Specific implementation mode
It is with reference to the accompanying drawings and embodiments, right in order to make the object, technical solution and advantage of the application be more clearly understood
The application is further elaborated.It should be appreciated that specific embodiment described herein is only used to explain the application, not
For limiting the application.
Robotic Hand-Eye Calibration method provided by the present application, can be applied in application environment as shown in Figure 1, and Fig. 1 is
The applied environment figure of Robotic Hand-Eye Calibration method in one embodiment.Wherein, the end of robot 10 includes that there are one removable
Dynamic arm tool 11 can be equipped with camera 20 and standard smelting tool 30, in machine by arm tool 11 in the end of robot 10
The lower section of 10 end of people is placed there are one fixed characteristic point 40, in addition, arm tool 11 is mounted on machine by ring flange 50
The end of device people 10, therefore the direction for calibrating smelting tool 30 is parallel with the normal direction of ring flange 50.
In one embodiment, as shown in Fig. 2, Fig. 2 is the flow chart of Robotic Hand-Eye Calibration method in one embodiment,
A kind of Robotic Hand-Eye Calibration method is provided, the application environment being applied in this way in Fig. 1 illustrates, the end of robot
End is equipped with camera and standard smelting tool, is provided with characteristic point on fixed calibration object, the above method includes the following steps:
Step S210:Obtain the first machine coordinates, the second machine coordinates and calibration pixel coordinate, wherein the first machine is sat
It is designated as moving control coordinate along reference direction alignment characteristics Dian Shi robots in the end of calibration smelting tool, reference direction is machine
The normal direction for the ring flange installed on people, the second machine coordinates are to be located at designated position opportunity in camera fields of view in characteristic point
The mobile control coordinate of device people, calibration pixel coordinate are special when characteristic point random device people is moved to designated position in camera fields of view
Position coordinates of the sign point in camera fields of view.
Standard smelting tool is indirectly by the end that ring flange is mounted on robot, therefore calibrates the placing direction and method of smelting tool
The normal direction of blue disk is parallel.When calibrating the end of smelting tool along reference direction alignment characteristics point, it is ensured that the edge of calibration smelting tool
Its placing direction alignment characteristics point, and calibrate the end alignment characteristics point of smelting tool.
The mobile control coordinate of robot can be used for spatial position and the control work of the arm tool of recorder people end
Tool arm is moved to specified spatial position.Therefore, the first machine coordinates can be aligned in the end of calibration smelting tool along reference direction
The mobile control coordinate of robot at this time is acquired when characteristic point, and is stored.Second machine coordinates can be located at camera in characteristic point
The mobile control coordinate of robot is acquired in the visual field when designated position, and is stored.
The magazine camera fields of view of robot end can include characteristic point, and the focal length of camera can make spy
Sign point can be identified clearly.Calibration pixel coordinate can be moved to designated position in camera fields of view as characteristic point random device people
When, in by the camera fields of view acquired in camera, position coordinates of the identification feature point in camera fields of view, and store.
Designated position includes multiple, may include upper left side, top, upper right side, left, centre, the right side in camera fields of view
Side, lower left, lower section and lower right.For example, keeping characteristic point stationary in the earth space coordinates, machine can be moved
People and make characteristic point respectively appear in upper left side in camera fields of view, top, upper right side, left, centre, right, lower left,
Lower section and lower right obtain the second machine coordinates and calibration pixel coordinate under 9 designated positions, can also obtain wherein arbitrary
The second machine coordinates and calibration pixel coordinate of number designated position.
Step S220:According to the first machine coordinates, the second machine coordinates and calibration pixel coordinate, perspective transform square is obtained
Battle array.
It is appreciated that from another angle when the position of robot is in the first machine coordinates, in order to enable characteristic point
It is moved to designated position in camera fields of view, actually need to control robot and robot is made to be moved to the second machine coordinates.
Therefore there are perspective transform relationships between the second machine coordinates and calibration pixel coordinate, according to the first machine coordinates, the second machine
Coordinate and calibration pixel coordinate can calculate the perspective transformation matrix of reflection perspective transform relationship.
Step S230:The mobile control coordinate of robot is obtained according to perspective transformation matrix and the position in camera fields of view is sat
Target mapping relations.
Step S240:The pixel coordinate of target point is converted into target machine coordinate according to mapping relations, wherein target point
Pixel coordinate include the position coordinates of target point in the camera, it is corresponding that target machine coordinate is that robot is moved to target point
Mobile control coordinate.
According to perspective transformation matrix, the pixel coordinate of target point can be converted into target machine coordinate, in order to realize
Robot is moved to target point, completes the hand and eye calibrating of robot.
Above-mentioned Robotic Hand-Eye Calibration method, the mobile control coordinate by obtaining robot and the position in camera fields of view
The mapping relations of coordinate may be implemented that the pixel coordinate of target point is converted into target machine coordinate according to mapping relations, complete
The hand and eye calibrating of robot can improve the precision that robot reaches crawl object position.
In one embodiment, as shown in figure 3, Fig. 3 is the flow chart that target machine coordinate is converted in one embodiment, root
The step of pixel coordinate of target point is converted into target machine coordinate according to mapping relations, includes the following steps:
Step S241:Obtain the pixel coordinate of initial machine coordinate and target point, wherein initial machine coordinate is to obtain
Mobile control coordinate where target point when camera fields of view.
