CN107168110A - A kind of material grasping means and system - Google Patents
A kind of material grasping means and system Download PDFInfo
- Publication number
- CN107168110A CN107168110A CN201611132346.2A CN201611132346A CN107168110A CN 107168110 A CN107168110 A CN 107168110A CN 201611132346 A CN201611132346 A CN 201611132346A CN 107168110 A CN107168110 A CN 107168110A
- Authority
- CN
- China
- Prior art keywords
- captured
- crawl
- feed tank
- body feed
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- 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
- G05B19/04—Programme control other than numerical control, i.e. in sequence controllers or logic controllers
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
Obtain the position of body feed tank, at least one material to be captured is included in the body feed tank, using the relative position relation and material information to be captured of body feed tank and material capture, the crawl information of material to be captured is obtained, grabbing device is using capturing information scratching material to be captured.The present invention contacts the body feed tank that information is captured with position feature information, material to be captured with material to be captured by known relative position relation, allow grabbing device at the same position and capture different quality, the object of shape, size, substantially increase the adaptability of material grasping system.
Description
Technical field
The present invention relates to storage, Material Sorting field.It is related to a kind of method of material crawl.
Background technology
Existing material grasping means to be captured mainly has pneumatic suction, electromagnetism to draw, machinery crawl three major types, wherein gas
Dynamic target of drawing has higher flatness requirement for the surface smoothness of object;Electromagnetism suction method has to the magnetic of object
Higher demand.
Existing grasping means is mainly the shape of vision system Direct Recognition object, by various grasping means to object
Captured, its systematicness to body form has higher requirement with uniformity.
Existing grasping means can be captured to doing market object according to mark, its size marked, mark
The attribute such as shape, color can only be marked for a type objects, same mark be difficult to object of different shapes such as
Pencil carries out same mark and crawl with drummed oil.
Existing grasping means is for having higher distinguish in summary in single shape, the crawl of the object of single kind
Knowledge and magnanimity and success rate, but if necessary to be captured to the object of sizes, shape, quality simultaneously, its Grasping skill is just
It is very limited.
The content of the invention
In view of the above-mentioned problems, the present invention provides one kind material grasping means to be captured, concrete technical scheme is acquisition material
At least one material to be captured is included in the position of case, the body feed tank, the relative position of body feed tank and material to be captured is utilized
Relation and material information to be captured, obtain the crawl information of material to be captured, grabbing device is waited to capture using information scratching is captured
Material.
Further, the camera by least one camera or with depth information obtains an at least sub-picture, identification
Fisrt feature in image, the position of body feed tank is obtained using fisrt feature.
Further, the body feed tank includes second feature, is the identity of body feed tank, by recognizing that second feature is obtained
Obtain the relative position relation and material information to be captured of body feed tank and material to be captured.
Further, the crawl information includes crawl amount of deflection, is obtained by the fisrt feature;
The crawl information includes crawl position, grasp force, crawl size, is obtained by the second feature.
Further, the material information to be captured includes the dimension information of material to be captured, the quality of material to be captured
Information, material to be captured surface to be captured are relative to the coefficient of friction on grabbing device crawl surface, the crawl chi of material to be captured
Very little information, the position of centre of gravity information of material to be captured, the dimension information of body feed tank.
Further, the camera or the camera with depth information are fixed in grabbing device space coordinate system
Fixed position or be fixed on grabbing device.
Further, the material to be captured in the body feed tank has consistent direction to be captured.
Further, the material information to be captured includes grasp force correction coefficient.
Further, the grabbing device includes force snesor, judges whether the value that force snesor is obtained is grasp force
Value, is if it is captured, if not then adjusting to the value of grasp force;
The grabbing device includes position sensor, judges whether the value that position sensor is obtained is the value for capturing size, if
To be captured, if not then adjust to crawl size value.
Further, it is described wait capture material for it is a variety of when, and material to be captured crawl attribute with certain threshold value
In the range of, using the different material to be captured of the relative position relation difference between various materials case and material to be captured;It is described
Capturing attribute includes grasp force, crawl size.
Further, after described material to be captured is crawled, the position corresponding to the material to be captured in body feed tank
It is designated as having captured.
The present invention additionally provides a kind of grasping system, including first acquisition unit, body feed tank position, the body feed tank are obtained
It is interior to include at least one material to be captured;
Second acquisition unit, using relative position relation and the material information to be captured of body feed tank and material to be captured, is treated
Capture the crawl information of material;
Control unit, control grabbing device is according to crawl information scratching material to be captured.
Further, the first acquisition unit by least one camera or the camera with depth information obtain to
Fisrt feature in a few sub-picture, identification image, the position of body feed tank is obtained using fisrt feature.
Further, the body feed tank includes second feature, is the identity of body feed tank, second acquisition unit is by knowing
Not described second feature obtains the relative position relation and material information of body feed tank and material to be captured.
Further, the crawl information includes crawl position, grasp force, crawl size, is obtained by the second feature;
The crawl information also includes crawl amount of deflection, is obtained by the fisrt feature.
Further, the material information to be captured includes dimension information, quality information, the thing to be captured of material to be captured
Expect that surface to be captured captures the coefficient of friction on surface, the crawl dimension information of material to be captured relative to grabbing device, waits to capture
The position of centre of gravity information of material.
Further, the camera or the camera with depth information are fixed in grabbing device space coordinate system
Fixed position or be fixed on grabbing device.
Further, the material to be captured in the body feed tank has consistent direction to be captured.
Further, the material information includes grasp force correction coefficient.
