CN106774208A - Group's visual machine collaborative assembly method and model system - Google Patents
Group's visual machine collaborative assembly method and model system Download PDFInfo
- Publication number
- CN106774208A CN106774208A CN201611209458.3A CN201611209458A CN106774208A CN 106774208 A CN106774208 A CN 106774208A CN 201611209458 A CN201611209458 A CN 201611209458A CN 106774208 A CN106774208 A CN 106774208A
- Authority
- CN
- China
- Prior art keywords
- vision robot
- robot
- intelligent vision
- feeding
- control platform
- 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
- 238000000034 method Methods 0.000 title claims abstract description 42
- 230000000007 visual effect Effects 0.000 title claims abstract description 22
- 230000033228 biological regulation Effects 0.000 claims abstract description 13
- 238000003860 storage Methods 0.000 claims description 28
- 238000007599 discharging Methods 0.000 claims description 17
- 238000004140 cleaning Methods 0.000 claims description 8
- 239000000284 extract Substances 0.000 claims description 7
- 238000005520 cutting process Methods 0.000 claims description 6
- 238000001514 detection method Methods 0.000 claims description 5
- 230000009466 transformation Effects 0.000 claims description 4
- 239000002699 waste material Substances 0.000 claims description 4
- 239000002994 raw material Substances 0.000 claims description 3
- 238000004422 calculation algorithm Methods 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 claims description 2
- 238000006243 chemical reaction Methods 0.000 claims description 2
- 230000009977 dual effect Effects 0.000 claims description 2
- 238000006467 substitution reaction Methods 0.000 claims description 2
- 230000001629 suppression Effects 0.000 claims description 2
- 230000001186 cumulative effect Effects 0.000 claims 1
- 239000000463 material Substances 0.000 claims 1
- 238000004519 manufacturing process Methods 0.000 abstract description 12
- 238000010923 batch production Methods 0.000 abstract description 2
- 150000001875 compounds Chemical class 0.000 abstract description 2
- 230000000694 effects Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 206010033799 Paralysis Diseases 0.000 description 1
- 208000037656 Respiratory Sounds Diseases 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000002195 synergetic effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
- G05B19/41805—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by assembly
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Engineering & Computer Science (AREA)
- Manipulator (AREA)
- General Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Automatic Assembly (AREA)
Abstract
One population visual machine collaborative assembly method, including control platform, vision robot's work station and undertake between control platform and vision robot's work station communicate wireless module, parts number and species of the control platform according to needed for the sequence information of input is calculated, the quantity of intelligent vision robot devoted oneself to work, and according to actual working condition, dynamic regulation is operated in the quantity of the intelligent vision robot of disparate modules；Present invention also offers the model system of methods described, the present invention uses intelligent vision robot work compound, according to sequence information planning, corresponding program can be formulated according to different demands, hardware configuration need not be changed and increase sensor, flexibility is strong, Multi-varieties and Small-batch Production pattern can be met, and due to introducing vision sensor, operation can be simplified, work independently, can individually be repaired during failure, the work of whole system is not influenceed, robustness is good, production efficiency is greatly improved, production equipment input and floor space are saved again.
Description
Technical field
The invention belongs to technical field of industrial automatic control, more particularly to a population visual machine collaborative assembly method and mould
Type system.
Background technology
As economic fast development, competition become fiercer, the cycle of model change gradually reduces, for many
The demand of kind small lot batch manufacture pattern is increasingly strong, causes more concerns of the people to intelligent automation production line.At present
Although many automatic production lines can well complete task, cost is very high, and can be only done single work, complicated
Operation needs Duo Tai robots sequential working, once middle which link breaks down, whole system will paralyse, and produce effect
Rate is low, and floor space is big, high cost, it is necessary to multiple sensors collaboration completes to recognize work, when producing object and change, it is necessary to
Hardware is changed, increases cost, therefore very flexible, it is difficult to adapt to the production model of multi-varieties and small-batch.
The content of the invention
In order to overcome the shortcoming of above-mentioned prior art, it is an object of the invention to provide a population visual machine collaborative assembly
Method and model system, its is practical, adapts to various production objects, and control is flexible, and scalability is strong, and required space is small, respectively
Part independently working, can individually be repaired during failure, have no effect on the work of whole system, and robustness is good, can be real
Now efficiently, Flexible Production.
To achieve these goals, the technical solution adopted by the present invention is：
One population visual machine collaborative assembly method, including control platform 1, vision robot's work station 2 and undertake control
The wireless module 3 communicated between platform processed 1 and vision robot's work station 2, it is characterised in that the control platform 1 is according to defeated
The sequence information for entering calculate needed for parts number and species, the quantity of intelligent vision robot devoted oneself to work, and according to reality
The working condition on border, dynamic regulation is operated in the quantity of the intelligent vision robot of disparate modules.
Independent task is completed per class intelligent vision robot, is not interfere with each other, when a certain class robot breaks down, can
Individually to be repaired to it, the operation of whole system is had no effect on, with good robustness.When assembling object changes
When, control instruction is sent to vision robot's work station 2 by control platform 1, allow vision robot's work station 2 to run for this
The linkage editor of object, so can well meet the production model of multi-varieties and small-batch, and system flexibility is very strong.
