selecting threshold value；
Step 2-4, morphological erosion, used herein be 3*3 template processed so that the burr at striation edge and
Some isolate scattered point and are eliminated；Its template is：
Step 2-5, image thinning, by horizontally and vertically scanning direction, find out the point that all pixels value is 1, find out light
Band, in the ordinate of each row up-and-down boundary, is designated as y respectively1iAnd y2i.The width of light belt is designated as Δ yi, the ordinate note of center line
For ymidi, so center point coordinate is (xi, ymidi)；I.e.：
Step 2-6, the determination of weld bead feature points, its algorithm steps is as follows：
Step 2-6-1, constantly calculate the slope of line segment with n pixel as long measure from left to right, by whole line
The slope of section is represented with the type of curve.
Step 2-6-2, it is clear that the slope variation of line segment from slope variation curve, slope is learnt by analysis
Point corresponding to maximum is the position of the required weld bead feature points asked on curve.
Step 3：By the image coordinate of the curved welding seam obtaining in step 2 characteristic point through three-dimensional reconstruction, it is transformed into machine
Coordinate under device people's coordinate system.
Step 3-1, demarcated and line-structured light plane equation parameter M obtained by calibrating using camera interior and exterior parameter, false
The coordinate under image coordinate system for the p that sets up an office is (u1, v1), thus the focal length that can calculate p point in video camera is normalized into as flat
The imaging point P in face1c1Coordinate：
Step 3-2, hypothesis Plane Equation are：Ax+by+cz+1=0, wherein, a, b, c are each term system of plane equation
Number.Due to spatial point P1c1On the straight line that the optical axis center point of video camera is constituted with imaging point, you can know：
Step 3-3, the equation using this straight line and structure light plane equation, you can obtain characteristic point in camera coordinate system
Under three-dimensional coordinate pc(x, y, z), can obtain：
Step 3-4, by demarcate understand video camera to wrist transition matrix T.The six degree of freedom joint angle of robot can
To learn on teaching machine, can learn, through the normal solution of robot, the position auto-control that robot end is with respect to robot base
B.Finally, by formula pb=B*T*pc, p can be tried to achievecCoordinate pb under basis coordinates system for the point.
Step 4：To realize the interpolation of space weld curve using the theory of B-spline.Our roots first in welding processing
According to image processing algorithm and the three-dimensional reconstruction of early stage, get the three-dimensional coordinate of each characteristic point of curved welding seam, that is, type
Value point coordinates, then calculates control vertex.We are according to given offset point sequence Pi(i=1,2 ..., n) calculate control vertex
Vj(j=1 ..., 2, n+1, n+2) is so as to the B-spline Curve of definition passes through point range Pi(i=1,2 ..., n) and with PiFor song
The node of line segment.
Curve fitting process generally requires obtain n data point.6 data points being obtained by early stage in real example
To verify, curve hop count is divided into 5 sections, according to the coefficient of B-spline basic function：
Middle B-spline matrix is expressed as：
Then the first point of each section of curve is：
So, the 3rd section of curve can be drawn according to formula (8)：
4th section of curve：
5th section of (final stage) curve：
The end of final stage curve is put：P6=V8.
To sum up can draw：
When two ends take free end condition, equation group is：
Plus end-point condition, its head end summit and terminal vertex,
Synthesis is various above just can to obtain all of control vertex.
Step 5：Determine the attitude in each characteristic point position for the welding gun.One reference Point C is respectively chosen on straight line and curve
And B, on the tangent extending line of curved welding seam characteristic point starting point, angle BAC approximate representation is that the side of oblique pipe is opened to wherein B point
Angle.
Welded Joint Curve expression formula U is obtained according to B-spline curves interpolation, by U derivation, thus obtaining along curve tangent line side
To direction vector Xx.
According to welding procedure and requirement, the position that welding gun is located should be the position of oblique tube side face subtended angle angular bisector.?
