CN107363813A - A kind of desktop industrial robot teaching system and method based on wearable device - Google Patents
A kind of desktop industrial robot teaching system and method based on wearable device Download PDFInfo
- Publication number
- CN107363813A CN107363813A CN201710707674.9A CN201710707674A CN107363813A CN 107363813 A CN107363813 A CN 107363813A CN 201710707674 A CN201710707674 A CN 201710707674A CN 107363813 A CN107363813 A CN 107363813A
- Authority
- CN
- China
- Prior art keywords
- arm
- module
- data
- angle
- msup
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/0081—Programme-controlled manipulators with master teach-in means
Landscapes
- Engineering & Computer Science (AREA)
- Robotics (AREA)
- Mechanical Engineering (AREA)
- Manipulator (AREA)
- Numerical Control (AREA)
Abstract
The invention discloses a kind of desktop industrial robot teaching system and method based on wearable device, the teaching system includes training data fetching portion, data processing section and industrial robot.The teaching system is by obtaining the Eulerian angles on human arm at each sensing station, calculate the rotational angle of the position and each joint that obtain arm end, after carrying out coded representation and extensive output to above-mentioned discrete exercise data, the rotational angle of continuous arm end movement track and corresponding joint is obtained.The present invention carries out Inverse Kinematics Solution according to terminal position to robot, multigroup against solution so as to obtain, and assesses above-mentioned multigroup inverse solution according to the rotational angle of corresponding joint, chooses one group of joint angles against solution as robot closest to arm posture.Teaching system provided by the invention carries out teaching without teaching box or driving machinery arm, it is thus only necessary to the teaching to industrial robot is completed according to normal manual operation can, so as to improve teaching efficiency.
Description
Technical field
The present invention relates to technical field of automation, more particularly to a kind of desktop industrial robot based on wearable device to show
Teaching system and method.
Background technology
The flexibility and its reasonability of mechanism design that various sixdegree-of-freedom simulations are moved due to it, current automatic
Change and be widely used in production system.The main task of robot be that instead of the mankind carry out some repeatability, environment it is poor, endanger
Dangerous high manual operations, its premise completed are to send instruction to robot in advance, it is specified that the action that robot should complete
With the particular content of operation, this process is exactly the teaching to robot.Teaching playback is the programming side that robot generally uses
Formula, this needs operating personnel and adjusted repeatedly with teaching box the job parameter of robot at each taught point.Whole teaching process
After end, robot according to record data operation repeatedly.Teach-by-doing teaching is also a kind of form of teaching playback system, operation
Member carries out operation track demonstration by the joystick manipulated installed in robot end to robot, and work opportunity device people passes through
The data of storage, run according to the track of previous teaching.Existing teaching playback system to the requirement of the operative skill of operator compared with
Height, teaching process are cumbersome, time-consuming inefficient.In order to improve teaching speed, most of mechanical arm control system all provides
The teaching box of simple operation, but this mode still can not fundamentally lift the teaching efficiency of sixdegree-of-freedom simulation.
The content of the invention
To solve the above problems, the present invention provide a kind of desktop industrial robot teaching system based on wearable device and
Method, at least partly solves above-mentioned technical problem.
Therefore, the present invention provides a kind of desktop industrial robot teaching system based on wearable device, including teaching number
According to fetching portion, data processing section and industrial robot, the data processing section respectively with the training data acquisition unit
Divide and connected with the industrial robot；
The training data fetching portion includes wearable device, and the wearable device includes data control node, number
According to sending module and 6 sensor nodes, 6 sensor nodes are averagely arranged on left arm and right arm respectively, 3 of every arm
Sensor node is separately positioned on the remote shoulder joint end of upper arm, the remote elbow joint end of forearm and arm end, the data control node
It is arranged on human body back；
The sensor node is used to obtain movable information of the arm joint among motion process；
The data control node is used for the movable information for converging each sensor node, and is estimated according to default posture
Calculating method carries out fusion treatment to the movable information, to obtain Euler's angular data of sensor correspondence position；
The data transmission blocks are used to the Eulerian angles data are transferred at the data by wireless transmission method
Manage part；
The data processing section include data reception module, arm motion information computational module, data encoding study with
Extensive output module, the inverse solution module of robot kinematics and optimal solution evaluation module, the arm motion information computational module point
Be not connected with the data reception module and data encoding study with extensive output module, the data encoding learn with it is general
Change output module to be connected with the inverse solution module of the robot kinematics and the optimal solution evaluation module respectively, the optimal solution is commented
Estimate module to be connected with the inverse solution module of the robot kinematics；
The data reception module is used to receive Euler's angular data, and the Eulerian angles data is transferred to described
Arm motion information computational module；
The arm motion information computational module is used to calculate arm terminal position, arm end according to the Eulerian angles
Posture and arm joint rotational angle；
Data encoding study and extensive output module be used for discrete arm terminal position, arm terminal angle and
Arm joint rotational angle carries out coded representation and extensive output, to obtain continuous arm terminal position, arm terminal angle
With arm joint rotational angle；
The inverse solution module of the robot kinematics is used to carry out machine to continuous arm terminal position and arm terminal angle
Device people's Inverse Kinematics Solution, with multigroup inverse solution corresponding to acquisition；
The optimal solution evaluation module be used for according to minimum evaluation factor selected among multigroup inverse solution one group it is inverse
The articulation angle as robot is solved, the evaluation factor is according to multigroup inverse solution and continuous arm joint angle of rotation
Degree is formed；
The industrial robot includes industrial machine human body and bottom control module, the industrial machine human body and institute
State the connection of bottom control module；
The bottom control module is used to carry out the industrial machine human body according to the articulation angle of robot
Control, to cause the industrial robot to reproduce the arm motion of demonstrator.