Vision robot's work station 2 includes a conveyer 23 for annular, and feeding is disposed with along conveyer 23
Area 231, assembly section 233,238 and discharging area 235, feeding area 231 are disposed about cloth near element storage area 22, discharging area 235
It is set to product storage area 26, the feeding intelligent vision robot 21 of feeding area 231 and the blanking intelligent vision robot of discharging area 235
27 all have feeding module and cutting module, and control platform 1 is operated in both according to real-time assembling situation come dynamic regulation
The quantity of the intelligent vision robot of module, the assembling intelligent vision robot of assembly section only has load module, and each intelligence is regarded
Feel robot is equipped with the wireless module 211 communicated between control platform 1 and intelligent vision robot and and for seeking
The vision sensor 213 of path and element identification is looked for, the collaborative assembly method includes four collaborations of aspect：All intelligence are regarded
Feel robot cooperateed with the start and stop of conveyer 23, the collaboration between feeding intelligent vision robot 21 and conveyer 23, assembling
Intelligent vision robot 24,25,28,29 and cooperateing between conveyer 23, blanking intelligent vision robot 27 and conveyer
Collaboration between 23.
The control platform 1 is operated in the intelligent vision machine of disparate modules according to actual working condition, dynamic regulation
The process of the quantity of people is：
Control platform 1 sends order of starting working to vision robot's work station 2, and allows vision robot's work station 2 to transport
The program of the corresponding assembling object of row, all intelligent vision robots start simultaneously at work, and feeding intelligent vision robot 21 leads to
The image information for crossing the return of vision sensor 213 recognizes that path moves to element storage area 22 and picks required element, then
Move to feeding area 231 and start feeding, assembling intelligent vision robot recognizes required element 232 by vision sensor 242, and
Finished product 236 is placed on conveyer 23, blanking intelligent vision robot 27 recognizes qualified finished product by vision sensor 273
236, it is grabbed down from conveyer 23 and is placed on products storage area 26, all intelligent vision robots each step work in,
Job information is sent to control platform 1 by wireless module 3, control platform 1 is integrated information, to obtain conveyer
On number of elements, the finished product quantity on conveyer, add up qualified finished product quantity, intelligently regarded according to these information dynamic regulations
The quantity of the quantity for feeling robot and the intelligent vision robot for being operated in feeding module and cutting module, preferably to complete
Fittage.
The feeding process of the feeding intelligent vision robot 21 will cooperate with feeding, institute according to the speed of conveyer 23
State assembling intelligent vision robot 24,25,28,29 assembling process to be cooperateed with according to the speed of conveyer 23 crawl needed for
Element is assembled, and the blanking process of the blanking intelligent vision robot 27 will be cooperateed with according to the speed of conveyer 23 grabs
Lower finished product.
After the blanking intelligent vision robot 27 completes finished product blanking each time, blanking information is sent to control platform
1, when finished product is qualified, control platform counts the qualified quantity of finished product 261 and adds 1, and puts it into box 271, when finished product is unqualified,
Blanking intelligent vision robot 27 is put into waste product area after unqualified finished product is captured, and control platform 1 notifies that feeding is intelligently regarded
Feel robot 21, the quantity of each class component 232,234,237,239 of required sorting adds 1, counted when the qualified finished product 261 of assembling is counted on
When amount meets desired finished product quantity on order, control platform 1 is sent completely assignment instructions to vision robot's work station 2, institute
Assembling intelligent vision robot is stated to clean out the workpiece of rigging position 243, feeding intelligent vision robot 21 and blanking and intelligently regard
Feel that robot 27 is responsible for knocked-down element on cleaning conveyer 23, raw material storeroom 22, Ran Houyun is put back to after being reclaimed
Move to appointed place and send completion cleaning work instruction to control platform 1, control platform 1 sends halt instruction, all intelligence
Vision robot enters holding state, while the stop motion of conveyer 23.
Present invention also offers a kind of model system based on group's visual machine collaborative assembly method, its feature exists
In：
Control platform 1 is served as by a computer；
By a wireless receiving transmitter as wireless module 3；
It is main by colony intelligence vision robot, the conveyer 23 of annular, element storage area 22 and products storage area 26
Vision robot's work station 2 of part.
In vision robot's work station 2, conveyer 23 is arranged in center, the region at the major axis two ends of conveyer 23
It is respectively feeding area 231 and discharging area 235, the region parallel to major axis both sides is assembly section 1 and assembly section 2 238, unit
Part storage area 22 is arranged near feeding area 231, and products storage area 26 is arranged near discharging area 235, the feeding of feeding area 231
The blanking intelligent vision robot 27 of intelligent vision robot 21 and discharging area 235 all has feeding module and cutting module, control
Platform processed 1 is operated in the quantity of the intelligent vision robot of both modules according to real-time assembling situation come dynamic regulation, dress
Assembling intelligent vision robot 24,25,28,29 with area 1 and assembly section 2 238 only has load module, and each intelligence is regarded
Feel robot is equipped with the wireless module 211 communicated between control platform 1 and intelligent vision robot and and for seeking
Look for the vision sensor 213 of path and element identification.
In vision robot's work station 2, element to be assembled includes red base 221, black core 222, spring 223
With blue cap 224.First, the order comprising component kind and expectation processed finished products quantity by the input in control platform 1
Information, then control platform 1 by wireless module 3 to vision robot's workbench 2 send start-up operation instruct and run phase
The program answered, all intelligent vision robots 21,24,25,27,28,29 and conveyer 23 enter working condition, feeding intelligence
Vision robot 21 recognizes path by the image information that vision sensor 213 is obtained, and moves to element storage area 22, then
The image information that is obtained by vision sensor 213 extracts the features such as color, area, radius of circle come recognition component, according to control
The element that the instruction crawl that platform processed 1 sends is specified, then moves to feeding area 231, and the speed according to conveyer 23 starts
Feeding, the information that assembling intelligent vision robot 24,25,28,29 is returned by vision sensor 242, extract color, face successively
The feature such as product and radius of circle carries out template matches, and red base 221, black core 222, spring 223 and blue cap are recognized successively
224 start assembling, and the finished product 236 that will be assembled is placed on conveyer 231, and then blanking intelligent vision robot 27 passes through
The image information that vision sensor 273 is returned, extracts the features such as color, area, radius of circle and finished product is identified, and judge it
It is whether qualified, it is qualified, it is placed in box 271, it is unqualified, waste product area is placed on, while sending job information, control to control platform
Platform processed calculates qualified finished product number, when it meets order requirements, notifies that vision robot's workbench 1 performs clean-up task,
After the completion of clean-up task, control platform 1 sends halt instruction, all intelligent vision robots to vision robot's workbench 2
21st, 24,25,27,28,29 and conveyer 23 enter holding state.