While acquiring the pixel coordinate of target point, initial machine coordinate is acquired.
Step S242:According to initial machine coordinate, the pixel coordinate of target point and mapping relations, obtains target machine and sit
Mark.
Robot is in the position corresponding to initial machine coordinate, is closed according to the pixel coordinate of target point at this time and mapping
System, can obtain target machine coordinate.
Above-mentioned Robotic Hand-Eye Calibration method is obtained according to initial machine coordinate, the pixel coordinate of target point and mapping relations
Target machine coordinate is taken, the hand and eye calibrating of robot is completed, the precision that robot reaches crawl object position can be improved.
In one embodiment, as shown in figure 4, Fig. 4 is the flow chart that perspective transformation matrix obtains in one embodiment, root
The step of according to the first machine coordinates, the second machine coordinates and demarcating pixel coordinate, obtain perspective transformation matrix, including following step
Suddenly:
Step S221:Third machine coordinates are obtained according to the first machine coordinates and the second machine coordinates, wherein third machine
Coordinate is the coordinate of plane where the second machine coordinates are projected in the first machine coordinates.
By plane where the first machine coordinates, perspective transformation matrix is solved according to the third machine coordinates of acquisition, is reduced
The complexity and operand of calculating, and improve precision.
May include having the coordinate of X-axis, the coordinate of Y-axis and rotational coordinates in first machine coordinates.By the second machine coordinates
Plane where being projected in the first machine coordinates, i.e. angle corresponding to the rotational coordinates by the second machine coordinates according to the first machine coordinates
Degree is projected, then, the third machine coordinates that the second machine coordinates obtain after projection are the X-axis under the corresponding angle
With the coordinate of Y-axis.
For example, the first machine coordinates may include having the coordinate of X-axis, the coordinate of Y-axis and rotational coordinates, wherein rotational coordinates
It is 0, then after plane where the second machine coordinates are projected in the first machine coordinates that rotational coordinates is 0, to be interpreted as second
Machine coordinates are projected in the plane that rotation angle is 0.The third machine coordinates for projecting acquisition are the coordinate that rotation angle is 0,
The complexity and operand of calculating can be reduced, and improves precision.
Step S222:Perspective transformation matrix is obtained according to third machine coordinates and calibration pixel coordinate.
According to one-to-one relationship between third machine coordinates and calibration pixel coordinate, solves and obtain perspective transform square
Battle array, the mobile control coordinate and the position coordinates in camera fields of view that the perspective transformation matrix of acquisition can be used for representation robot
Mapping relations.
Above-mentioned Robotic Hand-Eye Calibration method obtains third machine according to the first machine coordinates and the second machine coordinates and sits
Mark obtains perspective transformation matrix according to third machine coordinates and calibration pixel coordinate, can reduce the complexity and fortune of calculating
Calculation amount, and improve precision.
In one embodiment, the step of perspective transformation matrix being obtained according to third machine coordinates and calibration pixel coordinate,
Include the following steps:
According to Qi=Pi* A obtains perspective transformation matrix, wherein i is the serial number of designated position, QiFor i-th of third machine
Coordinate, PiFor i-th of calibration pixel coordinate, A is perspective transformation matrix.
Above-mentioned Robotic Hand-Eye Calibration method according to third machine coordinates and the determination of calibration pixel coordinate and obtains perspective change
Matrix is changed, the complexity and operand of calculating can be reduced, and improves precision.
In one embodiment, the step of obtaining third machine coordinates according to the first machine coordinates and the second machine coordinates,
Include the following steps:
According toThird machine coordinates are obtained,
In, i is the serial number of designated position, Qi(xi,yi) it is i-th of third machine coordinates, xiAnd yiRespectively i-th of third machine coordinates
Abscissa and ordinate, vxi、vyiAnd vriAbscissa, ordinate and the rotational coordinates of respectively i-th second machine coordinates,
x0And y0The abscissa and ordinate of respectively the first machine coordinates.
Above-mentioned Robotic Hand-Eye Calibration method, x0And y0The abscissa and ordinate of respectively the first machine coordinates, the first machine
The rotational coordinates of device coordinate is 0, and the first machine coordinates for being 0 according to rotational coordinates calculate third machine coordinates, can be obtained same
The third machine coordinates that sample rotational coordinates is 0, can reduce the complexity and operand of calculating, and improve precision.
In one embodiment, as shown in figure 5, Fig. 5 is the flow chart that perspective transformation matrix solves in one embodiment, root
The step of obtaining perspective transformation matrix according to third machine coordinates and calibration pixel coordinate, includes the following steps:
Step S223:Perspective transform equation is established according to third machine coordinates and calibration pixel coordinate.
According under designated position third machine coordinates and calibration pixel coordinate in, the mobile control of representative robot
The relationship between position coordinates in coordinate and camera fields of view, establishes perspective transform equation, is used for subsequent perspective transformation matrix
Solution.
Step S224:Perspective transformation matrix is solved according to perspective transform equation.
According to perspective transform equation, determining equilibrium relationships can accurately solve perspective transformation matrix.
Above-mentioned Robotic Hand-Eye Calibration method, the perspective transform etc. established by third machine coordinates and calibration pixel coordinate
Formula can accurately solve perspective transformation matrix, in order to subsequently improve the precision that robot reaches crawl object position.