Optionally, in addition to information fusion module, described information Fusion Module calculate with the arm motion information respectively
Module and data encoding study are connected with extensive output module；
Described information Fusion Module is used for the arm terminal position, the arm terminal angle and the arm joint
Rotational angle carries out fusion treatment, to form default vector data collection；
Data encoding study and extensive output module are used for discrete vector data collection progress coded representation and general
Change output, to obtain continuous arm terminal position, arm terminal angle and arm joint rotational angle.
Optionally, the vector data collection is vectorial D, D={ x, y, z, αW, βW, γ W, θ1, θ2, θ3, θ4, θ5, θ6, wherein
X, y and z is position of the arm end in world coordinate system, αW、βWAnd γWFor posture of the arm end in world coordinate system,
θ1、θ2、θ3、θ4、θ5And θ6For the corresponding rotational angle of arm joint.
Optionally, the calculation formula of the evaluation factor is as follows：
Wherein, λi(i=1,2,6) and it is position of each arm joint to the mechanical arm of the industrial robot
With the factor of influence of posture；
Ni={ θR1, θR2, θR3, θR4, θR5, θR6(i=1,2,3) be same arm terminal position under i groups it is inverse
Solution；
DP={ θ '1, θ '2, θ '3, θ '4, θ '5, θ '6Be synchronization arm joint rotational angle.
Optionally, the Eulerian angles data include the Eulerian angles Ψ at the remote shoulder joint endB, the remote elbow joint end of forearm
Eulerian angles ΨF, the Eulerian angles Ψ of the arm endH, ΨB={ αB, βB, γB}T, ΨF={ αF, βF, γF}T, ΨH={ αH,
βH, γH}T, wherein αB、βBAnd γBRespectively Eulerian angles ΨBIn nutational angle, angle of precession and angle of rotation, αF、βFAnd γFRespectively
Eulerian angles ΨFIn nutational angle, angle of precession and angle of rotation, αH、βHAnd γHRespectively Eulerian angles ΨHIn nutational angle, angle of precession and
Angle of rotation.
The present invention also provides a kind of desktop industrial robot teaching method based on wearable device, and the teaching method is adopted
With the desktop industrial robot teaching system based on wearable device, the teaching system includes training data fetching portion, number
According to process part and industrial robot, the data processing section respectively with the training data fetching portion and the industrial machine
Device people connects；
The training data fetching portion includes wearable device, and the wearable device includes data control node, number
According to sending module and 6 sensor nodes, 6 sensor nodes are averagely arranged on left arm and right arm respectively, 3 of every arm
Sensor node is separately positioned on the remote shoulder joint end of upper arm, the remote elbow joint end of forearm and arm end, the data control node
It is arranged on human body back；
The data processing section include data reception module, arm motion information computational module, data encoding study with
Extensive output module, the inverse solution module of robot kinematics and optimal solution evaluation module, the arm motion information computational module point
Be not connected with the data reception module and data encoding study with extensive output module, the data encoding learn with it is general
Change output module to be connected with the inverse solution module of the robot kinematics and the optimal solution evaluation module respectively, the optimal solution is commented
Estimate module to be connected with the inverse solution module of the robot kinematics；
The industrial robot includes industrial machine human body and bottom control module, the industrial machine human body and institute
State the connection of bottom control module；
The teaching method includes：
The sensor node obtains movable information of the arm joint among motion process；
The data control node converges the movable information of each sensor node, and is calculated according to default Attitude estimation
Method carries out fusion treatment to the movable information, to obtain Euler's angular data of sensor correspondence position；
The Eulerian angles data are transferred to the data processing division by the data transmission blocks by wireless transmission method
Point；
The data reception module receives Euler's angular data, and the Eulerian angles data are transferred into the arm
Movable information computing module；
The arm motion information computational module calculates arm terminal position, arm terminal angle according to the Eulerian angles
With arm joint rotational angle；
The data encoding study is with extensive output module to discrete arm terminal position, arm terminal angle and arm
Articulation angle carries out coded representation and extensive output, to obtain continuous arm terminal position, arm terminal angle and hand
Shoulder joint rotational angle；
The inverse solution module of the robot kinematics carries out robot to continuous arm terminal position and arm terminal angle
Inverse Kinematics Solution, with multigroup inverse solution corresponding to acquisition；
The optimal solution evaluation module selects one group of inverse solution to make according to the evaluation factor of minimum among multigroup inverse solution
For the articulation angle of robot, the evaluation factor is according to multigroup inverse solution and continuous arm joint rotational angle shape
Into；
The bottom control module is controlled according to the articulation angle of robot to the industrial machine human body,
To cause the industrial robot to reproduce the arm motion of demonstrator.
Optionally, the teaching system also includes information fusion module, described information Fusion Module respectively with the arm
Movable information computing module and data encoding study are connected with extensive output module；
The data encoding study is with extensive output module to discrete arm terminal position, arm terminal angle and arm
Articulation angle include before the step of coded representation and extensive output：
Described information Fusion Module rotates the arm terminal position, the arm terminal angle and the arm joint
Angle carries out fusion treatment, to form default vector data collection；
The data encoding study is with extensive output module to discrete arm terminal position, arm terminal angle and arm
The step of progress coded representation of articulation angle and extensive output, includes：
Data encoding study and extensive output module are to discrete vector data collection progress coded representation and extensive defeated
Go out, to obtain continuous arm terminal position, arm terminal angle and arm joint rotational angle.