Compared with prior art, group's machine vision Intelligent assembly Synergistic method of the invention uses intelligent vision machine with system
Device people's work compound, feeding work is planned according to sequence information, can formulate phase according to the demand of the different production object of client
The program answered, it is not necessary to change hardware configuration and increase sensor, flexibility is strong, can well meet multi-varieties and small-batch
Production model, and due to introducing vision sensor, operation can be simplified, work independently, can be independent during failure
The work for not influenceing whole system is repaired, robustness is good, and greatly improves production efficiency, production equipment input is saved again
And floor space.
Brief description of the drawings
Fig. 1 is the Organization Chart of group's visual machine collaborative assembly simulation system of the present invention.
Fig. 2 is the workflow diagram of group's visual machine collaborative assembly method of the present invention.
Fig. 3 is the workpiece identification template matches flow chart of group's visual machine collaborative assembly simulation system of the present invention.
Specific embodiment
Describe embodiments of the present invention in detail with reference to the accompanying drawings and examples.
As shown in figure 1, group's visual machine collaborative assembly model system of the present invention, including control platform 1, visual machine is artificial
Make station 2 and undertake the wireless module 3 communicated between control platform 1 and vision robot's work station 2, control platform 1 is integrated and regarded
Feel the job information that robot workstation 2 passes back, sent to vision robot's workbench 2 according to the job information after integration and controlled
System order.
Vision robot's work station 2 include element storage area 22, feeding area 231, assembly section 1, assembly section 2 238,
The 4 kinds of elements 221,222,223,224 needed for assembling are placed in discharging area 235 and products storage area 26, wherein element storage area 22,
Feeding area 231 and discharging area 235 are respectively positioned at the two ends of the oval major axis of conveyer 23, and assembly section one 233, assembly section
2 238 both sides for being located parallel to the major axis of conveyer 23, around conveyer 23 laid 6 intelligent vision robots 21,24,
25、27、28、29：1 feeding intelligent vision robot, 21,4 assembling 24,25,28,29 and 1 blankings of intelligent vision robot
Intelligent vision robot 27, each intelligent vision robot is furnished with the wireless mould communicated between control platform 1 and robot
Block 211,241,272 and vision sensor 213,242,273, wherein feeding area intelligent vision robot 21 and discharging area intelligently regards
Feel that robot 27 is removable intelligent vision robot, because they will be completed from the crawl raw material workpiece of element storage area 22 221-
224 and the finished product 261 that will process be sent to products storage area 26.And feeding intelligent vision robot 21 and blanking intelligent vision
Feeding program module and blanking program module are respectively provided with the program of robot 27, control platform 1 passes intelligent vision robot
The job information for returning is integrated, and which program work module they are in dynamic regulation, preferably to complete assembly work.
As shown in Fig. 2 the workflow of group's visual machine collaborative assembly method, comprises the following steps：
First, sequence information is input into by control platform 1, then control platform 1 show which kind of is assembled according to sequence information
The number of elements and species of object and needs.
Secondly, control platform 1 sends order of starting working to vision robot's workbench 2, and makes visual machine artificial
Make the program that platform 2 runs corresponding assembling object, all intelligent vision robots 21,24,25,27,28,29 and conveyer
23 enter working condition simultaneously
All intelligent vision robots 21,24,25,27,28,29 cooperate according to the instruction of control platform 1, feeding intelligence
Energy vision robot 21 recognizes that path moves to element storage area 22 by the image information that vision sensor 213 is returned, and passes through
The features such as image zooming-out color, area, the radius of circle that vision sensor 213 is returned carry out template matches, the element needed for identification
221st, 222,223,224, the element required for picking is put into box 212, and is counted, when box 212 is filled, feeding
Intelligent vision robot 21 moves to feeding area 231 by vision sensor 213, and feeding process will be according to the speed of conveyer 23
Spend to cooperate with feeding.Then, the element 214 in box 212 is placed on conveyer 23 successively, and the workpiece placed every time
Species will notify control platform 1, when not having element in box 212, feeding intelligent vision robot 21 by wireless module 211
Element storage area 22 is moved to again by vision sensor 213, starts to carry work, then transported by vision sensor 213
Move to feeding area 231 to be cooperateed with conveyer 23 and carry out feeding work, be so repeated, until receiving control platform 1
Assignment instructions are completed, after completion assignment instructions are received, the feeding area 231 of conveyer 23 is moved to, conveyer is cleared up
On workpiece, be returned to element storage area 22, then feeding area 231 is moved to by the image information that vision sensor 213 is returned
And by radio communication 211 notify control platform 1 complete clean-up task, when receive control platform 1 be stopped order after,
Feeding Visual intelligent robot 21 enters holding state.