In one embodiment, the step of perspective transform equation being established according to third machine coordinates and calibration pixel coordinate,
Include the following steps:
Step S225:It determinesFor third machine coordinates and calibration pixel
Transformation of coordinates relationship, wherein i is the serial number of designated position, xiAnd yiThe abscissa of respectively i-th third machine coordinates and
Ordinate, ziMeet zi=a13ui+a23wi+a33, uiAnd wiThe abscissa and ordinate of respectively i-th calibration pixel coordinate,For perspective transformation matrix, a11、a12、a13、a21、a22、a23、a31、a32And a33Respectively perspective transformation matrix
Element.
According in the third machine coordinates and calibration pixel coordinate under designated position, third machine coordinates and calibration picture are obtained
Plain transformation of coordinates relationship, transformation relation can be used for representing the mobile control coordinate of robot and the position in camera fields of view is sat
Relationship between mark.
Wherein, 1 can regard given reference data as, which can be definite value, and the reference data is in addition to being 1
In addition, other definite values be can also be.It is calculated in addition, can facilitate and simplify when reference data is 1, improves accuracy.
Step S226:It is obtained according to transformation relationWithWherein,WithFor perspective transform equation, i is the serial number of designated position, xiAnd yi
The abscissa and ordinate of respectively i-th third machine coordinates, uiAnd wiThe abscissa of respectively i-th calibration pixel coordinate
And ordinate, a11、a12、a13、a21、a22、a23、a31、a32And a33The respectively element of perspective transformation matrix.
According to perspective transform equation acquired under multiple designated positions, can accurately solve in perspective transformation matrix
Element.
It is important that in addition, by the perspective transformation matrix solved after perspective transform equation, can calculate
Angle between plane where plane where ring flange and camera fields of view calculates the focal plane of camera and robot after installing
Error between plane where ring flange.Wherein, a13The angle between X-axis and U axis, a can be embodied23Y-axis and W can be embodied
Angle between axis, a11The proportionality coefficient between X-axis and U axis, a can be embodied12The ratio system between X-axis and W axis can be embodied
Number, a21The proportionality coefficient between Y-axis and U axis, a can be embodied22The proportionality coefficient between Y-axis and W axis, a can be embodied31It can be with
Embody the translational movement between X-axis and U axis, a32The translational movement between Y-axis and W axis, a can be embodied33a33It under special circumstances can be with
It is 1.X-axis is the reference axis where x, and Y-axis is the reference axis where y, and U axis is the reference axis where u, and W axis is the coordinate where w
Axis.
Above-mentioned Robotic Hand-Eye Calibration method, by determining third machine coordinates and demarcating the transformation relation of pixel coordinate,
And perspective transform equation is obtained, the element in perspective transformation matrix can be accurately solved, perspective transformation matrix is obtained.
In one embodiment, in mobile control coordinate and camera fields of view that robot is obtained according to perspective transformation matrix
The step of mapping relations of position coordinates, include the following steps:
According toObtain the shifting of robot
The mapping relations of dynamic control coordinate and the position coordinates in camera fields of view, wherein x and y is respectively that the mobile control of robot is sat
Target abscissa and ordinate, u and w are respectively the abscissa and ordinate of the position coordinates in camera fields of view, a11、a12、a13、
a21、a22、a23、a31、a32And a33The respectively element of perspective transformation matrix, xt、ytAnd rtThe respectively horizontal seat of initial machine coordinate
Mark, ordinate and rotational coordinates, initial machine coordinate are the mobile control coordinate when obtaining camera fields of view where target point.
Above-mentioned Robotic Hand-Eye Calibration method can accurately obtain machine after the perspective transformation matrix after solution
The mapping relations of the mobile control coordinate of people and the position coordinates in camera fields of view, in order to subsequently obtain target machine coordinate,
The hand and eye calibrating of robot is completed, and improves the precision that robot reaches crawl object position.
In one embodiment, the pixel coordinate of target point is converted into the step of target machine coordinate according to mapping relations
Suddenly, include the following steps:
Obtain the pixel coordinate of initial machine coordinate and target point;
According toObtain target machine
Device coordinate, wherein xfAnd yfThe respectively abscissa and ordinate of target machine coordinate, utAnd wtThe respectively pixel of target point
The abscissa and ordinate of coordinate, a11、a12、a13、a21、a22、a23、a31、a32And a33The respectively element of perspective transformation matrix,
xt、ytAnd rtThe respectively abscissa of initial machine coordinate, ordinate and rotational coordinates.
Above-mentioned Robotic Hand-Eye Calibration method is obtained according to initial machine coordinate, the pixel coordinate of target point and mapping relations
Target machine coordinate is taken, the hand and eye calibrating of robot is completed, the precision that robot reaches crawl object position can be improved.
While acquiring the pixel coordinate of target point, initial machine coordinate is acquired.Robot is in initial machine
In position corresponding to device coordinate, according to the pixel coordinate and mapping relations of target point at this time, target machine coordinate can be obtained.
In one embodiment, according to perspective transform equation solve perspective transformation matrix the step of after, further include with
Lower step:
According toObtain the perspective transformation matrix so that ε value minimums, wherein i
For the serial number of designated position, xiAnd yiThe abscissa and ordinate of respectively i-th third machine coordinates, uiAnd wiRespectively i-th
The abscissa and ordinate of a calibration pixel coordinate, a11、a12、a13、a21、a22、a23、a31、a32And a33Respectively perspective transform square
The element of battle array.