Optionally, the data encoding study and extensive output module are to discrete arm terminal position, arm end appearance
The step of state and arm joint rotational angle carry out coded representation and extensive output includes：
The data encoding study is with extensive output module using gauss hybrid models to discrete arm terminal position, hand
Arm terminal angle and arm joint rotational angle carry out coded representation, to realize discrete arm terminal position, arm end appearance
The representative learning of state and arm joint rotational angle；
The data encoding study is with extensive output module using Gaussian Mixture regression model to the arm end after coding
End position, arm terminal angle and arm joint rotational angle carry out data reconstruction and extensive output, to obtain continuous arm
Terminal position, arm terminal angle and arm joint rotational angle.
The present invention has following beneficial effects：
Among desktop industrial robot teaching system and method provided by the invention based on wearable device, the teaching
System includes training data fetching portion, data processing section and industrial robot.The teaching system is by obtaining teaching
Eulerian angles among journey on human arm at each sensing station, calculate turn in the position for obtaining arm end and each joint
Dynamic angle, above-mentioned discrete exercise data is carried out obtain after coded representation and extensive output continuous end movement track and
The rotational angle of corresponding joint.The present invention carries out Inverse Kinematics Solution according to terminal position to robot, so as to obtain multigroup inverse solution,
Above-mentioned multigroup inverse solution is assessed according to the rotational angle of corresponding joint, the one group of inverse solution chosen closest to arm posture is used as robot
Joint angles.Teaching system provided by the invention realizes teaching personnel and carries out teaching without teaching box or driving machinery arm,
Need only to according to teaching of the normal manual operation can completion to industrial robot, so as to improve teaching efficiency.Institute
State that the teaching process of teaching system is simple and convenient, to teaching personnel without higher operative skill requirement, need to only dress wearable
Equipment.The teaching system causes the posture among robot kinematics close to the posture of human arm motion, from
And the motion process of human arm can be farthest reduced, motion smoothing is stable, and kinematic accuracy is controllable, and teaching effect is good
It is good.
Brief description of the drawings
Fig. 1 is the structure for the desktop industrial robot teaching system based on wearable device that the embodiment of the present invention one provides
Schematic diagram；
Fig. 2 is the information for the desktop industrial robot teaching system based on wearable device that the embodiment of the present invention one provides
Transmit schematic diagram；
Fig. 3 is the schematic diagram that the human arm that the embodiment of the present invention one provides is reduced to three connecting rod seven freedom models；
Fig. 4 is the flow for the desktop industrial robot teaching method based on wearable device that the embodiment of the present invention two provides
Schematic diagram.
Embodiment
To make those skilled in the art more fully understand technical scheme, the present invention is carried below in conjunction with the accompanying drawings
The desktop industrial robot teaching system and method based on wearable device supplied are described in detail.
Embodiment one
Fig. 1 is the structure for the desktop industrial robot teaching system based on wearable device that the embodiment of the present invention one provides
Schematic diagram, Fig. 2 are the information for the desktop industrial robot teaching system based on wearable device that the embodiment of the present invention one provides
Transmit schematic diagram.As depicted in figs. 1 and 2, the desktop industrial robot teaching system based on wearable device that the present embodiment provides
System includes training data fetching portion, data processing section and industrial robot, and the data processing section shows with described respectively
Religion data acquiring portion connects with the industrial robot.The teaching system is by obtaining among teaching process on human arm
Eulerian angles at each sensing station, calculate the rotational angle of the position and each joint that obtain arm end, to it is above-mentioned from
Scattered exercise data carries out coded representation and extensive output obtains the rotation of continuous end movement track and corresponding joint afterwards
Angle.The present embodiment carries out Inverse Kinematics Solution according to terminal position to robot, so as to obtain multigroup inverse solution, according to corresponding joint
Rotational angle assess above-mentioned multigroup inverse solution, choose joint angles of the one group of inverse solution closest to arm posture as robot.
The training data fetching portion that the present embodiment provides includes wearable device, and the wearable device includes data control
Node, data transmission blocks and 6 sensor nodes processed, 6 sensor nodes are averagely arranged on left arm and right arm respectively, every
3 sensor nodes of arm are separately positioned on the remote shoulder joint end of upper arm, the remote elbow joint end of forearm and arm end, the data
Control node is arranged on human body back.The sensor node is used to obtain motion letter of the arm joint among motion process
Breath, the data control node is used for the movable information for converging each sensor node, and is calculated according to default Attitude estimation
Method carries out fusion treatment to the movable information, and to obtain Euler's angular data of sensor correspondence position, the data send mould
Block is used to the Eulerian angles data are transferred into the data processing section by wireless transmission method.What the present embodiment provided shows
Teaching system realizes teaching personnel and carries out teaching without teaching box or driving machinery arm, it is thus only necessary to according to normal manual operation
Can completes the teaching to industrial robot, so as to improve teaching efficiency.The teaching for the teaching system that the present embodiment provides
Process is simple and convenient, to teaching personnel without higher operative skill requirement, only need to dress wearable device.The present embodiment
The teaching system of offer causes the posture among robot kinematics close to the posture of human arm motion, so as to most
The motion process of human arm is reduced to big degree, motion smoothing is stable, and kinematic accuracy is controllable, and teaching works well.
The data control node that the present embodiment provides is convergence control node.Wearable device based on inertial sensor includes
6 sensor nodes and 1 convergence control node.Sensor node gathers movable information of the arm joint in motion process,
The sensor node mainly includes three axis accelerometer, three-axis gyroscope and three axle magnetometers, is respectively used to gather arm pass
3-axis acceleration (a of sectionx, ay, az), three axle angular velocity of rotation (gx, gy, gz) and three-axle magnetic field intensity (mx, my, mz).Convergence control
Node processed converges the data of each sensor node, and above-mentioned data are handled, then is sent to by wireless transmission method
To data processing section.