Assembling intelligent vision robot 24,25,28,29 completes assembly work in a certain order, captures before this red
Color base 221, then captures black core 222, spring 223 and is sequentially placed into red base 221, finally buckles blue cap 224,
Tighten, complete an assembly work for finished product 236；Assembling intelligent vision robot 24,25,28,29 passes through vision sensor 242
The image of the workpiece 232,234,237,239 moved on echo-plex device 23, extracts color, area, radius of circle, average gray
Etc. feature, template matches are carried out, judge whether it is the workpiece needed for current assemble sequence, if sending the control command of crawl,
Then specified assembly area 243 is put it to, and the workpiece information that control platform 1 is captured is informed by wireless module 241, connect
The identification work into workpiece needed for next operation, if it is not, not sending control command then, required workpiece is waited until always
Occur, then send control command；When the intelligent vision robot 24,25,28,29 of assembly section 1, assembly section 2 238 is assembled
After what a finished product 236, control command is sent, be put into after being captured on conveyer 23, and notify that control platform 1 transmits dress
Put and be placed on 23 a finished product 236, subsequently into next round assembly work, and so on carry out being sent until control platform 1
Assignment instructions are completed, after being connected to completion task order, stops assembly work, if assembly section 1, assembly section 2 238 do not exist
Element, then assembly section 1, the intelligent vision robot 24,25,28,29 of assembly section 2 238 enter holding state, if in the presence of
, then be placed on element on conveyer 23, subsequently into holding state by element.
Whether blanking intelligent vision robot 27 recognizes on conveyer 23 there is finished product 236 by vision sensor 273, when
After having finished product to occur, the image of the workpiece by being moved on the conveyer 23 that vision sensor 273 is returned extracts color, face
The features such as product, radius of circle carry out template matches, recognize whether it is finished product first, and in the case where being finished product, have seen table
Whether face has crackle qualified to judge whether, if qualified, the box 271 of blanking intelligent vision robot 27 is put into after being captured
In, and notify that the quantity of qualified finished product 261 plus 1 by control platform 1, if unqualified, waste product area is put into after being captured, and will
Underproof information is sent to control platform 1, and the quantity of required each element 221-224 is added 1 by control platform 1, and notifies feeding
The quantity of each element 221,222,223,224 adds 1 needed for intelligent vision robot 21, when the box of blanking intelligent vision robot 27
After son 271 is filled, the image information returned by vision sensor 273 moves to products storage area 26, by finished product 261 all
After being discharged into products storage area 26, then the discharging area 235 of conveyer 23 is moved to, so worked repeatedly, until receiving control
After the completion task of platform processed 1, if having finished product in box 271, finished product is placed into discharge to products storage area 26, if not having,
The discharging area 235 for moving to conveyer 23 enters holding state.
Job information is passed through wireless by all intelligent vision robots 21,24,25,27,28,29 in the work of each step
Module 3 sends control platform 1 to, and control platform 1 is integrated information, to obtain the number of elements on conveyer, transmission
Finished product quantity on device, add up qualified finished product quantity, the quantity according to these information dynamic regulation intelligent vision robots with
And the quantity of the intelligent vision robot of feeding module and cutting module is operated in, preferably to complete fittage.
Finally, all intelligent vision robots 21,24,25,27,28,29 pass through radio communication 211,241 by job information
Control platform 1 is sent to, control platform 1 is compared with sequence information after the job information of return is integrated, judged whether
Into fittage, if completing, assignment instructions are sent completely to vision robot's work station 2, each region intelligent robot starts
Cleaning work is carried out, cleaning sends cleaning work and completes instruction after completing to control platform 1, and moves to designated area, then
Control platform 1 to vision robot's workbench 2 send halt instruction, all intelligent vision robots 21,24,25,27,28,
29 enter holding state, the stop motion of conveyer 23.
As shown in figure 3, intelligent vision robot is in workpiece identification, image is returned to by vision sensor, extract color,
The information such as elemental area, radius of circle, circle number, then compare with template successively, it is first determined whether it is red to meet color
More than threshold value, radius of circle within the specified range, if meeting, the target in image is red base for color, red pixel area
221, if it is not satisfied, continuing to determine whether to meet color for black and black area, average gray and radius of circle are in specified range
Interior, if meeting, the target in image is black core 222, if it is not satisfied, then judging whether to meet radius of circle, circle number and putting down
Within the specified range, if meeting, the target in image is spring 223 to equal gray scale, if it is not satisfied, then continuing to determine whether to meet
Color is less than threshold value for blue, blue pixel area less than threshold value, radius of circle, if meeting, the target in image is blue cap
Son 224, if it is not satisfied, continuing to determine whether to meet color for blue, blue pixel area is more than threshold more than threshold value, radius of circle
Value, if meeting, the target in image is finished product 236, if it is not satisfied, then matched without any template.Wherein feeding intelligence
Can vision robot 21 and the red base 221 of the assembling needs identification of intelligent vision robot 24,25,28,29, black core 222, bullet
Spring 223 and blue cap 224, blanking intelligent vision robot 27 only need to recognize finished product 236.
In the feature in extracting image, it can be directly that can determine whether by the RGB channel of access images to extract color, and
Whether then by the RGB scopes for accessing each pixel to meet and specifying, satisfaction is then added up elemental area, and what is finally obtained is tired
Plus and as elemental area.And the calculating on radius of circle and circle number will use loop truss, method is as follows：
First have to pre-process the image for obtaining：
Median filter process is carried out to image.Median filter uses each pixel value in center pixel square neighborhood
Median pixel value is replaced, and can remove the marginal information that denoising retains image again.Can be defined as in two dimensional image：
Yi=med { xij}=med { x(i+m),(j+n)(m,n)∈A,(i,j)∈I2}
Binary conversion treatment is carried out to image.Binaryzation makes image become simple, and data volume reduces, and can highlight interested
Objective contour.Definition m is predetermined threshold value, and f (x, y) is the gray value of pixel coordinate (x, y), and g (x, y) is the gray value for obtaining.