Further, the system also includes judging unit, and the grabbing device includes force snesor and position sensing
Device;
The judging unit judge force snesor obtain value whether be grasp force value, if control unit control crawl dress
Crawl is put, grabbing device is then adjusted if not control unit to the value of grasp force;
The judging unit judges whether the value that position sensor is obtained is the value for capturing size, if control unit control is grabbed
Take device to capture, grabbing device is then adjusted if not control unit to the value for capturing size.
Further, it is described wait capture material for it is a variety of when, and material to be grabbed crawl attribute with certain threshold value model
In enclosing, using a variety of materials to be captured to be captured material different from the relative position relation difference of body feed tank, the crawl category
Property includes:Grasp force, crawl size.
Further, after described material to be captured is crawled, the position corresponding to the material to be captured in body feed tank
It is designated as having captured.
The beneficial effects of the invention are as follows:The present invention will capture the material of information with position feature information, material to be captured
Case is contacted with material to be captured by known relative position relation, allows grabbing device while positioning and capturing not homogeneity
Amount, shape, the object of size, substantially increase the adaptability of material grasping system.
The present invention, without the method for Direct Recognition object to be captured, is carried significantly using indirect identification material position to be captured
The robustness that high visual identifying system is recognized for object space to be captured.
The present invention obtains material information to be captured by second feature, required for reducing Real time identification material to be captured
Amount of calculation, improves crawl speed.
The present invention installs position sensor, force snesor to feed back crawl information on grabbing device, to capturing
Journey formation closed loop, is improved to different quality, the adaptability of different shape material crawl to be captured.
After each material to be captured in body feed tank of the present invention is crawled, the material to be captured is to should body feed tank
Under position be designated as having captured;Crawl need not rejudge the change of image after completing every time, reduce amount of calculation, raising
Crawl speed.
Brief description of the drawings
Fig. 1 represents that recognition methods is material flow figure to be captured.
Fig. 2 is the functional block diagram of the grasping means.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is described.
The flow chart of material grasping means to be captured provided in an embodiment of the present invention shown in Fig. 1.Fig. 2 includes body feed tank
1st, fisrt feature 2, second feature 3, material to be captured 4, material limited block 5 to be captured, grabbing device 6, camera 7, processing storage
Device 8.
At least one material to be captured is placed in body feed tank;Fisrt feature and second feature are marked on body feed tank.
The position that body feed tank 1 is under world coordinate system is obtained by fisrt feature.
Using scaling board calibration for cameras 7, the inner parameter of camera 7 obtained surely.
Fixed position or be fixed on grabbing device 6 that camera 7 is fixed in the space coordinate system of grabbing device 6.
At least one material 4 to be captured is placed in body feed tank 1;The material to be captured 4 is spacing by material to be captured
The defined position of unit 5;The material to be captured 4 limits treating in direction, i.e., single body feed tank 1 by material limited block 5 to be captured
Capturing material has consistent direction to be captured;The material position-limiting unit 5 to be captured can be dividing plate, limited block, glue etc..
The direction to be captured of material to be captured in the body feed tank 1 and the preferred coordinates in body feed tank plane coordinate system
Axle has fixed angular relationship;The fixed angular relationship establishes the relation between body feed tank and material direction to be captured.
The fisrt feature 2 be specific texture, color, shape, or its combination, pass through camera 7 obtain image, processing
Memory 8 handles the coordinate system that the fisrt feature 2 obtains body feed tank 1, i.e. body feed tank plane coordinate system;Body feed tank coordinate system is put down
Face is overlapped with the upper plane of body feed tank 1.
Processing memory 8 handles the fisrt feature 2 and obtains body feed tank plane coordinate system relative between camera coordinates system
Conversion Relations.
Phase between a certain position that camera 7 is fixed within world coordinate system, i.e. camera coordinates system and world coordinate system
Mutual relation is known.
World coordinate system is overlapped with grabbing device task-oriented coordinates system.
Mutually changed by camera coordinates system, world coordinate system, body feed tank plane coordinate system, image coordinate system, obtain thing
The Conversion Relations of hopper coordinate system and world coordinate system, including in body feed tank plane coordinate system with material 4 to be captured
In the reference axis of fixed angular relationship relative to the rotation amount between corresponding reference axis under its world coordinate system, rotation amount note
For crawl amount of deflection.
Specifically:As shown in Fig. 2 the fisrt feature on body feed tank 1 is four round dots in plane on body feed tank.Wherein
Each round dot assignment has different coordinate values;
Four points in the plane of body feed tank 1 are recognized by camera 7 and body feed tank 1 and camera coordinates are obtained using the calculating of its coordinate value
Conversion Relations between system(R1, T1);
It regard the work coordinate system of grabbing device 6 as world coordinate system;
Camera 7 is fixed on a certain fixed position under world coordinate system, i.e., mutual between camera coordinates system and world coordinate system
Transformational relation (R2,T2) it is known;
Using above-mentioned(R1, T1)With (R2,T2), calculate the mutual conversion obtained between body feed tank coordinate system and world coordinate system and close
It is (R3, T3);
It is hereby achieved that body feed tank plane coordinate system in material 4 capture be in fixed angular relationship reference axis relative to it
Rotation amount under world coordinate system between corresponding reference axis.The rotation amount is designated as crawl amount of deflection M.
The position relationship of each material 4 to be captured and body feed tank plane coordinate system, can make up a position in body feed tank 1
Relation function f(X, y).Obtained body feed tank plane coordinates is wherein recognized relative to the fisrt feature 2 comprising material 4 to be captured
The coordinate value information fastened;And the corresponding seized condition attribute of the coordinate, the property value can be divided into two states, Yi Zhongshi
Do not capture, one kind is to have captured.
When it is described wait capture material for it is a variety of when, and material capture captures attribute with certain threshold range,
Using the different material to be captured of the relative position relation difference between various materials case and material to be captured, i.e., multiple positions are closed
It is function f(X, y);The crawl attribute includes grasp force, crawl size.