Above-mentioned Robotic Hand-Eye Calibration method, by acquisition so that the perspective transformation matrix of ε value minimums, can obtain optimal
Change and closest to actual perspective transformation matrix, further decrease error, improve the accuracy of perspective transformation matrix, in order to can
To improve the precision that robot reaches crawl object position.
In another embodiment, as shown in fig. 6, Fig. 6 is the stream of Robotic Hand-Eye Calibration method in another embodiment
Cheng Tu, Robotic Hand-Eye Calibration method includes the following steps in the present embodiment:
The first machine coordinates and the second machine coordinates of robot are obtained, and obtain the calibration of characteristic point in camera fields of view
Pixel coordinate matches corresponding second machine coordinates and calibration pixel coordinate under 9 groups of designated positions.By camera and standard smelting tool peace
Mounted in the end of robot, placed there are one fixed characteristic point in the lower section of robot end, adjustment camera makes phase
The focal plane of machine can plane where clear identification feature point and characteristic point, this feature point can be easily identified and uniquely,
The shape of characteristic point can be circle, circular hole etc..The center of the end alignment characteristics point of smelting tool will be calibrated, and records this opportunity
The mobile control coordinate of device people is the first machine coordinates.It is moved by robot and drives camera fields of view variation, and make characteristic point
Upper left side, top, upper right side, left, centre, right, lower left, lower section and the bottom right in camera fields of view can be respectively appeared in
It is square, while obtaining the calibration pixel coordinate under this 9 designated positions, and acquire this 9 calibration pixel coordinates, obtain corresponding
9 the second machine coordinates;The calibration picture of characteristic point can be handled image and be extracted by acquiring the image of camera at this time
Plain coordinate.According to the first machine coordinates, 9 the second machine coordinates are normalized, and obtains corresponding 9 third machines and sits
Mark, normalization also refer to, and the second machine coordinates are projected in the plane that rotation angle is 0, acquired third machine is sat
It is designated as the coordinate that rotation angle is 0.According toObtain the
Three machine coordinates, wherein i is the serial number of designated position, Qi(xi,yi) it is i-th of third machine coordinates, xiAnd yiRespectively i-th
The abscissa and ordinate of a third machine coordinates, vxi、vyiAnd vriThe abscissa of respectively i-th second machine coordinates is indulged
Coordinate and rotational coordinates, x0And y0The abscissa and ordinate of respectively the first machine coordinates.
Perspective transformation matrix is calculated according to matched second machine coordinates and calibration pixel coordinate.It is sat according to the second machine
After mark obtains third machine coordinates, established according to the third machine coordinates of the same designated position and corresponding calibration pixel coordinate
Perspective transform transformation model so that Qi=Pi* A, wherein i is the serial number of designated position, QiFor i-th of third machine
Coordinate, PiFor i-th of calibration pixel coordinate, A is perspective transformation matrix.By being determined in perspective transform transformation modelFor the transformation relation of third machine coordinates and calibration pixel coordinate, wherein i
For the serial number of designated position, xiAnd yiThe abscissa and ordinate of respectively i-th third machine coordinates, ziMeet zi=a13ui+
a23wi+a33, uiAnd wiThe abscissa and ordinate of respectively i-th calibration pixel coordinate,For perspective transform square
Battle array, a11、a12、a13、a21、a22、a23、a31、a32And a33The respectively element of perspective transformation matrix.It is obtained according to transformation relationWithWherein,With
For perspective transform equation.According to perspective transform equation andIt obtains so that ε values are minimum
Perspective transformation matrix.
The pixel coordinate of the target point of camera fields of view is converted to target machine according to perspective transformation matrix
Device coordinate completes the hand and eye calibrating of robot in order to realize that robot is moved to target point.According toObtain robot mobile control coordinate with
The mapping relations of position coordinates in camera fields of view, wherein x and y be respectively the mobile control coordinate of robot abscissa and
Ordinate, u and w are respectively the abscissa and ordinate of the position coordinates in camera fields of view, xt、ytAnd rtRespectively initial machine
Abscissa, ordinate and the rotational coordinates of coordinate, initial machine coordinate are the movement when obtaining camera fields of view where target point
Control coordinate.According toObtain target machine
Device coordinate, wherein xfAnd yfThe respectively abscissa and ordinate of target machine coordinate, utAnd wtThe respectively pixel of target point
The abscissa and ordinate of coordinate.
Above-mentioned Robotic Hand-Eye Calibration method proposes the second machine coordinates and the calibration of the characteristic point under 9 designated positions
Pixel coordinate can calculate accurate and optimal perspective transformation matrix, and the pixel by target point according to mapping relations may be implemented
Coordinate is converted into target machine coordinate, completes the hand and eye calibrating of robot, can improve robot and reach crawl object institute in place
The precision set.The perspective transformation matrix by being solved after perspective transform equation is calculated simultaneously, can calculate ring flange
Angle between plane where place plane and camera fields of view calculates the focal plane of camera and robot ring flange after installing
Error between the plane of place.Wherein, a13The angle between X-axis and U axis, a can be embodied23It can embody between Y-axis and W axis
Angle.