6 sensor nodes that the present embodiment provides are distributed in left and right two-arm, each 3 of left and right two-arm.3 of every arm
Sensor node is respectively distributed to the remote shoulder joint end of upper arm, the remote elbow joint end of forearm and the centre of the palm.The centre of the palm that the present embodiment provides is
For arm end.Each sensor node is made up of 1 three axis accelerometer, 1 three-axis gyroscope and 1 three axle magnetometer.
Three axis accelerometer is used for gathering the 3-axis acceleration (a of each jointx, ay, az), three-axis gyroscope is used for gathering each pass
Three axle angular velocity of rotation (g at sectionx, gy, gz), three axle magnetometers are used for gathering the three-axle magnetic field intensity (m of each jointx, my,
mz).Convergence control node is distributed in human body back, for collecting the acceleration in each joint, angular speed and magnetometer data, and
And fusion treatment is carried out to above-mentioned data.Gyro sensor measures carrier along X, Y and the angular speed in Z coordinate system direction respectively,
Integrated in conjunction with the sampling time, calculate the posture of sensor, be indicated by the form of Eulerian angles.The present embodiment knot
Resultant acceleration sensor and magnetometric sensor are modified and Combined Calculation to attitude information, are then based on Kalman filtering
Attitude estimation algorithm carries out fusion calculation to sensing data, then data are packed by radioing to data processing section.
The data processing section that the present embodiment provides includes data reception module, arm motion information computational module, data
Coding study and extensive output module, the inverse solution module of robot kinematics and optimal solution evaluation module, the arm motion information
Computing module is connected with the data reception module and data encoding study with extensive output module respectively, and the data are compiled
Code study is connected with the inverse solution module of the robot kinematics and the optimal solution evaluation module respectively with extensive output module, institute
Optimal solution evaluation module is stated to be connected with the inverse solution module of the robot kinematics.The data reception module is used to receive the Europe
Angular data is drawn, and the Eulerian angles data are transferred to the arm motion information computational module, the arm motion information
Computing module is used to calculate arm terminal position, arm terminal angle and arm joint rotational angle according to the Eulerian angles,
The data encoding study is used for discrete arm terminal position, arm terminal angle and arm joint with extensive output module
Rotational angle carries out coded representation and extensive output, is closed with obtaining continuous arm terminal position, arm terminal angle and arm
Rotational angle is saved, the inverse solution module of the robot kinematics is used to carry out continuous arm terminal position and arm terminal angle
The inverse solution of robot kinematics, with multigroup inverse solution corresponding to acquisition, the optimal solution evaluation module be used for according to the assessment of minimum because
Son selects articulation angle of one group of inverse solution as robot among multigroup inverse solution, and the evaluation factor is according to
Multigroup inverse solution and continuous arm joint rotational angle are formed.The teaching system that the present embodiment provides realizes teaching personnel need not
Teaching box or driving machinery arm carry out teaching, it is thus only necessary to are completed according to normal manual operation can to industrial robot
Teaching, so as to improve teaching efficiency.The teaching process for the teaching system that the present embodiment provides is simple and convenient, and to teaching, personnel do not have
There is higher operative skill requirement, only need to dress wearable device.The teaching system that the present embodiment provides causes robot
Posture among motion process close to human arm motion posture, so as to farthest reduce the fortune of human arm
Dynamic process, motion smoothing is stable, and kinematic accuracy is controllable, and teaching works well.
In the present embodiment, data reception module receives the training data from perspective data acquisition module, then by teaching number
According to passing to arm joint rotational angle computing module.The calculating content of arm motion information computational module is divided into two parts：Hand
Arm terminal position and arm joint angle.Among teaching process, the pose of arm end is important training data, for showing
Teach the end effector of industrial robot.In the present embodiment, arm end is the centre of the palm.The correct teaching of end effector pose
It is the important prerequisite that industrial robot completes operation task.The characteristic of the training data gathered according to wearable device, arm
End pose calculating sub module is respectively calculated to the position of arm and posture.The data that inertial sensor is gathered are inertia
Relative to the Eulerian angles of world coordinate system, the data that are gathered of wearable device that the present embodiment provides are for sensor position
Relative to the Eulerian angles of world coordinate system at elbow joint, wrist joint and the centre of the palm.Therefore, the pose of arm end can be by slapping
The data of inertial sensor at the heart are drawn.Position at the centre of the palm needs to combine used at elbow joint, wrist joint and the centre of the palm
Property sensing data calculate obtain.
The present embodiment establishes following coordinate system：Lifted before teaching personnel both arms are horizontal, the centre of the palm is upward.It is former by coordinate of shoulder joint
Point establishes world coordinate system WCS, and along arm forward, vertically upward, Y-axis is determined Z axis X-axis by the right-hand rule, world coordinate system WCS
Not with arm motion.The present embodiment establishes coordinate system S, coordinate system E and coordinate system respectively in shoulder joint, elbow joint and wrist joint
W：Coordinate system S is X-axis forward using shoulder joint as the origin of coordinates, along orientation arm, and Z axis determines Y vertically upward, using right-hand rule
Axle, coordinate system S are connected in large arm, moved with large arm, i.e., coordinate system S and world coordinate system WCS is overlapped in initial position；Sit
Mark system E is X-axis forward using elbow joint as the origin of coordinates, along orientation arm, and Z axis determines Y-axis vertically upward, using right-hand rule,
Coordinate system E is connected on forearm, with arm movements；Coordinate system W is X-axis forward using wrist joint as the origin of coordinates, along volar direction,
Z axis determines Y-axis vertically upward, using right-hand rule, and coordinate system W is connected on palm, moved with palm.Therefore, elbow joint
The data that the inertial sensor of (at the remote shoulder joint of large arm) is gathered are Eulerian angles of the coordinate system S-phase for world coordinate system WCS
ΨS={ αS, βS, γS}T；The data that the inertial sensor of wrist joint (the remote elbow joint of forearm) is gathered are that coordinate system E is relative
In world coordinate system WCS Eulerian angles ΨE={ αE, βE, γE}T；The data that inertial sensor at the centre of the palm is gathered are coordinate systems
W relative to world coordinate system WCS Eulerian angles ΨW={ αW, βW, γW}T.The big arm lengths of human arm are r1, forearm lengths are
r2, computational length r of the sensor to carpal length as palm at palm3。
The present embodiment regards arm as one group of head and the tail in world coordinate system WCS when calculating the position of arm end
Connected vector：WhereinTo represent the vector of large arm,To represent the vector of forearm,To represent hand
The vector of the palm.Therefore, the position of arm end is 3 vectorial and vector coordinates.The calculating of directed quantity understands that one group first
The vectorial and vector that tail is connected can be equivalent to the vectorial and vectorial of one group of common starting point, and 3 vectorial and vector
Coordinate can be obtained by the projection of 3 vectors respectively in the X, Y, Z direction.I.e.