Holder is handled according to the Conversion Relations (R between body feed tank coordinate system and world coordinate system3, T3), and join
Vertical function f (x, y) obtains the location of material 4 to be captured correspondence world coordinates under body feed tank plane coordinate system so as to calculate
Coordinate (X under systemA,YA,ZA)。
The camera 7 can be binocular camera, or the camera with depth information, by binocular camera or with depth
The camera identification of degree information obtains waiting to capture under body feed tank plane coordinate system, crawl amount of deflection M and body feed tank plane coordinate system
Coordinate (X under the location of material correspondence world coordinate systemA,YA,ZA)。
Specifically:As shown in Fig. 2 the fisrt feature on body feed tank 1 is four round dots in plane on body feed tank;
By binocular camera 7 or the camera with depth information 7 identification the plane of body feed tank 1 on four points, and calculate obtain
Position relationship between each round dot, the coordinate system, body feed tank 1 and phase of body feed tank 1 are obtained using the position relationship between round dot
Conversion Relations between machine coordinate system(R1, T1);
Camera 7 is integrally fixed at a certain fixed position under world coordinate system, mutual between camera coordinates system and world coordinate system
Transformational relation (R2,T2) it is known;
Using above-mentioned(R1, T1)With (R2,T2), calculate the mutual conversion obtained between body feed tank coordinate system and world coordinate system and close
It is (R3, T3);
It is hereby achieved that body feed tank plane coordinate system in material 4 capture be in fixed angular relationship reference axis relative to it
Rotation amount under world coordinate system between corresponding reference axis.The rotation amount is that grabbing device 6 needs to adjust crawl direction
Amount of deflection, is designated as crawl amount of deflection M.
The position relationship of each material 4 to be captured and body feed tank plane coordinate system is, it is known that and composition one in body feed tank 1
Position relationship function f(X, y).Obtained body feed tank plane is wherein recognized relative to the fisrt feature 2 comprising material 4 to be captured
Coordinate value information on coordinate system;And the seized condition attribute under the coordinate, the property value can be divided into two states, a kind of
It is not capture, one kind is to have captured.
When it is described wait capture material for it is a variety of when, and material capture captures attribute with certain threshold range,
Using the different material to be captured of the relative position relation difference between various materials case and material to be captured, i.e., multiple positions are closed
It is function f(X, y);The crawl attribute includes grasp force, crawl size.
Holder is handled according to the Conversion Relations (R between body feed tank coordinate system and world coordinate system3, T3), and join
Vertical function f (x, y) obtains the location of material 4 to be captured correspondence world coordinates under body feed tank plane coordinate system so as to calculate
Coordinate (X under systemA,YA,ZA)。
Using indirect identification material position to be captured, without the method for Direct Recognition object to be captured, substantially increase and regard
Feel the robustness that identifying system is recognized for object space to be captured.
Using the relative position relation and material information to be captured of body feed tank and material to be captured, crawl information is obtained, its
Acquisition methods are as follows.
Second feature 3 is preset on body feed tank 1;The second feature 3 can be distributed on each surface of body feed tank 1, such as
Shown in figure, second feature 3 is distributed on the side of body feed tank 1, and the image for including second feature 3 is obtained by camera 7.
The body feed tank includes second feature, is the identity of body feed tank, by recognizing that second feature obtains body feed tank
With the relative position relation and material information to be captured of material to be captured.
The identity is sat comprising each material 4 to be captured in crawl force information, body feed tank 1 relative to body feed tank plane
Mark the position relationship f (x, y) for being, the physical attribute information of material to be captured 4, the size of body feed tank 1.
The dimension information of the body feed tank 1 includes the inside dimension of body feed tank 1:Long l1, width w1, high h1。
The physical attribute information of the material to be captured 4, includes the size of material to be captured 4:Limiting length l, width w,
With height h;The quality m of material to be captured;Material surface to be captured to be captured captures the friction system on surface relative to grabbing device 6
Number μ;Material 4 to be captured captures the dimension information of position;Correction coefficient k;Material 4 to be captured is relative to itself limiting length l, width
W, height h position of centre of gravity.
The dimension information of the crawl of material to be captured 4 position includes crawl position crawl depth, the crawl initial grasp width in position.
By the information obtained by fisrt feature 2 and second feature 3, calculated using following methods and obtain material 4 to be captured
Crawl information.
According to the quality m of material 4 to be captured, material to be captured surface to be captured in the attribute information of the material to be captured
The friction coefficient μ that surface is contacted with grabbing device 6 calculates the grasp force F obtained required for the theoretical crawl of material to be captured1
Value;
Considered based on system sexual factors such as quality of material error to be captured, mechanical mechanism error, material position of centre of gravitys to be captured, it is real
Border grasp force needs to increase correction coefficient k, resulting in the grasp force F required for material crawl to be captured.
Obtaining value of the material 4 to be captured under task-oriented coordinates system by fisrt feature 2 and the calculating of second feature 3 is
(XA,YA,ZA)。
According to the high h in the inside of body feed tank 1 in second feature 31, when material to be captured is with length direction vertical material case coordinate
When plane is put, the limiting length l of material to be captured in itself can calculate the Z for obtaining actual crawl positionBCoordinate value, thus
Coordinate value (the X of position is captured to material 4 to be capturedA,YA,ZB)。
So far crawl amount of deflection required for material to be captured, crawl size, grasp force, crawl position have been obtained capturing.
Material information to be captured is obtained by second feature, the calculating required for Real time identification material to be captured is reduced
Amount, improves crawl speed.
The body feed tank 1 is the body feed tank of at least one specification.