Can be expressed as：
Rim detection is carried out to image using Canny operators.The principle of Canny rim detections is limited with single order local derviation
Difference calculates amplitude and the direction of gradient.If f (x, y) is image, the gradient of f (x, y) is approximate with 2 × 2 first differences point
Formula calculates two array f of the partial derivative of x and yx' (x, y) and fy′(x,y)：
fx′(x,y)≈Gx=[f (x+1, y)-f (x, y)+f (x+1, y+1)-f (x, y+1)]/2
fy′(x,y)≈Gy=[f (x, y+1)-f (x, y)+f (x+1, y+1)-f (x+1, y)]/2
By first-order difference convolution maskWithObtain amplitude and the direction of 2 gradients：
In order to obtain rational edge, non-maxima suppression is carried out to gradient magnitude, carried out with dual threashold value-based algorithm detection and
The real edge of connection.
Expansion process is carried out to edge image.Expansion process is to carry out convolutional calculation to image and core, obtains kernel covering area
The max pixel value in domain so that the highlight regions in image increase, and are compensated to pixel, form UNICOM domain.It is to be located to define X
The image of reason, B is structural element, and X is by the result that B expands：
Then to using random Hough transformation loop truss by pretreated image, method is as follows：
The equation of circle is in two-dimensional space：
(x-a)2+(y-b)2=r2
In formula：(a, b) is central coordinate of circle, and r is round radius.Determine 3 unknown parameters of a, b, r, it is necessary to be taken on circle
3 point (x1,y1)、(x2,y2)、(x3,y3), obtain equation group during 3 points are substituted into above formula：
(x1-a)2+(y1-b)2=r2
(x2-a)2+(y2-b)2=r2
(x3-a)2+(y3-b)2=r2
Solve equation group and can obtain central coordinate of circle (a, b) and radius r.
The principle of random Hough transformation determines central coordinate of circle to randomly select 3 points in all of marginal point in the picture
(a1,b1) and radius r1.Then take in the 4th point of (x4,y4) substitute into first equation in, obtain radius r4, by r4In substitution following formula：
r4-r1=δ1
δ is error amount set in advance, works as δ1During less than δ, it is defined as candidate's circle.After determining candidate's circle, take a little generation
Enter to calculate, work as δ1-iAccumulator adds 1 during less than δ, when the value of accumulator reaches predetermined threshold, is defined as 1 proper circle.
In the present invention, the element letter that control platform 1 is returned according to all intelligent vision robots 21,24,25,27,28,29
Breath, finished product quantity information can dynamically update existing number of elements on conveyer 23, existing processing on conveyer 23
Good finished product, the accumulative finished product quantity for processing, when existing number of elements exceedes threshold value on conveyer 23, control platform
1 will send the order of pause feeding to feeding intelligent vision robot 21, and allow feeding intelligent vision robot 21 to go to assist
Blanking intelligent robot 27 goes to complete finished product blanking work, when existing number of elements is less than a certain threshold value on conveyer 23
When, control platform 1 is notified that blanking intelligent robot 27 goes to assist feeding robot 21 to complete feeding work, when conveyer 23
When the upper existing finished product quantity for processing exceedes a certain threshold value, control platform 1 is notified that feeding intelligent vision robot 21 is temporary
Stop feeding work, and notify that feeding intelligent vision robot 211 assists blanking intelligent vision robot 27 to complete finished product blankers
Make, when the quantity of qualified finished product meets order requirements, control platform 1 to all intelligent vision robots 21,24,25,27,
28th, completion assignment instructions are sent 29, and wait the flush instructions of conveyer 23 that feeding area intelligent vision robot 21 sends,
After receiving the instruction, control platform 1 sends halt instruction, all intelligent vision robots to vision robot's workbench 2
21st, 24,25,27,28,29 enter holding state, and conveyer 23 stops.
Claims (10)
1. a population visual machine collaborative assembly method, including control platform (1), vision robot's work station (2) and undertake
The wireless module (3) communicated between control platform (1) and vision robot's work station (2), it is characterised in that the control platform
(1) parts number and species needed for the sequence information according to input is calculated, the quantity of intelligent vision robot devoted oneself to work,
And according to actual working condition, dynamic regulation is operated in the quantity of the intelligent vision robot of disparate modules.
2. group's visual machine collaborative assembly method according to claim 1, it is characterised in that complete per class intelligent vision robot
Into independent task, it does not interfere with each other, when assembling object changes, by control platform (1) to vision robot's work station
(2) control instruction is sent, allows vision robot's work station (2) to run the linkage editor for this object.
3. group's visual machine collaborative assembly method according to claim 1, it is characterised in that vision robot's work station
(2) including a conveyer for annular (23), feeding area (231), assembly section (233,238) are disposed with along conveyer (23)
With discharging area (235), feeding area (231) are disposed about element storage area (22), and discharging area (235) are disposed about products storage area
(26), the blanking intelligent vision robot (27) of the feeding intelligent vision robot (21) of feeding area (231) and discharging area (235)
All there is feeding module and cutting module, control platform (1) is operated in both according to real-time assembling situation come dynamic regulation
The quantity of the intelligent vision robot of module, the assembling intelligent vision robot of assembly section only has load module, and each intelligence is regarded
Feel that robot is equipped with wireless module (211) and and the use communicated between control platform (1) and intelligent vision robot
In the vision sensor (213) for finding path and element identification, the collaborative assembly method includes four collaborations of aspect：It is all
Intelligent vision robot is cooperateed with the start and stop of conveyer (23), between feeding intelligent vision robot (21) and conveyer (23)
Collaboration, assembling intelligent vision robot (24,25,28,29) and cooperateing between conveyer (23), blanking intelligent vision machine
Cooperateing between people (27) and conveyer (23).