It should be understood that although each step in the flow chart of Fig. 2 to 6 is shown successively according to the instruction of arrow,
Be these steps it is not that the inevitable sequence indicated according to arrow executes successively.Unless expressly stating otherwise herein, these steps
There is no stringent sequences to limit for rapid execution, these steps can execute in other order.Moreover, in Fig. 2 to 6 at least
A part of step may include that either these sub-steps of multiple stages or stage are not necessarily in same a period of time to multiple sub-steps
Quarter executes completion, but can execute at different times, the execution in these sub-steps or stage be sequentially also not necessarily according to
Secondary progress, but can either the sub-step of other steps or at least part in stage in turn or replace with other steps
Ground executes.
In one embodiment, as shown in fig. 7, the structure that Fig. 7 is Robotic Hand-Eye Calibration system in one embodiment is shown
It is intended to, provides a kind of Robotic Hand-Eye Calibration system, the end of robot is equipped with camera and standard smelting tool, fixed calibration
Characteristic point is provided on object, system includes coordinate obtaining module 310, perspective transformation matrix acquisition module 320 and coordinate transform
Module 330, wherein:
Coordinate obtaining module 310, for obtaining the first machine coordinates, the second machine coordinates and calibration pixel coordinate, wherein
First machine coordinates are to move control coordinate, standard along reference direction alignment characteristics Dian Shi robots in the end of calibration smelting tool
By the normal direction for the ring flange installed in robot, the second machine coordinates are to be located in camera fields of view to refer in characteristic point in direction
The mobile control coordinate of Zhi Shi robots is positioned, calibration pixel coordinate is when characteristic point random device people is moved to camera fields of view middle finger
Position coordinates of the characteristic point in camera fields of view when positioning is set;
Perspective transformation matrix acquisition module 320, for being sat according to the first machine coordinates, the second machine coordinates and calibration pixel
Mark obtains perspective transformation matrix;
Coordinate transformation module 330, the mobile control coordinate for obtaining robot according to perspective transformation matrix are regarded with camera
The pixel coordinate of target point is converted into target machine coordinate by the mapping relations of the position coordinates of Yezhong according to mapping relations,
In, the pixel coordinate of target point includes the position coordinates of target point in the camera, and target machine coordinate is that robot is moved to mesh
The corresponding mobile control coordinate of punctuate.
Above-mentioned Robotic Hand-Eye Calibration system, the mobile control coordinate by obtaining robot and the position in camera fields of view
The mapping relations of coordinate may be implemented that the pixel coordinate of target point is converted into target machine coordinate according to mapping relations, complete
The hand and eye calibrating of robot can improve the precision that robot reaches crawl object position.
In one embodiment, coordinate transformation module 330 is additionally operable to obtain initial machine coordinate and the pixel of target point is sat
Mark, wherein initial machine coordinate is the mobile control coordinate when obtaining camera fields of view where target point;It is sat according to initial machine
Mark, the pixel coordinate of target point and mapping relations obtain target machine coordinate.
Above-mentioned Robotic Hand-Eye Calibration system is obtained according to initial machine coordinate, the pixel coordinate of target point and mapping relations
Target machine coordinate is taken, the hand and eye calibrating of robot is completed, the precision that robot reaches crawl object position can be improved.
In one embodiment, perspective transformation matrix acquisition module 320 is additionally operable to according to the first machine coordinates and the second machine
Device coordinate obtains third machine coordinates, wherein third machine coordinates are that the second machine coordinates are projected in where the first machine coordinates
The coordinate of plane;Perspective transformation matrix is obtained according to third machine coordinates and calibration pixel coordinate.
Above-mentioned Robotic Hand-Eye Calibration system obtains third machine according to the first machine coordinates and the second machine coordinates and sits
Mark obtains perspective transformation matrix according to third machine coordinates and calibration pixel coordinate, can reduce the complexity and fortune of calculating
Calculation amount, and improve precision.
In one embodiment, perspective transformation matrix acquisition module 320 is additionally operable to according to Qi=Pi* A obtains perspective transform square
Battle array, wherein i is the serial number of designated position, QiFor i-th of third machine coordinates, PiFor i-th of calibration pixel coordinate, A is perspective
Transformation matrix.
Above-mentioned Robotic Hand-Eye Calibration system according to third machine coordinates and the determination of calibration pixel coordinate and obtains perspective change
Matrix is changed, the complexity and operand of calculating can be reduced, and improves precision.
In one embodiment, perspective transformation matrix acquisition module 320 is additionally operable to basisObtain third machine coordinates, wherein i is designated position
Serial number, Qi(xi,yi) it is i-th of third machine coordinates, xiAnd yiThe abscissa of respectively i-th third machine coordinates and vertical seat
Mark, vxi、vyiAnd vriAbscissa, ordinate and the rotational coordinates of respectively i-th second machine coordinates, x0And y0Respectively
The abscissa and ordinate of one machine coordinates.
Above-mentioned Robotic Hand-Eye Calibration system, x0And y0The abscissa and ordinate of respectively the first machine coordinates, the first machine
The rotational coordinates of device coordinate is 0, and the first machine coordinates for being 0 according to rotational coordinates calculate third machine coordinates, can be obtained same
The third machine coordinates that sample rotational coordinates is 0, can reduce the complexity and operand of calculating, and improve precision.