The quantity that the material to be captured 4 is in body feed tank 1 can be monitored in real time, and embodiment is as follows:
The location of each material 4 to be captured, which has, in body feed tank 1 treats crawl and does not capture attribute, such as should before capturing
The crawl attribute of the location of material to be captured is marked as 1, after the material to be captured is crawled, and the position is labeled
For 0;
The body feed tank 1 of memory cell in to(for) material quantity to be captured stores the material quantity to be captured in correspondence body feed tank 1
For n, when when capturing the present position of material 4 and being marked as 0, n counting just reduces 1;
When n is less than certain certain numerical value, material material system to be captured to be captured points out the material to be captured to handle shape in short supply
State;
When n is 0, represent that the material to be captured in the body feed tank 1 is crawled completely, the empty hopper 1 needs to carry out
Again feeding.
After each material to be captured in the body feed tank is crawled, the material to be captured to should be under body feed tank position
Put and be designated as having captured;Crawl need not rejudge the change of image after completing every time, reduce amount of calculation, improve crawl
Speed.
There is force snesor, position sensor, force snesor is to feed back crawl force information on grabbing device 6;Position is passed
Sensor is to feed back crawl dimension information.By feeding back crawl process formation closed loop, improve and different quality, shaped objects are grabbed
The adaptability taken.
Processing memory 8 captures material to be captured using the crawl information control grabbing device of acquisition.The crawl information
Including:Grab bits of information, crawl force information, crawl coordinate information, crawl amount of deflection.Specific crawl process is as follows:
Processing memory 8 is moved to target crawl position using coordinate information control grabbing device is captured;
According to crawl amount of deflection with grabbing bits of information, grabbing device adjustment crawl angle and initial grasp width;
According to bits of information is grabbed, grabbing device moves to predetermined crawl depth;
According to crawl force information, grabbing device treats crawl material and applies predetermined grasp force;
Grabbing device captures material to be captured.
This method will capture the body feed tank of information with position feature information, material to be captured and material to be captured passes through
Know relative position relation contact, allow grabbing device at the same position and capture different quality, the object of shape, size, greatly
The big adaptability for improving material grasping system.
Describing and explaining for the inventive method is applied to present system.
The flow chart of material grasping means to be captured provided in an embodiment of the present invention shown in Fig. 1.Fig. 2 includes body feed tank
1st, fisrt feature 2, second feature 3, material to be captured 4, material limited block 5 to be captured, grabbing device 6, camera 7, processing storage
Device 8.
At least one material to be captured is placed in body feed tank;Fisrt feature and second feature are marked on body feed tank.
In a particular embodiment, what the image that the first acquisition unit is obtained including camera and to camera was handled
Memory 8 is handled, the first acquisition unit obtains the position that body feed tank 1 is under world coordinate system by fisrt feature.
Using scaling board calibration for cameras 7, the inner parameter of camera 7 obtained surely.
Fixed position or be fixed on grabbing device 6 that camera 7 is fixed in the space coordinate system of grabbing device 6.
At least one material 4 to be captured is placed in body feed tank 1;The material to be captured 4 is spacing by material to be captured
The defined position of unit 5;The material to be captured 4 limits treating in direction, i.e., single body feed tank 1 by material limited block 5 to be captured
Capturing material has consistent direction to be captured;The material position-limiting unit 5 to be captured can be dividing plate, limited block, glue etc..
The direction to be captured of material to be captured in the body feed tank 1 and the preferred coordinates in body feed tank plane coordinate system
Axle has fixed angular relationship;The fixed angular relationship establishes the relation between body feed tank and material direction to be captured.
The fisrt feature 2 be specific texture, color, shape, or its combination, pass through camera 7 obtain image, processing
Memory 8 handles the coordinate system that the fisrt feature 2 obtains body feed tank 1, i.e. body feed tank plane coordinate system;Body feed tank coordinate system is put down
Face is overlapped with the upper plane of body feed tank 1.
Processing memory 8 handles the fisrt feature 2 and obtains body feed tank plane coordinate system relative between camera coordinates system
Conversion Relations.
Phase between a certain position that camera 7 is fixed within world coordinate system, i.e. camera coordinates system and world coordinate system
Mutual relation is known.
World coordinate system is overlapped with grabbing device task-oriented coordinates system.
Mutually changed by camera coordinates system, world coordinate system, body feed tank plane coordinate system, image coordinate system, obtain thing
The Conversion Relations of hopper coordinate system and world coordinate system, including in body feed tank plane coordinate system with material 4 to be captured
In the reference axis of fixed angular relationship relative to the rotation amount between corresponding reference axis under its world coordinate system, rotation amount note
For crawl amount of deflection.
Specifically:As shown in Fig. 2 the fisrt feature on body feed tank 1 is four round dots in plane on body feed tank.Wherein
Each round dot assignment has different coordinate values;
Four points in the plane of body feed tank 1 are recognized by camera 7 and body feed tank 1 and camera coordinates are obtained using the calculating of its coordinate value
Conversion Relations between system(R1, T1);
It regard the work coordinate system of grabbing device 6 as world coordinate system;
Camera 7 is fixed on a certain fixed position under world coordinate system, i.e., mutual between camera coordinates system and world coordinate system
Transformational relation (R2,T2) it is known;
Using above-mentioned(R1, T1)With (R2,T2), calculate the mutual conversion obtained between body feed tank coordinate system and world coordinate system and close
It is (R3, T3);
It is hereby achieved that body feed tank plane coordinate system in material 4 capture be in fixed angular relationship reference axis relative to it
Rotation amount under world coordinate system between corresponding reference axis.The rotation amount is designated as crawl amount of deflection M.