4. group's visual machine collaborative assembly method according to claim 3, it is characterised in that control platform (1) basis
Actual working condition, the process that dynamic regulation is operated in the quantity of the intelligent vision robot of disparate modules is：
Control platform (1) sends order of starting working to vision robot's work station (2), and allows vision robot's work station (2)
The program of the corresponding assembling object of operation, all intelligent vision robots start simultaneously at work, feeding intelligent vision robot
(21) image information returned by vision sensor (213) recognizes that path is moved to required for element storage area (22) is picked
Element, then moves to feeding area (231) and starts feeding, and assembling intelligent vision robot is recognized by vision sensor (242)
Required element (232), and finished product (236) is placed on conveyer (23), blanking intelligent vision robot (27) is by vision
Sensor (273) recognizes qualified finished product (236), it is grabbed down from conveyer (23) and is placed on products storage area (26), owns
Intelligent vision robot sends job information to control platform (1) in the work of each step by wireless module (3), controls
Platform (1) is integrated information, to obtain the number of elements on conveyer, the finished product quantity on conveyer, accumulative conjunction
Lattice finished product quantity, quantity according to these information dynamic regulation intelligent vision robots and is operated in feeding module and blanking die
The quantity of the intelligent vision robot of block, preferably to complete fittage.
5. group's visual machine collaborative assembly method according to claim 4, it is characterised in that the feeding intelligent vision machine
The feeding process of people (21) will cooperate with feeding according to the speed of conveyer (23), the assembling intelligent vision robot (24,
25th, 28 element needed for, assembling process 29) will cooperate with crawl according to the speed of conveyer (23) is assembled, it is described under
The blanking process of material intelligent vision robot (27) will be cooperateed with according to the speed of conveyer (23) grabs lower finished product.
6. group's visual machine collaborative assembly method according to claim 4, it is characterised in that the blanking intelligent vision machine
After people (27) completes finished product blanking each time, blanking information is sent to control platform (1), when finished product is qualified, control platform system
Count qualified finished product (261) quantity and add 1, and put it into box (271), when finished product is unqualified, blanking intelligent vision robot
(27) waste product area is put into after unqualified finished product is captured, and control platform (1) notifies feeding intelligent vision robot (21), institute
Each class component (232,234, the 237,239) quantity that need to be sorted adds 1, is ordered when qualified finished product (261) the quantity satisfaction of assembling is counted on
On list during desired finished product quantity, control platform (1) is sent completely assignment instructions, the dress to vision robot's work station (2)
Workpiece, feeding intelligent vision robot (21) and the blanking for cleaning out rigging position (243) with intelligent vision robot are intelligently regarded
Feel that robot (27) is responsible for knocked-down element in cleaning conveyer (23), raw material storeroom (22) put back to after being reclaimed,
Then move to appointed place and send completion cleaning work to control platform (1) and instruct, control platform (1) sends stopping and refers to
Order, all intelligent vision robots enter holding state, while conveyer (23) stop motion.
7. the model system of group's visual machine collaborative assembly method described in claim 1 is based on, it is characterised in that：
Control platform (1) is served as by a computer；
By a wireless receiving transmitter as wireless module (3)；
Based on colony intelligence vision robot, the conveyer (23) of annular, element storage area (22) and products storage area (26)
Want vision robot's work station (2) of part.
8. model system according to claim 7, it is characterised in that：
In vision robot's work station (2), conveyer (23) is arranged in center, the area at conveyer (23) major axis two ends
Domain is respectively feeding area (231) and discharging area (235), and the region parallel to major axis both sides is assembly section one (233) and assembly section
Two (238), element storage area (22) is arranged near feeding area (231), and it is attached that products storage area (26) are arranged in discharging area (235)
Closely, the blanking intelligent vision robot (27) of the feeding intelligent vision robot (21) of feeding area (231) and discharging area (235) is all
With feeding module and cutting module, control platform (1) is operated in both moulds according to real-time assembling situation come dynamic regulation
The quantity of the intelligent vision robot of block, assembly section one (233) and assembly section two (238) assembling intelligent vision robot (24,
25th, 28,29) there was only load module, each intelligent vision robot is equipped with and control platform (1) and intelligent vision robot
Between the wireless module (211) that communicates and and for finding the vision sensor (213) that path and element are recognized.
9. model system according to claim 7, it is characterised in that：
In vision robot's work station (2), element to be assembled includes red base (221), black core (222), spring
And blue cap (224) (223)；
Intelligent vision robot returns to image in workpiece identification by vision sensor, extracts color, elemental area, circle half
Footpath, circle number information, then compare with template successively, it is first determined whether it is red, red pixel area to meet color
More than threshold value, radius of circle within the specified range, if meeting, the target in image is red base (221), if it is not satisfied, after
It is continuous to judge whether to meet color for black and black area, average gray and radius of circle are within the specified range, if meeting, image
In target be black core (222), if it is not satisfied, then judging whether to meet radius of circle, circle number and average gray in specified model
In enclosing, if meeting, the target in image is spring (223), if it is not satisfied, then continue to determine whether to meet color for it is blue,
Blue pixel area is less than threshold value less than threshold value, radius of circle, if meeting, the target in image is blue cap (224), if not
Meet, continue to determine whether to meet color for blue, blue pixel area is more than threshold value more than threshold value, radius of circle, if meeting,
Target in image is finished product (236), if it is not satisfied, then matched without any template；Wherein feeding intelligent vision machine
People (21) and assembling intelligent vision robot (24,25,28,29) need the red base (221) of identification, black core (222), spring
(223) and blue cap (224), blanking intelligent vision robot (27) only needs to identification finished product (236).