VectorCoordinate in coordinate system S isSP1=[r1,0,0]T；
VectorCoordinate in coordinate system E isEP2=[r2,0,0]T；
VectorCoordinate in coordinate system W iswP3=[r3,0,0]T。
In the present embodiment, Eulerian angles Ψ of the coordinate system S-phase for world coordinate system WCSS={ αS, βS, γS}T, coordinate system E
Relative to world coordinate system WCS Eulerian angles ΨE={ αE, βE, γE}T, coordinate system W relative to world coordinate system WCS Eulerian angles
ΨW={ αW, βW, γW}T.Its spin matrix R can be obtained by the Eulerian angles conversion between two coordinate systems, rotational order according to
Z-X-Z：
The present embodiment is by ΨS={ αS, βS, γS}T、ΨE={ αE, βE, γE}TAnd ΨW={ αW, βW, γW}TSubstitute into above-mentioned
Formula can obtain：
Spin matrix of the coordinate system S-phase for world coordinate system WCSCoordinate system E is relative to world coordinate system WCS's
Spin matrixCoordinate system W relative to world coordinate system WCS spin matrixTherefore, arm end is in world coordinates
It is that location matrix in WCS is：
In the present embodiment, arm terminal angle is centre of the palm posture, because sensor is fixed on palm, so sensor
There is identical rotation status with the centre of the palm, i.e., the Eulerian angles calculated by sensing data can be used for representing the posture of palm.
Therefore, the attitude matrix in the centre of the palm isIn order to facilitate the processing of subsequent module, still down transmitted in the form of Eulerian angles, i.e.,
ΨW={ αW, βW, γW}T。
Fig. 3 is the schematic diagram that the human arm that the embodiment of the present invention one provides is reduced to three connecting rod seven freedom models.Such as
Shown in Fig. 3, the shoulder joint of human arm, elbow joint, wrist joint have 3,2,2 frees degree respectively, according to human arm
Movement characteristic and rigid body assume theory, and human arm can be reduced to three connecting rod seven freedom models.Arm joint angle of rotation
The training data that computing module receives data reception module transmission is spent, calculates the rotational angle in 7 joints of arm.This implementation
The exercise data that example is obtained by wearable device, the Eulerian angles Ψ of large arm, forearm and palm is measured respectivelyS, ΨE, ΨW, ΨS
={ αS, βS, γS}T, ΨE={ αE, βE, γE}T, ΨW={ αW, βW, γW}T, wherein αB, βB, γBRespectively in large arm Eulerian angles
Nutational angle, angle of precession and angle of rotation；αF, βF, γFNutational angle, angle of precession and angle of rotation respectively in forearm Eulerian angles；αH,
βH, γHNutational angle, angle of precession and angle of rotation respectively in palm Eulerian angles.To calculate the joint angle of human arm, this implementation
Example is as follows using the nonsingular mapping algorithm of joint space based on quaternary number, calculation procedure：
(1) anglec of rotation of shoulder joint is：θ1=βS, θ2=αS, θ3=γS
(2) by calculating forearm relative to the Eulerian angles and palm of large arm relative to the Eulerian angles of forearm, arm is obtained
Remaining 4 joint angle.The present embodiment is converted to the Eulerian angles of measurement the quaternary number of the large arm of human arm, forearm and wrist
QS, QE, QW, to avoid singularity.Eulerian angles ΨrTo quaternary number QrConversion formula be：
Wherein, q1r, q2r, q3r, q4rFor quaternary number Qr4 parameters；αr, βr, γrRespectively Eulerian angles ΨrIn nutating
Angle, angle of precession and angle of rotation.
(3) hypothesis forearm is Q relative to the quaternary number of large armES, palm is Q relative to the quaternary number of forearmWE, then QE=
QESQS, QW=QWEQE, i.e., it is Q with respect to quaternary numberES=QEQS - 1, QWE=QWQE - 1
(4) will be with respect to quaternary number QESAnd QWEBe converted to Relative Euler angle ΨES={ αES, βES, γES}T, ΨWE={ αWE,
βWE, γWE}T, wherein ΨESEulerian angles for forearm relative to large arm, αES, βES, γESRespectively its corresponding nutational angle, precession
Angle and angle of rotation；ΨWEEulerian angles for palm relative to forearm, αWE, βWE, γWERespectively its corresponding nutational angle, angle of precession
And angle of rotation.Quaternary number QSTo Eulerian angles ΨSConversion formula be：
(5) anglec of rotation of elbow joint is θ4=αES, θ5=γES
(6) the carpal anglec of rotation is θ6=αWE, θ7=βWE
Finally, be calculated 7 joint angles are passed to information fusion module by the present embodiment.