The position relationship of each material 4 to be captured and body feed tank plane coordinate system, can make up a position in body feed tank 1
Relation function f(X, y).Obtained body feed tank plane coordinates is wherein recognized relative to the fisrt feature 2 comprising material 4 to be captured
The coordinate value information fastened;And the corresponding seized condition attribute of the coordinate, the property value can be divided into two states, Yi Zhongshi
Do not capture, one kind is to have captured.
When it is described wait capture material for it is a variety of when, and material capture captures attribute with certain threshold range,
Using the different material to be captured of the relative position relation difference between various materials case and material to be captured, i.e., multiple positions are closed
It is function f(X, y);The crawl attribute includes grasp force, crawl size.
Holder is handled according to the Conversion Relations (R between body feed tank coordinate system and world coordinate system3, T3), and join
Vertical function f (x, y) obtains the location of material 4 to be captured correspondence world coordinates under body feed tank plane coordinate system so as to calculate
Coordinate (X under systemA,YA,ZA)。
The camera 7 can be binocular camera, or the camera with depth information, by binocular camera or with depth
The camera identification of degree information obtains waiting to capture under body feed tank plane coordinate system, crawl amount of deflection M and body feed tank plane coordinate system
Coordinate (X under the location of material correspondence world coordinate systemA,YA,ZA)。
Specifically:As shown in Fig. 2 the fisrt feature on body feed tank 1 is four round dots in plane on body feed tank;
By binocular camera 7 or the camera with depth information 7 identification the plane of body feed tank 1 on four points, and calculate obtain
Position relationship between each round dot, the coordinate system, body feed tank 1 and phase of body feed tank 1 are obtained using the position relationship between round dot
Conversion Relations between machine coordinate system(R1, T1);
Camera 7 is integrally fixed at a certain fixed position under world coordinate system, mutual between camera coordinates system and world coordinate system
Transformational relation (R2,T2) it is known;
Using above-mentioned(R1, T1)With (R2,T2), calculate the mutual conversion obtained between body feed tank coordinate system and world coordinate system and close
It is (R3, T3);
It is hereby achieved that body feed tank plane coordinate system in material 4 capture be in fixed angular relationship reference axis relative to it
Rotation amount under world coordinate system between corresponding reference axis.The rotation amount is that grabbing device 6 needs to adjust crawl direction
Amount of deflection, is designated as crawl amount of deflection M.
The position relationship of each material 4 to be captured and body feed tank plane coordinate system is, it is known that and composition one in body feed tank 1
Position relationship function f(X, y).Obtained body feed tank plane is wherein recognized relative to the fisrt feature 2 comprising material 4 to be captured
Coordinate value information on coordinate system;And the seized condition attribute under the coordinate, the property value can be divided into two states, a kind of
It is not capture, one kind is to have captured.
When it is described wait capture material for it is a variety of when, and material capture captures attribute with certain threshold range,
Using the different material to be captured of the relative position relation difference between various materials case and material to be captured, i.e., multiple positions are closed
It is function f(X, y);The crawl attribute includes grasp force, crawl size.
Holder is handled according to the Conversion Relations (R between body feed tank coordinate system and world coordinate system3, T3), and join
Vertical function f (x, y) obtains the location of material 4 to be captured correspondence world coordinates under body feed tank plane coordinate system so as to calculate
Coordinate (X under systemA,YA,ZA)。
Using indirect identification material position to be captured, without the method for Direct Recognition object to be captured, substantially increase and regard
Feel the robustness that identifying system is recognized for object space to be captured.
Using the relative position relation and material information to be captured of body feed tank and material to be captured, crawl information is obtained, its
Acquisition methods are as follows.
Second feature 3 is preset on body feed tank 1;The second feature 3 can be distributed on each surface of body feed tank 1, such as
Shown in figure, second feature 3 is distributed on the side of body feed tank 1, and the image for including second feature 3 is obtained by camera 7.
In a particular embodiment, what the image that the second acquisition unit is obtained including camera and to camera was handled
Equipment, the body feed tank includes second feature, is the identity of body feed tank, and the second acquisition unit is special by identification second
Levy the relative position relation and material information to be captured for obtaining body feed tank and material to be captured.
The identity is sat comprising each material 4 to be captured in crawl force information, body feed tank 1 relative to body feed tank plane
Mark the position relationship f (x, y) for being, the physical attribute information of material to be captured 4, the size of body feed tank 1.
The dimension information of the body feed tank 1 includes the inside dimension of body feed tank 1:Long l1, width w1, high h1。
The physical attribute information of the material to be captured 4, includes the size of material to be captured 4:Limiting length l, width w,
With height h;The quality m of material to be captured;Material surface to be captured to be captured captures the friction system on surface relative to grabbing device 6
Number μ;Material 4 to be captured captures the dimension information of position;Correction coefficient k;Material 4 to be captured is relative to itself limiting length l, width
W, height h position of centre of gravity.
The dimension information of the crawl of material to be captured 4 position includes crawl position crawl depth, the crawl initial grasp width in position.
By the information obtained by fisrt feature 2 and second feature 3, calculated using following methods and obtain material 4 to be captured
Crawl information.
According to the quality m of material 4 to be captured, material to be captured surface to be captured in the attribute information of the material to be captured
The friction coefficient μ that surface is contacted with grabbing device 6 calculates the grasp force F obtained required for the theoretical crawl of material to be captured1
Value;
Considered based on system sexual factors such as quality of material error to be captured, mechanical mechanism error, material position of centre of gravitys to be captured, it is real
Border grasp force needs to increase correction coefficient k, resulting in the grasp force F required for material crawl to be captured.