10. model system according to claim 9, it is characterised in that：
In the feature in extracting image, color is directly judged by the RGB channel of access images, and elemental area is by visiting
Ask whether each pixel meets the RGB scopes specified, satisfaction is then added up, final obtain cumulative and as pixel faces
Product；The calculating of radius of circle and circle number will use loop truss, and method is as follows：
First have to pre-process the image for obtaining：
Median filter process is carried out to image, median filter is middle by each pixel value in center pixel square neighborhood
Pixel value is replaced, and can remove the marginal information that denoising retains image again, is defined in two dimensional image：
Yi=med { xij}=med { x(i+m),(j+n)(m,n)∈A,(i,j)∈I2}
Binary conversion treatment is carried out to image, binaryzation makes image become simple, data volume reduces, and can highlight target interested
Profile, definition m is predetermined threshold value, and f (x, y) is the gray value of pixel coordinate (x, y), and g (x, y) is the gray value for obtaining, and is represented
For：
Rim detection is carried out to image using Canny operators, the principle of Canny rim detections is with the finite difference of single order local derviation
To calculate amplitude and the direction of gradient, if f (x, y) is image, the gradient of f (x, y) divided with 2 × 2 first differences approximate expression come
Calculate two array f ' of the partial derivative of x and yx(x, y) and f 'y(x,y)：
f′x(x,y)≈Gx=[f (x+1, y)-f (x, y)+f (x+1, y+1)-f (x, y+1)]/2
f′y(x,y)≈Gy=[f (x, y+1)-f (x, y)+f (x+1, y+1)-f (x+1, y)]/2
By first-order difference convolution maskWithObtain amplitude and the direction of 2 gradients：
In order to obtain rational edge, non-maxima suppression is carried out to gradient magnitude, detected and connected with dual threashold value-based algorithm
Real edge；
Expansion process is carried out to edge image, expansion process is to carry out convolutional calculation to image and core, obtains kernel covering region
Max pixel value so that the highlight regions in image increase, and are compensated to pixel, form UNICOM domain.It is processed to define X
Image, B is structural element, and X is by the result that B expands：
Then to using random Hough transformation loop truss by pretreated image, method is as follows：
The equation of circle is in two-dimensional space：
(x-a)2+(y-b)2=r2
In formula：(a, b) is central coordinate of circle, and r is round radius.Determine a, b, r3 unknown parameter, it is necessary to take 3 points on circle
(x1,y1)、(x2,y2)、(x3,y3), obtain equation group during 3 points are substituted into above formula：
(x1-a)2+(y1-b)2=r2
(x2-a)2+(y2-b)2=r2
(x3-a)2+(y3-b)2=r2
Solve equation group and can obtain central coordinate of circle (a, b) and radius r,
The principle of random Hough transformation determines central coordinate of circle (a to randomly select 3 points in all of marginal point in the picture1,
b1) and radius r1, then take in the 4th point of (x4,y4) substitute into first equation in, obtain radius r4, by r4In substitution following formula：
r4-r1=δ1
δ is error amount set in advance, works as δ1During less than δ, it is defined as candidate's circle.After determining candidate's circle, take and substitute into a little meter
Calculate, work as δ1-iAccumulator adds 1 during less than δ, when the value of accumulator reaches predetermined threshold, is defined as 1 proper circle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611209458.3A CN106774208B (en) | 2016-12-23 | 2016-12-23 | Group's visual machine collaborative assembly method and model system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611209458.3A CN106774208B (en) | 2016-12-23 | 2016-12-23 | Group's visual machine collaborative assembly method and model system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106774208A true CN106774208A (en) | 2017-05-31 |
| CN106774208B CN106774208B (en) | 2017-12-26 |
Family
ID=58920344
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611209458.3A Expired - Fee Related CN106774208B (en) | 2016-12-23 | 2016-12-23 | Group's visual machine collaborative assembly method and model system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106774208B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108459572A (en) * | 2018-03-20 | 2018-08-28 | 广东美的制冷设备有限公司 | Monitoring method, device, system, robot and air conditioner production equipment |
| CN109060823A (en) * | 2018-08-03 | 2018-12-21 | 珠海格力智能装备有限公司 | Method and device for detecting coating quality of heat dissipation paste of radiator |
| CN109299720A (en) * | 2018-07-13 | 2019-02-01 | 沈阳理工大学 | A kind of target identification method based on profile segment spatial relationship |
| CN110561415A (en) * | 2019-07-30 | 2019-12-13 | 苏州紫金港智能制造装备有限公司 | Double-robot cooperative assembly system and method based on machine vision compensation |
| CN110580373A (en) * | 2018-06-07 | 2019-12-17 | 能力中心-虚拟车辆研究公司 | Preprocessing collaborative simulation method and device |
| CN111843981A (en) * | 2019-04-25 | 2020-10-30 | 广州中国科学院先进技术研究所 | Multi-robot cooperative assembly system and method |
| CN112157408A (en) * | 2020-08-13 | 2021-01-01 | 盐城工学院 | Industrial robot double-machine cooperation carrying system and method |
| CN112363470A (en) * | 2020-11-05 | 2021-02-12 | 苏州工业园区卡鲁生产技术研究院 | User-cooperative robot control system |
| CN112589401A (en) * | 2020-11-09 | 2021-04-02 | 苏州赛腾精密电子股份有限公司 | Assembling method and system based on machine vision |
| CN114115151A (en) * | 2021-11-24 | 2022-03-01 | 山东哈博特机器人有限公司 | Industrial robot cooperative assembly method and system based on MES |