VectorWithUnitization：
The direction vector of primary Calculation welding gun cusp Z axis：
Ask for the direction vector of characteristic point Y direction further：Xy=Xz′×Xx；
Obviously can draw, XxAnd XyIt is orthogonal, and XxAnd Xz' it cannot be guaranteed that vertical, so needing to Xz' side
To vectorial travel direction correction：Xz=Xz×Xy；
Three direction vectors of each characteristic point according to various determination above and then welding gun corresponding to this feature point can be obtained
The attitude matrix of position：
Step 6：Posture of welding torch discrete.According to the attitude matrix of each characteristic point obtaining in step 5, and then solve every two
The attitude of each discrete point between individual characteristic point；
Assume that the attitude matrix of the first and second characteristic points is respectively R1And R2, R1And R2Between need discrete N number of point.
We can obtain here：R1To R2Posture changing matrix：Rot1=R1\R2；
Solved using robot kinematics and obtain R1And R2Between Eulerian angles gradient E between each discrete point1, and then
Attitude gradient R between each two discrete point01；
Finally give the attitude matrix R of first discrete point11=R1*R01；
The attitude matrix that second discrete point can be obtained in the same manner is R12=R1*R01*R01, ask for by that analogy next
The attitude matrix of each discrete point.
The above-mentioned curved welding seam automatic welding technique being intended for oblique pipe based on line-structured light, can quickly identify
Curved welding seam the position of welding gun and attitude when can effectively control arc welding robot welding.Break and obtained by teaching in the past
Take the mode of curved welding seam data point, take leave of the welding method that arc welding robot welds machinery during parametric curve model,
The development tool realizing propulsion industrial automation is of great significance.
Claims (6)
1. a kind of curved welding seam solder technology method being intended for oblique pipe based on line-structured light, its step is as follows：
Industrial camera in step one, control Six-DOF industrial robot driving mechanical arm end carries out image to curved welding seam
Collection；
Step 2, Image semantic classification work is carried out according to the weld image that industrial camera is gathered, through feature point extraction algorithm,
Obtain the image coordinate of each characteristic point of weld seam and reference point；
Step 3, ask for the characteristic point that obtains using step 2 and reference point image coordinate carries out three-dimensional reconstruction, obtain these
Characteristic point and the robot coordinate of reference point；
Step 4, carry out space curve interpolation using the robot coordinate that step 3 is asked for；
Step 5, the angle being built according to each characteristic point on the space curve having obtained and each reference point, to determine weldering
Rifle is in the attitude of each characteristic point position；
Step 6, curved welding seam often between two neighboring characteristic point posture of welding torch discrete.
2. method according to claim 1 is it is characterised in that in step 2：Curved welding seam original image is in shooting
During, there are substantial amounts of smog and the noise jamming of sunlight reflection, image has cleverly used ROI area before being pre-processed
Domain extraction algorithm, this algorithm effectively eliminates substantial amounts of noise jamming, improves operation efficiency.Carry out intermediate value on this basis
Filtering, Log filtering, image binaryzation, Image erosion and image thinning operation, through feature point extraction algorithm, obtain each song
The image coordinate of line weld bead feature points.
3. method according to claim 1 is it is characterised in that in step 3, understand video camera to wrist by demarcating
Transition matrix T.The six degree of freedom joint angle of robot can be learnt from teaching machine, can learn through the normal solution of robot
Robot end is with respect to the position auto-control B of robot base.Finally, by formula pb=B*T*pc, p can be tried to achievecPoint is in basis coordinates
Coordinate pb under system.Wherein pcBe with camera coordinate system be with reference under coordinate points.
4. method according to claim 1 is it is characterised in that in step 4, machine vision and robot motion
Learn and cleverly combine, abandoned traditional method obtaining weld bead feature points by the method for teaching.According to B-spline
The coefficient of basic function obtains the B-spline matrix of each curved section, by using the characteristic point space coordinates acquired in step 3, i.e. type
Value point, finally obtains the 3 d space coordinate of each control vertex on curve.