Obtaining value of the material 4 to be captured under task-oriented coordinates system by fisrt feature 2 and the calculating of second feature 3 is
(XA,YA,ZA)。
According to the high h in the inside of body feed tank 1 in second feature 31, when material to be captured is with length direction vertical material case coordinate
When plane is put, the limiting length l of material to be captured in itself can calculate the Z for obtaining actual crawl positionBCoordinate value, thus
Coordinate value (the X of position is captured to material 4 to be capturedA,YA,ZB)。
So far crawl amount of deflection required for material to be captured, crawl size, grasp force, crawl position have been obtained capturing.
Material information to be captured is obtained by second feature, the calculating required for Real time identification material to be captured is reduced
Amount, improves crawl speed.
The body feed tank 1 is the body feed tank of at least one specification.
The quantity that the material to be captured 4 is in body feed tank 1 can be monitored in real time, and embodiment is as follows:
The location of each material 4 to be captured, which has, in body feed tank 1 treats crawl and does not capture attribute, such as should before capturing
The crawl attribute of the location of material to be captured is marked as 1, after the material to be captured is crawled, and the position is labeled
For 0;
The body feed tank 1 of memory cell in to(for) material quantity to be captured stores the material quantity to be captured in correspondence body feed tank 1
For n, when when capturing the present position of material 4 and being marked as 0, n counting just reduces 1;
When n is less than certain certain numerical value, material material system to be captured to be captured points out the material to be captured to handle shape in short supply
State;
When n is 0, represent that the material to be captured in the body feed tank 1 is crawled completely, the empty hopper 1 needs to carry out
Again feeding.
After each material to be captured in the body feed tank is crawled, the material to be captured to should be under body feed tank position
Put and be designated as having captured;Crawl need not rejudge the change of image after completing every time, reduce amount of calculation, improve crawl
Speed.
The system of the present invention also includes judging unit, and the grabbing device includes force snesor and position sensor;Institute
State judging unit judge force snesor obtain value whether be grasp force value, if control unit control grabbing device grab
Take, grabbing device is then adjusted if not control unit to the value of grasp force;The judging unit judges that position sensor is obtained
Value whether be capture size value, if control unit control grabbing device crawl, then adjusted if not control unit
Grabbing device extremely captures the value of size.There is force snesor, position sensor, force snesor is grabbed to feed back on grabbing device 6
Power taking information;Position sensor is to feed back crawl dimension information.By feeding back crawl process formation closed loop, improve to difference
Quality, the adaptability of shaped objects crawl.
Control unit, material to be captured is captured using the crawl information control grabbing device of acquisition.The crawl packet
Include:Grab bits of information, crawl force information, crawl coordinate information, crawl amount of deflection.Specific crawl process is as follows:
Control unit, target crawl position is moved to using coordinate information control grabbing device is captured;
According to crawl amount of deflection with grabbing bits of information, grabbing device adjustment crawl angle and initial grasp width;
According to bits of information is grabbed, grabbing device moves to predetermined crawl depth;
According to crawl force information, grabbing device treats crawl material and applies predetermined grasp force;
Control unit control grabbing device captures material to be captured.
The system will capture the body feed tank of information with position feature information, material to be captured and material to be captured passes through
Know relative position relation contact, allow grabbing device at the same position and capture different quality, the object of shape, size, greatly
The big adaptability for improving material grasping system.
Claims (22)
1. the grasping means of material to be captured in a kind of body feed tank, it is characterised in that
Obtain and at least one material to be captured is included in the position of body feed tank, the body feed tank, utilize body feed tank and thing to be captured
The relative position relation of material and material information to be captured, obtain the crawl information of material to be captured, and grabbing device is believed using crawl
Breath captures material to be captured.
2. according to the method described in claim 1, it is characterised in that by least one camera or the phase with depth information
Machine obtains the fisrt feature in an at least sub-picture, identification image, and the position of body feed tank is obtained using fisrt feature.
3. method according to claim 2, it is characterised in that the body feed tank includes second feature, is the body of body feed tank
Part mark, by recognizing that second feature obtains body feed tank and the relative position relation of material to be captured and material to be captured letter
Breath.
4. method according to claim 3, it is characterised in that the crawl information includes crawl amount of deflection, by described
Fisrt feature is obtained;The crawl information includes crawl position, grasp force, crawl size, is obtained by the second feature.
5. according to the method described in claim 1, it is characterised in that the material information to be captured includes the chi of material to be captured
Very little information, the quality information of material to be captured, material to be captured surface to be captured relative to grabbing device capture surface friction
Crawl dimension information, the position of centre of gravity information of material to be captured, the dimension information of body feed tank of coefficient, material to be captured.
6. method according to claim 2, it is characterised in that the camera or the camera with depth information are fixed on
Fixed position in grabbing device space coordinate system is fixed on grabbing device.
7. according to the method described in claim 1, it is characterised in that the material to be captured in the body feed tank has consistent treat
Capture direction.
8. according to the method described in claim 1, it is characterised in that the material information to be captured includes grasp force and corrects system
Number.
9. method according to claim 4, it is characterised in that the grabbing device includes force snesor, judgment sensing
The value that device is obtained whether be grasp force value, if it is capture, if not then adjusting to the value of grasp force;
The grabbing device includes position sensor, judges whether the value that position sensor is obtained is the value for capturing size, if
To be captured, if not then adjust to crawl size value.
10. according to the method described in claim 1, it is characterised in that it is described wait capture material for it is a variety of when, and material to be captured
Crawl attribute with certain threshold range, using the relative position relation area between various materials case and material to be captured
Not different materials to be captured;The crawl attribute includes grasp force, crawl size.
11. according to the method described in claim 1, it is characterised in that after described material to be captured is crawled, this waits to capture
Position corresponding to material in body feed tank is designated as having captured.