In one embodiment, perspective transformation matrix acquisition module 320 is additionally operable to according to third machine coordinates and calibration picture
Plain coordinate establishes perspective transform equation;Perspective transformation matrix is solved according to perspective transform equation.
Above-mentioned Robotic Hand-Eye Calibration system, the perspective transform etc. established by third machine coordinates and calibration pixel coordinate
Formula can accurately solve perspective transformation matrix, in order to subsequently improve the precision that robot reaches crawl object position.
In one embodiment, perspective transformation matrix acquisition module 320 is additionally operable to determineFor the transformation relation of third machine coordinates and calibration pixel coordinate, wherein i
For the serial number of designated position, xiAnd yiThe abscissa and ordinate of respectively i-th third machine coordinates, ziMeet zi=a13ui+
a23wi+a33, uiAnd wiThe abscissa and ordinate of respectively i-th calibration pixel coordinate,For perspective transform square
Battle array, a11、a12、a13、a21、a22、a23、a31、a32And a33The respectively element of perspective transformation matrix;It is obtained according to transformation relationWithWherein,With
For perspective transform equation, i is the serial number of designated position, xiAnd yiThe abscissa of respectively i-th third machine coordinates and vertical seat
Mark, uiAnd wiThe abscissa and ordinate of respectively i-th calibration pixel coordinate, a11、a12、a13、a21、a22、a23、a31、a32With
a33The respectively element of perspective transformation matrix.
Above-mentioned Robotic Hand-Eye Calibration system, by determining third machine coordinates and demarcating the transformation relation of pixel coordinate,
And perspective transform equation is obtained, the element in perspective transformation matrix can be accurately solved, perspective transformation matrix is obtained.
In one embodiment, coordinate transformation module 330 is additionally operable to basisObtain robot mobile control coordinate with
The mapping relations of position coordinates in camera fields of view, wherein x and y be respectively the mobile control coordinate of robot abscissa and
Ordinate, u and w are respectively the abscissa and ordinate of the position coordinates in camera fields of view, a11、a12、a13、a21、a22、a23、a31、
a32And a33The respectively element of perspective transformation matrix, xt、ytAnd rtThe respectively abscissa of initial machine coordinate, ordinate and rotation
Turn coordinate, initial machine coordinate is the mobile control coordinate when obtaining camera fields of view where target point.
Above-mentioned Robotic Hand-Eye Calibration system can accurately obtain machine after the perspective transformation matrix after solution
The mapping relations of the mobile control coordinate of people and the position coordinates in camera fields of view, in order to subsequently obtain target machine coordinate,
The hand and eye calibrating of robot is completed, and improves the precision that robot reaches crawl object position.
In one embodiment, coordinate transformation module 330 is additionally operable to obtain initial machine coordinate and the pixel of target point is sat
Mark;According toTarget machine is obtained to sit
Mark, wherein xfAnd yfThe respectively abscissa and ordinate of target machine coordinate, utAnd wtThe respectively pixel coordinate of target point
Abscissa and ordinate, a11、a12、a13、a21、a22、a23、a31、a32And a33The respectively element of perspective transformation matrix, xt、yt
And rtThe respectively abscissa of initial machine coordinate, ordinate and rotational coordinates.
Above-mentioned Robotic Hand-Eye Calibration system is obtained according to initial machine coordinate, the pixel coordinate of target point and mapping relations
Target machine coordinate is taken, the hand and eye calibrating of robot is completed, the precision that robot reaches crawl object position can be improved.
In one embodiment, perspective transformation matrix acquisition module 320 is additionally operable to basisObtain the perspective transformation matrix so that ε value minimums, wherein i is designated position
Serial number, xiAnd yiThe abscissa and ordinate of respectively i-th third machine coordinates, uiAnd wiRespectively i-th calibration pixel is sat
Target abscissa and ordinate, a11、a12、a13、a21、a22、a23、a31、a32And a33The respectively element of perspective transformation matrix.
Above-mentioned Robotic Hand-Eye Calibration system, by acquisition so that the perspective transformation matrix of ε value minimums, can obtain optimal
Change and closest to actual perspective transformation matrix, further decrease error, improve the accuracy of perspective transformation matrix, in order to can
To improve the precision that robot reaches crawl object position.
Specific restriction about Robotic Hand-Eye Calibration system may refer to above for Robotic Hand-Eye Calibration method
Restriction, details are not described herein.Modules in above-mentioned Robotic Hand-Eye Calibration system can be fully or partially through software, hard
Part and combinations thereof is realized.Above-mentioned each module can be embedded in or in the form of hardware independently of in the processor in computer equipment,
It can also in a software form be stored in the memory in computer equipment, the above modules are executed in order to which processor calls
Corresponding operation.
In one embodiment, a kind of computer equipment is provided, which can be server, internal junction
Composition can be as shown in Figure 8.The computer equipment include the processor connected by system bus, memory, network interface and
Database.Wherein, the processor of the computer equipment is for providing calculating and control ability.The memory packet of the computer equipment
Include non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system, computer program and data
Library.The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The calculating
The database of machine equipment is for storing data.The network interface of the computer equipment is used to pass through network connection with external terminal
Communication.To realize a kind of Robotic Hand-Eye Calibration method when the computer program is executed by processor.