5. method according to claim 1, can be in real time according to side surface of workpiece it is characterised in that in step 5
Angle is come controlling the attitude of welding gun so that the position of welding gun remains at the angular bisector position of side surface of workpiece angle.If
A certain weld bead feature points are A, respectively look for reference point B and C on this characteristic point both sides, angle B AC now constituting is workpiece side
The angle in face, Welded Joint Curve expression formula U being obtained according to B-spline curves interpolation, by U derivation, thus obtain characteristic point exist
Direction vector X along curve tangential direction (i.e. X-direction)x.VectorWithUnitization：Just
Step calculates the direction vector of welding gun cusp Z axis：Ask for the direction vector of characteristic point Y direction further：Xy=Xz′
×Xx；Obviously can draw, XxAnd XyIt is orthogonal, and XxAnd Xz' it cannot be guaranteed that vertical, so needing to Xz' direction
Vectorial travel direction correction：Xz=Xz×Xy.And XzDirection vector location also be curved welding seam angle angular bisector
Position.Determination according to this feature three direction vectors of the point and then attitude square of welding torch position corresponding to this feature point can be obtained
Battle array R.
6. method according to claim 1 is it is characterised in that in step 6, according to each feature obtaining in step 5
The attitude matrix of point, and then solve the attitude of often each discrete point between two neighboring characteristic point.Assume the first and second characteristic points
Attitude matrix be respectively R1And R2, R1And R2Between need discrete N number of point.We can obtain here：R1To R2Attitude
Transformation matrix：Rot1=R1\R2；Solved using robot kinematics and obtain R1And R2Between Eulerian angles ladder between each discrete point
Degree E1, and then obtain attitude gradient R between each two discrete point01；Finally give the attitude matrix R of first discrete point11=
R1*R01；Obtain the attitude matrix R of second discrete point in the same manner12=R1*R01*R01, by that analogy.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610922507.1A CN106425181A (en) | 2016-10-24 | 2016-10-24 | Curve welding seam welding technology based on line structured light |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610922507.1A CN106425181A (en) | 2016-10-24 | 2016-10-24 | Curve welding seam welding technology based on line structured light |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106425181A true CN106425181A (en) | 2017-02-22 |
Family
ID=58176588
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610922507.1A Pending CN106425181A (en) | 2016-10-24 | 2016-10-24 | Curve welding seam welding technology based on line structured light |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106425181A (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106826832A (en) * | 2017-02-27 | 2017-06-13 | 广东工业大学 | A kind of welding robot and its weld seam path learning method, system |
| CN107644441A (en) * | 2017-08-30 | 2018-01-30 | 南京大学 | Multi-foot robot complex road condition based on three-dimensional imaging is separated into point methods of stopping over |
| WO2018188276A1 (en) * | 2017-04-09 | 2018-10-18 | 北京工业大学 | Error modeling method for tail-end space curve trajectory of six-degree-of-freedom robot |
| CN108907526A (en) * | 2018-08-04 | 2018-11-30 | 苏州佩恩机器人有限公司 | A kind of weld image characteristic recognition method with high robust |
| CN110091333A (en) * | 2019-05-17 | 2019-08-06 | 上海交通大学 | The device and method of complex-curved surface weld feature identification and automatic grinding and polishing |
| CN110814584A (en) * | 2019-10-29 | 2020-02-21 | 创想智控科技（深圳）有限公司 | Welding method, device and system based on pre-scanning and electronic equipment |
| CN111558758A (en) * | 2020-05-21 | 2020-08-21 | 宁夏天地奔牛实业集团有限公司 | Automatic surfacing method for surface of mining sprocket chain nest |