12. a kind of grasping system, it is characterised in that including first acquisition unit, obtains in body feed tank position, the body feed tank and wraps
Containing at least one material to be captured;Second acquisition unit, is grabbed using the relative position relation of body feed tank and material to be captured with waiting
Material information is taken, the crawl information of material to be captured is obtained;Control unit, control grabbing device is waited to grab according to crawl information scratching
Take material.
13. system according to claim 12, it is characterised in that the first acquisition unit by least one camera or
There is person the camera of depth information to obtain the fisrt feature in an at least sub-picture, identification image, and thing is obtained using fisrt feature
The position of hopper.
14. system according to claim 12, it is characterised in that the body feed tank includes second feature, is body feed tank
Identity, second acquisition unit is by recognizing that the second feature obtains the relative position relation of body feed tank and material to be captured
And material information.
15. system according to claim 14, it is characterised in that the crawl information includes crawl position, grasp force, crawl
Size, is obtained by the second feature;
The crawl information also includes crawl amount of deflection, is obtained by the fisrt feature.
16. system according to claim 12, it is characterised in that the material information to be captured includes material to be captured
Dimension information, quality information, material to be captured surface to be captured capture the coefficient of friction on surface relative to grabbing device, wait to capture
The crawl dimension information of material, the position of centre of gravity information of material to be captured.
17. system according to claim 13, it is characterised in that the camera or the camera with depth information are fixed
Fixed position in grabbing device space coordinate system is fixed on grabbing device.
18. system according to claim 12, it is characterised in that the material to be captured in the body feed tank has consistent
Direction to be captured.
19. system according to claim 12, it is characterised in that the material information includes grasp force correction coefficient.
20. system according to claim 15, it is characterised in that the system also includes judging unit, the crawl dress
Put including force snesor and position sensor;
The judging unit judge force snesor obtain value whether be grasp force value, if control unit control crawl dress
Crawl is put, grabbing device is then adjusted if not control unit to the value of grasp force;
The judging unit judges whether the value that position sensor is obtained is the value for capturing size, if control unit control is grabbed
Take device to capture, grabbing device is then adjusted if not control unit to the value for capturing size.
21. system according to claim 12, it is characterised in that described when it is a variety of to capture material, and material to be grabbed
Crawl attribute with certain threshold range, distinguished not using the relative position relation of a variety of materials and body feed tank capture
Same material to be captured, the crawl attribute includes:Grasp force, crawl size.
22. system according to claim 12, it is characterised in that after described material to be captured is crawled, this waits to grab
The position corresponding to material in body feed tank is taken to be designated as having captured.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611132346.2A CN107168110A (en) | 2016-12-09 | 2016-12-09 | A kind of material grasping means and system |
PCT/CN2017/098243 WO2018036443A1 (en) | 2016-08-26 | 2017-08-21 | Material grabbing method, apparatus and system, and dynamometry apparatus and material case |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611132346.2A CN107168110A (en) | 2016-12-09 | 2016-12-09 | A kind of material grasping means and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107168110A true CN107168110A (en) | 2017-09-15 |
Family
ID=59848750
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611132346.2A Pending CN107168110A (en) | 2016-08-26 | 2016-12-09 | A kind of material grasping means and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107168110A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107160387A (en) * | 2016-12-13 | 2017-09-15 | 陈胜辉 | A kind of material grasping means and system |
CN108466268A (en) * | 2018-03-27 | 2018-08-31 | 苏州大学 | A kind of freight classification method for carrying, system and mobile robot and storage medium |
Citations (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
BE1017038A5 (en) * | 2006-03-07 | 2007-12-04 | Elaut N V | GRIPPING DEVICE AND METHOD FOR CONTROLLING THE GRIP FORCE. |
JP2009056844A (en) * | 2007-08-30 | 2009-03-19 | Toyota Motor Corp | Vehicular knee airbag device |
CN102826262A (en) * | 2011-06-17 | 2012-12-19 | 昆山金群力精密模具有限公司 | Automatic detecting and packing equipment |
CN102896884A (en) * | 2012-10-09 | 2013-01-30 | 温州大学 | Automatic printing production line |
CN102902271A (en) * | 2012-10-23 | 2013-01-30 | 上海大学 | Binocular vision-based robot target identifying and gripping system and method |
CN103753585A (en) * | 2014-01-10 | 2014-04-30 | 南通大学 | Method for intelligently adjusting manipulator and grasping force on basis of visual image analysis |
CN104635630A (en) * | 2013-11-07 | 2015-05-20 | 大连东方之星信息技术有限公司 | Archive management system |
CN104808370A (en) * | 2015-05-22 | 2015-07-29 | 合肥京东方光电科技有限公司 | Oppositely-closed equipment and alignment method |
CN104842362A (en) * | 2015-06-18 | 2015-08-19 | 厦门理工学院 | Method for grabbing material bag by robot and robot grabbing device |
CN204712055U (en) * | 2015-04-13 | 2015-10-21 | 上海金东唐科技股份有限公司 | The material grabbing mechanism of automatic fine tuning test machine |