It will be understood by those skilled in the art that structure shown in Fig. 8, is only tied with the relevant part of application scheme
The block diagram of structure does not constitute the restriction for the computer equipment being applied thereon to application scheme, specific computer equipment
May include either combining certain components than more or fewer components as shown in the figure or being arranged with different components.
In one embodiment, a kind of computer equipment is provided, including memory, processor and storage are on a memory
And the computer program that can be run on a processor, processor realize following steps when executing computer program:
Obtain the first machine coordinates, the second machine coordinates and calibration pixel coordinate, wherein the first machine coordinates are to calibrate
Control coordinate is moved in the end of smelting tool along reference direction alignment characteristics Dian Shi robots, and reference direction in robot by installing
Ring flange normal direction, the second machine coordinates be when characteristic point is located at designated position in camera fields of view robot movement
Control coordinate, calibration pixel coordinate be when characteristic point random device people is moved to designated position in camera fields of view characteristic point in camera
Position coordinates in the visual field;
According to the first machine coordinates, the second machine coordinates and calibration pixel coordinate, perspective transformation matrix is obtained;
The mapping of the mobile control coordinate and the position coordinates in camera fields of view of robot is obtained according to perspective transformation matrix
The pixel coordinate of target point is converted into target machine coordinate, wherein the pixel coordinate packet of target point by relationship according to mapping relations
The position coordinates of target point in the camera are included, target machine coordinate is that robot is moved to the corresponding mobile control seat of target point
Mark.
In one embodiment, following steps are also realized when processor executes computer program:
Obtain the pixel coordinate of initial machine coordinate and target point, wherein initial machine coordinate is in acquisition target point institute
Mobile control coordinate in camera fields of view;According to initial machine coordinate, the pixel coordinate of target point and mapping relations, mesh is obtained
Mark machine coordinates.
In one embodiment, following steps are also realized when processor executes computer program:
Third machine coordinates are obtained according to the first machine coordinates and the second machine coordinates, wherein third machine coordinates are the
The coordinate of plane where two machine coordinates are projected in the first machine coordinates;It is obtained according to third machine coordinates and calibration pixel coordinate
Perspective transformation matrix.
In one embodiment, following steps are also realized when processor executes computer program:
According to Qi=Pi* A obtains perspective transformation matrix, wherein i is the serial number of designated position, QiFor i-th of third machine
Coordinate, PiFor i-th of calibration pixel coordinate, A is perspective transformation matrix.
In one embodiment, following steps are also realized when processor executes computer program:
According toThird machine coordinates are obtained,
In, i is the serial number of designated position, Qi(xi,yi) it is i-th of third machine coordinates, xiAnd yiRespectively i-th of third machine coordinates
Abscissa and ordinate, vxi、vyiAnd vriAbscissa, ordinate and the rotational coordinates of respectively i-th second machine coordinates,
x0And y0The abscissa and ordinate of respectively the first machine coordinates.
In one embodiment, following steps are also realized when processor executes computer program:
Perspective transform equation is established according to third machine coordinates and calibration pixel coordinate;It is solved according to perspective transform equation saturating
Depending on transformation matrix.
In one embodiment, following steps are also realized when processor executes computer program:
It determinesFor the transformation of third machine coordinates and calibration pixel coordinate
Relationship, wherein i is the serial number of designated position, xiAnd yiThe abscissa and ordinate of respectively i-th third machine coordinates, ziIt is full
Sufficient zi=a13ui+a23wi+a33, uiAnd wiThe abscissa and ordinate of respectively i-th calibration pixel coordinate,For
Perspective transformation matrix, a11、a12、a13、a21、a22、a23、a31、a32And a33The respectively element of perspective transformation matrix;
It is obtained according to transformation relationWithWherein,WithFor perspective transform equation, i is the serial number of designated position, xiAnd yi
The abscissa and ordinate of respectively i-th third machine coordinates, uiAnd wiThe abscissa of respectively i-th calibration pixel coordinate
And ordinate, a11、a12、a13、a21、a22、a23、a31、a32And a33The respectively element of perspective transformation matrix.
In one embodiment, following steps are also realized when processor executes computer program:
Obtain the pixel coordinate of initial machine coordinate and target point;According toObtain target machine coordinate, wherein
xfAnd yfThe respectively abscissa and ordinate of target machine coordinate, utAnd wtThe respectively abscissa of the pixel coordinate of target point
And ordinate, a11、a12、a13、a21、a22、a23、a31、a32And a33The respectively element of perspective transformation matrix, xt、ytAnd rtRespectively
For the abscissa, ordinate and rotational coordinates of initial machine coordinate.
In one embodiment, following steps are also realized when processor executes computer program:
According toObtain the perspective transformation matrix so that ε value minimums, wherein i
For the serial number of designated position, xiAnd yiThe abscissa and ordinate of respectively i-th third machine coordinates, uiAnd wiRespectively i-th
The abscissa and ordinate of a calibration pixel coordinate, a11、a12、a13、a21、a22、a23、a31、a32And a33Respectively perspective transform square
The element of battle array.