In the present embodiment, arm terminal position calculating sub module and arm joint angle calculation submodule pass through calculating respectively
Obtain the rotational angle of arm terminal position, arm terminal angle and each joint of arm.Above-mentioned two data are same group of sensings
Device data are obtained by different computational methods, and they are the arm motion data of synchronization.Meanwhile human arm has 7
The individual free degree, but the industrial robot for moving reproduction only has 6 frees degree, and 6 are only needed when assessing optimal solution certainly
By spending, it is therefore desirable to give up 1 free degree of human arm.In view of the influence degree to human arm posture, give up wrist
Last free degree at place, i.e. θ7.Give up θ7The attitude accuracy of industrial robot can't be influenceed, because θ7Only in optimal solution
Evaluation stage can just use.Meanwhile in order to facilitate the processing of subsequent module, the present embodiment it is one group of number by this two groups of data fusions
According to being represented with vectorial D：
D={ x, y, z, αW, βW, γW, θ1, θ2, θ3, θ4, θ5, θ6}
Wherein x, y, z and αW, βW, γWPosition and posture of the arm end in world coordinate system WCS, θ are represented respectively1,
θ2, θ3, θ4, θ5, θ6For the corresponding joint angles of arm.All movable informations among whole teaching process are in the form of data set
Subsequent module is passed to, is designated as data set { D }.
In the present embodiment, training data obtains under certain time interval, and the data of acquisition are discrete.At it
During motion reproduction afterwards, gathered used by the inverse solution module of robot kinematics during pose not necessarily teaching
Point.Therefore, it is necessary to generate continuous arm motion information using discrete arm motion information.The present embodiment will above gained
Movable information, i.e., vectorial D at different moments is utilized respectively GMM gauss hybrid models and encoded, and realizes representative learning, then lead to
Cross GMR Gaussian Mixtures regression model and carry out data reconstruction and extensive output, to obtain continuous arm terminal position and each pass
The rotational angle of section.Any time is provided, the present embodiment can obtain corresponding arm motion data, be represented with D ＇.
Assuming that i-th of data is di={ Di, Ti, i=1,2,3, and N }, wherein N is an arm motion process
Among the data times that are gathered of training data acquisition module, DiThe arm motion information obtained for ith, Ti is time value.
If each data point di obeys following probability distribution：
Wherein, p (k) is prior probability, and p (di | k) is conditional probability distribution, Gaussian distributed.Therefore, whole teaching
Data acquisition system can represent that K is the Gaussian Profile number for forming gauss hybrid models with gauss hybrid models.Prior probability and
The formula of conditional probability distribution is as follows：
P (k)=πk (10)
Wherein, D is the dimension for the GMM for encoding training data.Therefore, gauss hybrid models it needs to be determined that parameter be { πk,
μk, ∑k, prior probability, expectation and the variance of k-th of composition are represented respectively.Using EM algorithm, i.e. EM algorithms are estimated
GMM parameter, parameter learning is carried out by the maximal possibility estimation that parameter is found in probabilistic model.The time value of training data
TiAs query point, the Di ' of corresponding movable information is estimated using GMR.Known p (di | k) meet Gaussian Profile, i.e. { Dk,
Tk}~N (μk, ∑k), wherein,
μk={ μD, k, μT, k} (12)
In given TkUnder conditions of, DkConditional probability also meet Gaussian Profile, i.e. Dk|Tk~N (μ 'D, k, ∑ 'D, k).Its
In
μ’D, k=μD, k+∑DT, k(∑T, k)-1(Tk-μT, k) (14)
∑’D, k=∑D, k-∑DT, k(∑T, k)-1∑TD, k (15)
As shown from the above formula, the mean μ of the gauss hybrid models of K gauss component 'DWith variance ∑ 'D, wherein
By Dk|Tk~N (μ 'D, k, ∑ 'D, k) distribution estimation conditional expectation E (D | T), i.e. μ 'DFor reconstruction attractor value corresponding to T,
Therefore extensive data point is d '={ D ', T }, and the point is not included among the arm motion information of aforementioned modules offer, still
All substantive characteristics of arm motion are encapsulated, in covariance constraints ∑ 'DUnder can generate continuous arm motion information.
According to requirement of the motion replication technology to industrial robot motion, according to corresponding time interval, T is usediTo look into
Ask point inquiry arm motion information D '.The subsequent module that the present embodiment provides is to the arm end among arm movable information vector D '
End pose and arm joint corner have different purposes.The accurate calculating of arm terminal position is to ensure teaching system with transmission
An important ring for accuracy, and angle of rolling over is the important assessment foundation that optimal solution is assessed, and is to ensure industrial machinery arm appearance
State is approximately basic with human arm posture.Therefore, data characterization study will be with arm end pose phase with extensive output module
Data x ', y ', z ', the α ' of passW、β’W、γ’WThe inverse solution module of robot kinematics is passed to, will be related to arm joint corner
Data θ '1、θ’2、θ’3、θ’4、θ’5, θ '6Pass to optimal solution evaluation module.The teaching system that the present embodiment provides causes machine
Posture among people's motion process close to human arm motion posture, so as to farthest reduce human arm
Motion process, motion smoothing is stable, and kinematic accuracy is controllable, and teaching works well.
The inverse solution module of robot kinematics that the present embodiment provides obtains data characterization study and transmitted with extensive output module
Arm end pose data x ', y ', z ', α ' W, β 'W、γ’WAfterwards, by Eulerian angles α 'W, β 'W, γ 'WIt is converted into by preceding method
Spin matrixThe position auto―control among robot learning is formed in conjunction with arm terminal position x, y, zWherein P '=and [x ', y ', z '] '.With the inverse solution of robot kinematics, each pass of corresponding robot is tried to achieve
The rotational angle of section, as a result with vectorial NiRepresent, Ni={ θR1, θR2, θR3, θR4, θR5, θR6, wherein i is the number of inverse solution.