| CN111702380A (en) * | 2020-05-18 | 2020-09-25 | 天津大学 | Welding process control method of welding robot |
| CN112238304A (en) * | 2019-07-18 | 2021-01-19 | 山东淄博环宇桥梁模板有限公司 | Method for automatically welding small-batch customized special-shaped bridge steel templates by mechanical arm based on image visual recognition of welding seams |
| CN112659123A (en) * | 2020-12-14 | 2021-04-16 | 南京工程学院 | Double-robot curve welding collaborative planning method based on line structure light vision |
| CN112834505A (en) * | 2020-12-31 | 2021-05-25 | 芜湖哈特机器人产业技术研究院有限公司 | Three-dimensional visual detection positioning device and method for pasted welding line of pipeline workpiece |
| CN113063348A (en) * | 2021-03-15 | 2021-07-02 | 南京工程学院 | Structured light self-perpendicularity arc-shaped weld scanning method based on three-dimensional reference object |
| CN113418927A (en) * | 2021-06-08 | 2021-09-21 | 长春汽车工业高等专科学校 | Automobile mold visual detection system and detection method based on line structured light |
| CN113681133A (en) * | 2021-08-30 | 2021-11-23 | 南京衍构科技有限公司 | Intelligent welding method of redundant degree of freedom robot with vision |
| CN114161048A (en) * | 2021-12-30 | 2022-03-11 | 常熟理工学院 | Iron tower foot parametric welding method and device based on 3D vision |
| CN115990896A (en) * | 2023-02-15 | 2023-04-21 | 浙江奥桑机械设备有限公司 | Welding robot control system based on image recognition |
| CN116175035A (en) * | 2023-03-20 | 2023-05-30 | 中国十七冶集团有限公司 | Intelligent welding method for steel structure high-altitude welding robot based on deep learning |
-
2016
- 2016-10-24 CN CN201610922507.1A patent/CN106425181A/en active Pending
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106826832A (en) * | 2017-02-27 | 2017-06-13 | 广东工业大学 | A kind of welding robot and its weld seam path learning method, system |
| WO2018188276A1 (en) * | 2017-04-09 | 2018-10-18 | 北京工业大学 | Error modeling method for tail-end space curve trajectory of six-degree-of-freedom robot |
| CN107644441A (en) * | 2017-08-30 | 2018-01-30 | 南京大学 | Multi-foot robot complex road condition based on three-dimensional imaging is separated into point methods of stopping over |
| CN108907526A (en) * | 2018-08-04 | 2018-11-30 | 苏州佩恩机器人有限公司 | A kind of weld image characteristic recognition method with high robust |
| CN110091333A (en) * | 2019-05-17 | 2019-08-06 | 上海交通大学 | The device and method of complex-curved surface weld feature identification and automatic grinding and polishing |
| CN110091333B (en) * | 2019-05-17 | 2022-05-06 | 上海交通大学 | Device and method for identifying and automatically grinding and polishing weld joint features on surface of complex curved surface |
| CN112238304A (en) * | 2019-07-18 | 2021-01-19 | 山东淄博环宇桥梁模板有限公司 | Method for automatically welding small-batch customized special-shaped bridge steel templates by mechanical arm based on image visual recognition of welding seams |
| CN112238304B (en) * | 2019-07-18 | 2023-01-06 | 山东淄博环宇桥梁模板有限公司 | Method for automatically welding small-batch customized special-shaped bridge steel templates by mechanical arm based on image visual recognition of welding seams |
| CN110814584A (en) * | 2019-10-29 | 2020-02-21 | 创想智控科技（深圳）有限公司 | Welding method, device and system based on pre-scanning and electronic equipment |
| CN111702380A (en) * | 2020-05-18 | 2020-09-25 | 天津大学 | Welding process control method of welding robot |
| CN111558758A (en) * | 2020-05-21 | 2020-08-21 | 宁夏天地奔牛实业集团有限公司 | Automatic surfacing method for surface of mining sprocket chain nest |
| CN111558758B (en) * | 2020-05-21 | 2021-10-26 | 宁夏天地奔牛实业集团有限公司 | Automatic surfacing method for surface of mining sprocket chain nest |
| CN112659123A (en) * | 2020-12-14 | 2021-04-16 | 南京工程学院 | Double-robot curve welding collaborative planning method based on line structure light vision |
| CN112834505A (en) * | 2020-12-31 | 2021-05-25 | 芜湖哈特机器人产业技术研究院有限公司 | Three-dimensional visual detection positioning device and method for pasted welding line of pipeline workpiece |