In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, is calculated
Machine program realizes following steps when being executed by processor:
Obtain the first machine coordinates, the second machine coordinates and calibration pixel coordinate, wherein the first machine coordinates are to calibrate
Control coordinate is moved in the end of smelting tool along reference direction alignment characteristics Dian Shi robots, and reference direction in robot by installing
Ring flange normal direction, the second machine coordinates be when characteristic point is located at designated position in camera fields of view robot movement
Control coordinate, calibration pixel coordinate be when characteristic point random device people is moved to designated position in camera fields of view characteristic point in camera
Position coordinates in the visual field;
According to the first machine coordinates, the second machine coordinates and calibration pixel coordinate, perspective transformation matrix is obtained;
The mapping of the mobile control coordinate and the position coordinates in camera fields of view of robot is obtained according to perspective transformation matrix
The pixel coordinate of target point is converted into target machine coordinate, wherein the pixel coordinate packet of target point by relationship according to mapping relations
The position coordinates of target point in the camera are included, target machine coordinate is that robot is moved to the corresponding mobile control seat of target point
Mark.
In one embodiment, following steps are also realized when computer program is executed by processor:
Obtain the pixel coordinate of initial machine coordinate and target point, wherein initial machine coordinate is in acquisition target point institute
Mobile control coordinate in camera fields of view;According to initial machine coordinate, the pixel coordinate of target point and mapping relations, mesh is obtained
Mark machine coordinates.
In one embodiment, following steps are also realized when computer program is executed by processor:
Third machine coordinates are obtained according to the first machine coordinates and the second machine coordinates, wherein third machine coordinates are the
The coordinate of plane where two machine coordinates are projected in the first machine coordinates;It is obtained according to third machine coordinates and calibration pixel coordinate
Perspective transformation matrix.
In one embodiment, following steps are also realized when computer program is executed by processor:
According to Qi=Pi* A obtains perspective transformation matrix, wherein i is the serial number of designated position, QiFor i-th of third machine
Coordinate, PiFor i-th of calibration pixel coordinate, A is perspective transformation matrix.
In one embodiment, following steps are also realized when computer program is executed by processor:
According toThird machine coordinates are obtained,
In, i is the serial number of designated position, Qi(xi,yi) it is i-th of third machine coordinates, xiAnd yiRespectively i-th of third machine coordinates
Abscissa and ordinate, vxi、vyiAnd vriAbscissa, ordinate and the rotational coordinates of respectively i-th second machine coordinates,
x0And y0The abscissa and ordinate of respectively the first machine coordinates.
In one embodiment, following steps are also realized when computer program is executed by processor:
Perspective transform equation is established according to third machine coordinates and calibration pixel coordinate;It is solved according to perspective transform equation saturating
Depending on transformation matrix.
In one embodiment, following steps are also realized when computer program is executed by processor:
It determinesFor the transformation of third machine coordinates and calibration pixel coordinate
Relationship, wherein i is the serial number of designated position, xiAnd yiThe abscissa and ordinate of respectively i-th third machine coordinates, ziIt is full
Sufficient zi=a13ui+a23wi+a33, uiAnd wiThe abscissa and ordinate of respectively i-th calibration pixel coordinate,For
Perspective transformation matrix, a11、a12、a13、a21、a22、a23、a31、a32And a33The respectively element of perspective transformation matrix;
It is obtained according to transformation relationWithWherein,WithFor perspective transform equation, i is the serial number of designated position, xiAnd yi
The abscissa and ordinate of respectively i-th third machine coordinates, uiAnd wiThe abscissa of respectively i-th calibration pixel coordinate
And ordinate, a11、a12、a13、a21、a22、a23、a31、a32And a33The respectively element of perspective transformation matrix.
In one embodiment, following steps are also realized when computer program is executed by processor:
Obtain the pixel coordinate of initial machine coordinate and target point;According toObtain target machine coordinate, wherein
xfAnd yfThe respectively abscissa and ordinate of target machine coordinate, utAnd wtThe respectively abscissa of the pixel coordinate of target point
And ordinate, a11、a12、a13、a21、a22、a23、a31、a32And a33The respectively element of perspective transformation matrix, xt、ytAnd rtRespectively
For the abscissa, ordinate and rotational coordinates of initial machine coordinate.
In one embodiment, following steps are also realized when computer program is executed by processor:
According toObtain the perspective transformation matrix so that ε value minimums, wherein i
For the serial number of designated position, xiAnd yiThe abscissa and ordinate of respectively i-th third machine coordinates, uiAnd wiRespectively i-th
The abscissa and ordinate of a calibration pixel coordinate, a11、a12、a13、a21、a22、a23、a31、a32And a33Respectively perspective transform square
The element of battle array.
One of ordinary skill in the art will appreciate that realizing all or part of flow in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, the computer program can be stored in a non-volatile computer
In read/write memory medium, the computer program is when being executed, it may include such as the flow of the embodiment of above-mentioned each method.Wherein,
Any reference to memory, storage, database or other media used in each embodiment provided herein,
Including non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM
(PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include
Random access memory (RAM) or external cache.
Each technical characteristic of above example can be combined arbitrarily, to keep description succinct, not to above-described embodiment
In each technical characteristic it is all possible combination be all described, as long as however, the combination of these technical characteristics be not present lance
Shield is all considered to be the range of this specification record.
The several embodiments of the application above described embodiment only expresses, the description thereof is more specific and detailed, but simultaneously
It cannot therefore be construed as limiting the scope of the patent.It should be pointed out that coming for those of ordinary skill in the art
It says, under the premise of not departing from the application design, various modifications and improvements can be made, these belong to the protection of the application
Range.Therefore, the protection domain of the application patent should be determined by the appended claims.