For same terminal position, the inverse solution module of robot kinematics will obtain multigroup solution.Selected among multigroup solution
Optimal solution, conventional way are since large joint, and choosing the possibility solution closest with the value of the joint angle last moment is
Unique solution.This method ensure that the amount of exercise of industrial robot is minimum, and motion is more steady, but does not ensure that industrial machine
The similitude of posture and the posture of teaching arm among people's motion process.
In the present embodiment, Ni={ θR1, θR2, θR3, θR4, θR5, θR6, (i=1,2,3) be same terminal position under
The inverse solution of i groups, DP={ θ '1, θ '2, θ '3, θ '4, θ '5, θ '6Be synchronization arm joint angle, evaluation factor is：
Wherein λi(i=1,2,6) and it is factor of influence of each joint to mechanical arm pose.In human arm
Among motion, the influence of the swing of arm to arm position is far longer than the twisting of arm, and the twisting of arm is only to arm
Posture has considerable influence.Similarly, during industrial robot motion, the rotation in different joints influences on the position of mechanical arm
Effect is made a world of difference.The rotating effect in some joints is equal to the swing of human arm.For specific mechanical arm, according to specific
Architectural feature, analyze influence of the different joints to mechanical arm pose, be analogous to human arm, assign mechanical arm difference joint phase
The factor of influence answered, pose influence of the joint on mechanical arm is bigger, and factor of influence is bigger.
The numerical value of the evaluation factor of each group solution is bigger, illustrates the position of the mechanical arm pose and human arm corresponding to the group solution
Appearance difference is bigger.On the contrary, the numerical value of evaluation factor is smaller, illustrate the position of the mechanical arm pose and human arm corresponding to the group solution
Appearance is more close.Therefore, optimal solution evaluation module selects one group of minimum solution of evaluation factor in multigroup inverse solution so that robot transports
Posture among dynamic process close to human arm motion posture, so as to farthest reduce the motion of human arm
Process, motion smoothing is stable, and kinematic accuracy is controllable, and teaching works well.
The industrial robot that the present embodiment provides includes industrial machine human body and bottom control module, the industrial machine
Human body connects with the bottom control module, and the bottom control module is used for the articulation angle according to robot to institute
State industrial machine human body to be controlled, to cause the industrial robot to reproduce the arm motion of demonstrator.The present embodiment carries
Executable portion of the industrial machine human body and its bottom control module of confession as the motion reappearance of teaching system, bottom control mould
Block receives one group of solution that optimal solution evaluation module is chosen, and robot body is entered using this group solution as the rotational angle in each joint
Row control so that robot body reproduces the arm motion process of demonstrator, and teaching process is simple and convenient, and personnel do not have to teaching
Higher operative skill requirement, only need to dress wearable device.
The desktop industrial robot teaching system based on wearable device that the present embodiment provides obtains including training data
Partly, data processing section and industrial robot.The teaching system is each on human arm by obtaining among teaching process
Eulerian angles at sensing station, the rotational angle of the position and each joint that obtain arm end is calculated, to above-mentioned discrete
Exercise data carries out coded representation and extensive output obtains the rotational angle of continuous end movement track and corresponding joint afterwards.
The present embodiment carries out Inverse Kinematics Solution according to terminal position to robot, so as to obtain multigroup inverse solution, according to turning for corresponding joint
The dynamic above-mentioned multigroup inverse solution of angle estimator, chooses one group of inverse joint angles solved as robot closest to arm posture.This reality
Applying the teaching system of example offer realizes teaching personnel without teaching box or the progress teaching of driving machinery arm, it is thus only necessary to according to just
Normal manual operation can completes the teaching to industrial robot, so as to improve teaching efficiency.What the present embodiment provided shows
The teaching process of teaching system is simple and convenient, to teaching personnel without higher operative skill requirement, only need to dress wearable device
.The teaching system that the present embodiment provides causes the posture among robot kinematics close to the appearance of human arm motion
State, so as to farthest reduce the motion process of human arm, motion smoothing is stable, and kinematic accuracy is controllable, teaching effect
Fruit is good.
Embodiment two
The present embodiment provides a kind of desktop industrial robot teaching method based on wearable device, and the teaching method is adopted
With the desktop industrial robot teaching system based on wearable device, the teaching system includes training data fetching portion, number
According to process part and industrial robot, the data processing section respectively with the training data fetching portion and the industrial machine
Device people connects, and the training data fetching portion includes wearable device, and the wearable device includes data control node, number
According to sending module and 6 sensor nodes, 6 sensor nodes are averagely arranged on left arm and right arm respectively, 3 of every arm
Sensor node is separately positioned on the remote shoulder joint end of upper arm, the remote elbow joint end of forearm and arm end, the data control node
Human body back is arranged on, the data processing section includes data reception module, arm motion information computational module, data encoding
Study calculates with extensive output module, the inverse solution module of robot kinematics and optimal solution evaluation module, the arm motion information
Module is connected with the data reception module and data encoding study with extensive output module respectively, the data encoding
Practise with extensive output module respectively with the robot kinematics are inverse solves module and the optimal solution evaluation module is connected, it is described most
Excellent solution evaluation module and the robot kinematics inverse solution module be connected, the industrial robot including industrial machine human body with
Bottom control module, the industrial machine human body connect with the bottom control module.The teaching system that the present embodiment provides
Realize teaching personnel and carry out teaching without teaching box or driving machinery arm, it is thus only necessary to according to normal manual operation can
The teaching to industrial robot is completed, so as to improve teaching efficiency.The teaching process letter for the teaching system that the present embodiment provides
Folk prescription just, to teaching personnel without higher operative skill requirement, only need to dress wearable device.