| CN113063348A (en) * | 2021-03-15 | 2021-07-02 | 南京工程学院 | Structured light self-perpendicularity arc-shaped weld scanning method based on three-dimensional reference object |
| CN113418927A (en) * | 2021-06-08 | 2021-09-21 | 长春汽车工业高等专科学校 | Automobile mold visual detection system and detection method based on line structured light |
| CN113681133A (en) * | 2021-08-30 | 2021-11-23 | 南京衍构科技有限公司 | Intelligent welding method of redundant degree of freedom robot with vision |
| CN113681133B (en) * | 2021-08-30 | 2022-07-08 | 南京衍构科技有限公司 | Intelligent welding method of redundant degree of freedom robot with vision |
| CN114161048A (en) * | 2021-12-30 | 2022-03-11 | 常熟理工学院 | Iron tower foot parametric welding method and device based on 3D vision |
| CN114161048B (en) * | 2021-12-30 | 2023-11-21 | 常熟理工学院 | 3D vision-based parameterized welding method and device for tower legs of iron tower |
| CN115990896A (en) * | 2023-02-15 | 2023-04-21 | 浙江奥桑机械设备有限公司 | Welding robot control system based on image recognition |
| CN115990896B (en) * | 2023-02-15 | 2023-06-23 | 浙江奥桑机械设备有限公司 | Welding robot control system based on image recognition |
| CN116175035A (en) * | 2023-03-20 | 2023-05-30 | 中国十七冶集团有限公司 | Intelligent welding method for steel structure high-altitude welding robot based on deep learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106425181A (en) | Curve welding seam welding technology based on line structured light | |
| CN210046133U (en) | Welding seam visual tracking system based on laser structured light | |
| CN108637435B (en) | Three-dimensional weld tracking system and method based on vision and arc pressure sensing | |
| CN105844622A (en) | V-shaped groove welding seam detection method based on laser visual sense | |
| CN108280856A (en) | The unknown object that network model is inputted based on mixed information captures position and orientation estimation method | |
| CN111721259B (en) | Underwater robot recovery positioning method based on binocular vision | |
| CN113146172B (en) | Multi-vision-based detection and assembly system and method | |
| CN104636760B (en) | A kind of localization method of weld seam | |
| CN106934813A (en) | A kind of industrial robot workpiece grabbing implementation method of view-based access control model positioning | |
| CN109470149B (en) | Method and device for measuring position and posture of pipeline | |
| CN109447939B (en) | Weld width prediction method before molten pool forming based on mixed spectrum vision | |
| CN112529858A (en) | Welding seam image processing method based on machine vision | |
| CN111359915B (en) | Material sorting method and system based on machine vision | |
| CN112819772A (en) | High-precision rapid pattern detection and identification method | |
| CN107014291A (en) | A kind of vision positioning method of the accurate transfer platform of material | |
| CN114155301A (en) | Robot target positioning and grabbing method based on Mask R-CNN and binocular camera | |
| CN111784655A (en) | Underwater robot recovery positioning method | |
| CN112926503A (en) | Automatic captured data set generation method based on rectangle fitting | |
| CN115816460A (en) | Manipulator grabbing method based on deep learning target detection and image segmentation | |
| CN112419237A (en) | Automobile clutch master cylinder groove surface defect detection method based on deep learning | |
| CN104966302B (en) | A kind of detection localization method of any angle laser cross | |
| CN109894779A (en) | A kind of machine vision tracking system and method | |
| CN114618838A (en) | Intelligent laser cleaning device based on machine vision and operation method thereof | |
| CN117269168B (en) | New energy automobile precision part surface defect detection device and detection method | |
| CN112588621B (en) | Agricultural product sorting method and system based on visual servo |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170222 |
|
| WD01 | Invention patent application deemed withdrawn after publication |