| CN114161202A (en) * | 2021-12-29 | 2022-03-11 | 武汉交通职业学院 | Automatic industrial robot feeding and discharging system for numerical control machine tool |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104950684A (en) * | 2015-06-30 | 2015-09-30 | 西安交通大学 | Swarm robot collaborative scheduling measurement and control method and system platform |
| CN204725516U (en) * | 2015-01-19 | 2015-10-28 | 西安航天精密机电研究所 | A kind of single vision being applicable to pipelining coordinates multirobot navigation system |
| CN205734182U (en) * | 2016-06-30 | 2016-11-30 | 长沙长泰机器人有限公司 | For many group process equipment co-operating intelligent robot processing lines |
-
2016
- 2016-12-23 CN CN201611209458.3A patent/CN106774208B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN204725516U (en) * | 2015-01-19 | 2015-10-28 | 西安航天精密机电研究所 | A kind of single vision being applicable to pipelining coordinates multirobot navigation system |
| CN104950684A (en) * | 2015-06-30 | 2015-09-30 | 西安交通大学 | Swarm robot collaborative scheduling measurement and control method and system platform |
| CN205734182U (en) * | 2016-06-30 | 2016-11-30 | 长沙长泰机器人有限公司 | For many group process equipment co-operating intelligent robot processing lines |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108459572A (en) * | 2018-03-20 | 2018-08-28 | 广东美的制冷设备有限公司 | Monitoring method, device, system, robot and air conditioner production equipment |
| CN110580373A (en) * | 2018-06-07 | 2019-12-17 | 能力中心-虚拟车辆研究公司 | Preprocessing collaborative simulation method and device |
| CN110580373B (en) * | 2018-06-07 | 2023-09-15 | 虚拟汽车研究有限公司 | Preprocessing collaborative simulation method and device |
| CN109299720B (en) * | 2018-07-13 | 2022-02-22 | 沈阳理工大学 | Target identification method based on contour segment spatial relationship |
| CN109299720A (en) * | 2018-07-13 | 2019-02-01 | 沈阳理工大学 | A kind of target identification method based on profile segment spatial relationship |
| CN109060823A (en) * | 2018-08-03 | 2018-12-21 | 珠海格力智能装备有限公司 | Method and device for detecting coating quality of heat dissipation paste of radiator |
| CN111843981A (en) * | 2019-04-25 | 2020-10-30 | 广州中国科学院先进技术研究所 | Multi-robot cooperative assembly system and method |
| CN110561415A (en) * | 2019-07-30 | 2019-12-13 | 苏州紫金港智能制造装备有限公司 | Double-robot cooperative assembly system and method based on machine vision compensation |
| CN112157408A (en) * | 2020-08-13 | 2021-01-01 | 盐城工学院 | Industrial robot double-machine cooperation carrying system and method |
| CN112363470A (en) * | 2020-11-05 | 2021-02-12 | 苏州工业园区卡鲁生产技术研究院 | User-cooperative robot control system |
| CN112589401B (en) * | 2020-11-09 | 2021-12-31 | 苏州赛腾精密电子股份有限公司 | Assembling method and system based on machine vision |
| CN112589401A (en) * | 2020-11-09 | 2021-04-02 | 苏州赛腾精密电子股份有限公司 | Assembling method and system based on machine vision |
| CN114115151A (en) * | 2021-11-24 | 2022-03-01 | 山东哈博特机器人有限公司 | Industrial robot cooperative assembly method and system based on MES |
| CN114161202A (en) * | 2021-12-29 | 2022-03-11 | 武汉交通职业学院 | Automatic industrial robot feeding and discharging system for numerical control machine tool |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106774208B (en) | 2017-12-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106774208B (en) | Group's visual machine collaborative assembly method and model system | |
| CN107899814A (en) | A kind of robot spraying system and its control method | |
| US20180243776A1 (en) | Intelligent flexible hub paint spraying line and process | |
| CN109483573A (en) | Machine learning device, robot system and machine learning method | |
| CN106423656B (en) | Automatic spraying system and method based on cloud and images match | |
| CN110281231B (en) | Three-dimensional vision grabbing method for mobile robot for unmanned FDM additive manufacturing | |
| US20180210432A1 (en) | Industrial robot process cloud system and working method thereof | |
| CN202924613U (en) | Automatic control system for efficient loading and unloading work of container crane | |
| CN102923578A (en) | Automatic control system of efficient handing operation of container crane | |
| CN105500370B (en) | A kind of robot off-line teaching programing system and method based on body-sensing technology | |
| CN110456746A (en) | A kind of real-time scheduling method of multi items swinging cross automated production | |
| CN111906788B (en) | Bathroom intelligent polishing system based on machine vision and polishing method thereof | |
| CN104458748A (en) | Aluminum profile surface defect detecting method based on machine vision | |
| CN104299246B (en) | Production line article part motion detection and tracking based on video | |
| CN109726777A (en) | PCB appearance detection system and detection method Internet-based | |
| CN106681508A (en) | System for remote robot control based on gestures and implementation method for same | |
| CN114924513B (en) | Multi-robot cooperative control system and method | |
| CN107444644A (en) | A kind of unmanned plane movement supply platform and unmanned plane for orchard operation | |
| Niemueller et al. | Proposal for Advancements to the LLSF in 2014 and beyond | |
| CN107479552A (en) | Track machine people's self-organizing control system based on Agent | |
| CN206229582U (en) | A kind of full-automatic glue spraying streamline based on three-dimensional stereoscopic visual | |
| CN111352398A (en) | Intelligent precision machining unit | |
| CN117022971B (en) | Intelligent logistics stacking robot control system | |
| Christensen et al. | Integrating vision based behaviours with an autonomous robot | |
| CN204288242U (en) | Based on the Control During Paint Spraying by Robot trajectory extraction device that curved three-dimensional is rebuild |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20171226 Termination date: 20211223 |