CN105242194A (en) * | 2015-10-27 | 2016-01-13 | 苏州和瑞科自动化科技有限公司 | Circuit board test system provided with visual positioning system |
CN105253342A (en) * | 2015-10-10 | 2016-01-20 | 广州市万世德智能装备科技有限公司 | Method for multiple robots to cooperate to grab and pack bags into boxes |
CN105459136A (en) * | 2015-12-29 | 2016-04-06 | 上海帆声图像科技有限公司 | Robot vision grasping method |
CN105479461A (en) * | 2016-01-26 | 2016-04-13 | 先驱智能机械(深圳)有限公司 | Control method, control device and manipulator system |
CN205294235U (en) * | 2016-01-18 | 2016-06-08 | 刘文化 | System for robot snatchs whole material |
CN105945909A (en) * | 2016-05-13 | 2016-09-21 | 大族激光科技产业集团股份有限公司 | Error correction method and system for three-degree-of-freedom parallel robot |
CN106094516A (en) * | 2016-06-08 | 2016-11-09 | 南京大学 | A kind of robot self-adapting grasping method based on deeply study |
CN106181541A (en) * | 2015-05-29 | 2016-12-07 | 发那科株式会社 | Possesses the production system of the robot having position correction function |
-
2016
- 2016-12-09 CN CN201611132346.2A patent/CN107168110A/en active Pending
Patent Citations (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
BE1017038A5 (en) * | 2006-03-07 | 2007-12-04 | Elaut N V | GRIPPING DEVICE AND METHOD FOR CONTROLLING THE GRIP FORCE. |
JP2009056844A (en) * | 2007-08-30 | 2009-03-19 | Toyota Motor Corp | Vehicular knee airbag device |
CN102826262A (en) * | 2011-06-17 | 2012-12-19 | 昆山金群力精密模具有限公司 | Automatic detecting and packing equipment |
CN102896884A (en) * | 2012-10-09 | 2013-01-30 | 温州大学 | Automatic printing production line |
CN102902271A (en) * | 2012-10-23 | 2013-01-30 | 上海大学 | Binocular vision-based robot target identifying and gripping system and method |
CN104635630A (en) * | 2013-11-07 | 2015-05-20 | 大连东方之星信息技术有限公司 | Archive management system |
CN103753585A (en) * | 2014-01-10 | 2014-04-30 | 南通大学 | Method for intelligently adjusting manipulator and grasping force on basis of visual image analysis |
CN204712055U (en) * | 2015-04-13 | 2015-10-21 | 上海金东唐科技股份有限公司 | The material grabbing mechanism of automatic fine tuning test machine |
CN104808370A (en) * | 2015-05-22 | 2015-07-29 | 合肥京东方光电科技有限公司 | Oppositely-closed equipment and alignment method |
CN106181541A (en) * | 2015-05-29 | 2016-12-07 | 发那科株式会社 | Possesses the production system of the robot having position correction function |
CN104842362A (en) * | 2015-06-18 | 2015-08-19 | 厦门理工学院 | Method for grabbing material bag by robot and robot grabbing device |
CN105253342A (en) * | 2015-10-10 | 2016-01-20 | 广州市万世德智能装备科技有限公司 | Method for multiple robots to cooperate to grab and pack bags into boxes |
CN105242194A (en) * | 2015-10-27 | 2016-01-13 | 苏州和瑞科自动化科技有限公司 | Circuit board test system provided with visual positioning system |
CN105459136A (en) * | 2015-12-29 | 2016-04-06 | 上海帆声图像科技有限公司 | Robot vision grasping method |
CN205294235U (en) * | 2016-01-18 | 2016-06-08 | 刘文化 | System for robot snatchs whole material |
CN105479461A (en) * | 2016-01-26 | 2016-04-13 | 先驱智能机械(深圳)有限公司 | Control method, control device and manipulator system |
CN105945909A (en) * | 2016-05-13 | 2016-09-21 | 大族激光科技产业集团股份有限公司 | Error correction method and system for three-degree-of-freedom parallel robot |
CN106094516A (en) * | 2016-06-08 | 2016-11-09 | 南京大学 | A kind of robot self-adapting grasping method based on deeply study |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107160387A (en) * | 2016-12-13 | 2017-09-15 | 陈胜辉 | A kind of material grasping means and system |
CN108466268A (en) * | 2018-03-27 | 2018-08-31 | 苏州大学 | A kind of freight classification method for carrying, system and mobile robot and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107160387A (en) | A kind of material grasping means and system | |
US9920470B2 (en) | Method for capturing an item of laundry | |
CN110509281A (en) | The apparatus and method of pose identification and crawl based on binocular vision | |
CN109153521B (en) | Automated fabric sorting | |
JP5788460B2 (en) | Apparatus and method for picking up loosely stacked articles by robot | |
CN105437248B (en) | Robot system, robot device and workpiece pick-up method | |
CN108290286A (en) | Method for instructing industrial robot to pick up part | |
CN106839985B (en) | The automatic identification localization method of unmanned overhead traveling crane coil of strip crawl | |
CN104626169B (en) | Robot part grabbing method based on vision and mechanical comprehensive positioning | |
JP6000029B2 (en) | Handling system, handling method and program | |
CA2887261C (en) | Vision-assisted system and method for picking of rubber bales in a bin | |
JP5295828B2 (en) | Object gripping system and interference detection method in the system | |
CN104608149B (en) | The device of bulk product and method are taken out with robot | |
CN105654625B (en) | Method for classifying laundry, in particular dirty laundry | |
CN106044570B (en) | It is a kind of that automatic identification equipment and method are hung using the coil of strip of machine vision | |
CN109775376A (en) | The robot de-stacking method of irregular random material | |
CN109625986A (en) | Robot palletizing system and its palletizing method | |
DE102018111370B4 (en) | Workpiece holder system | |
CN105345585B (en) | The robot handling gripper and handling system of the engine cylinder body of view-based access control model | |
JPWO2009028489A1 (en) | Object detection method, object detection apparatus, and robot system | |
WO2020158060A1 (en) | Object grasping system | |
CN107168110A (en) | A kind of material grasping means and system | |
WO2021053750A1 (en) | Work robot and work system | |
WO2011074246A1 (en) | Textile article identification device and textile article gripping system | |
CN206340102U (en) | Magnetic reads the automatic vending machine of positioning |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170915 |