Fig. 4 is the flow for the desktop industrial robot teaching method based on wearable device that the embodiment of the present invention two provides
Schematic diagram.As shown in figure 4, the teaching method includes：The sensor node obtains arm joint among motion process
Movable information；The data control node converges the movable information of each sensor node, and according to default Attitude estimation
Algorithm carries out fusion treatment to the movable information, to obtain Euler's angular data of sensor correspondence position；The data are sent
The Eulerian angles data are transferred to the data processing section by module by wireless transmission method；The data reception module connects
Euler's angular data is received, and the Eulerian angles data are transferred to the arm motion information computational module；The arm
Movable information computing module calculates arm terminal position, arm terminal angle and arm joint angle of rotation according to the Eulerian angles
Degree；The data encoding study is with extensive output module to discrete arm terminal position, arm terminal angle and arm joint
Rotational angle carries out coded representation and extensive output, is closed with obtaining continuous arm terminal position, arm terminal angle and arm
Save rotational angle；The inverse solution module of the robot kinematics carries out machine to continuous arm terminal position and arm terminal angle
People's Inverse Kinematics Solution, with multigroup inverse solution corresponding to acquisition；The optimal solution evaluation module is according to the evaluation factor of minimum from described
Articulation angle of one group of inverse solution as robot is selected among multigroup inverse solution, the evaluation factor is according to multigroup inverse solution
Formed with continuous arm joint rotational angle；The bottom control module is according to the articulation angle of robot to the work
Industry robot body is controlled, to cause the industrial robot to reproduce the arm motion of demonstrator.On teaching method
Specific implementation process, specific descriptions of the embodiment one for the teaching process of teaching system are referred to, here is omitted.
Among the desktop industrial robot teaching method based on wearable device that the present embodiment provides, the teaching system
Including training data fetching portion, data processing section and industrial robot.The teaching system by obtain teaching process it
Eulerian angles on middle human arm at each sensing station, calculate the angle of rotation of the position and each joint that obtain arm end
Degree, carry out obtaining after coded representation and extensive output continuous end movement track and corresponding to above-mentioned discrete exercise data
The rotational angle in joint.The present embodiment carries out Inverse Kinematics Solution according to terminal position to robot, so as to obtain multigroup inverse solution, root
Above-mentioned multigroup inverse solution is assessed according to the rotational angle of corresponding joint, chooses one group of inverse solution closest to arm posture as robot
Joint angles.The teaching system that the present embodiment provides realizes teaching personnel and carries out teaching without teaching box or driving machinery arm,
Need only to according to teaching of the normal manual operation can completion to industrial robot, so as to improve teaching efficiency.This
The teaching process for the teaching system that embodiment provides is simple and convenient, to teaching personnel without higher operative skill requirement, only needs
Dress wearable device.The teaching system that the present embodiment provides causes posture among robot kinematics close to people
The posture of body arm motion, so as to farthest reduce the motion process of human arm, motion smoothing is stable, motion essence
Spend controllable, teaching works well.
It is understood that the principle that embodiment of above is intended to be merely illustrative of the present and the exemplary implementation that uses
Mode, but the invention is not limited in this.For those skilled in the art, the essence of the present invention is not being departed from
In the case of refreshing and essence, various changes and modifications can be made therein, and these variations and modifications are also considered as protection scope of the present invention.
Claims (8)
1. a kind of desktop industrial robot teaching system based on wearable device, it is characterised in that obtained including training data
Partly, data processing section and industrial robot, the data processing section respectively with the training data fetching portion and institute
State industrial robot connection；
The training data fetching portion includes wearable device, and the wearable device includes data control node, data are sent out
Module and 6 sensor nodes are sent, 6 sensor nodes are averagely arranged on left arm and right arm, 3 sensings of every arm respectively
Device node is separately positioned on the remote shoulder joint end of upper arm, the remote elbow joint end of forearm and arm end, and the data control node is set
In human body back；
The sensor node is used to obtain movable information of the arm joint among motion process；
The data control node is used for the movable information for converging each sensor node, and is calculated according to default Attitude estimation
Method carries out fusion treatment to the movable information, to obtain Euler's angular data of sensor correspondence position；
The data transmission blocks are used to the Eulerian angles data are transferred into the data processing division by wireless transmission method
Point；
The data processing section include data reception module, arm motion information computational module, data encoding study with it is extensive
Output module, the inverse solution module of robot kinematics and optimal solution evaluation module, the arm motion information computational module respectively with
The data reception module and data encoding study are connected with extensive output module, the data encoding learn with it is extensive defeated
Go out module to be connected with the inverse solution module of the robot kinematics and the optimal solution evaluation module respectively, the optimal solution assesses mould
Block is connected with the inverse solution module of the robot kinematics；
The data reception module is used to receive Euler's angular data, and the Eulerian angles data are transferred into the arm
Movable information computing module；
The arm motion information computational module is used to calculate arm terminal position, arm terminal angle according to the Eulerian angles
With arm joint rotational angle；
The data encoding study is used for discrete arm terminal position, arm terminal angle and arm with extensive output module
Articulation angle carries out coded representation and extensive output, to obtain continuous arm terminal position, arm terminal angle and hand
Shoulder joint rotational angle；
The inverse solution module of the robot kinematics is used to carry out robot to continuous arm terminal position and arm terminal angle
Inverse Kinematics Solution, with multigroup inverse solution corresponding to acquisition；
The optimal solution evaluation module is used to select one group of inverse solution to make among multigroup inverse solution according to the evaluation factor of minimum
For the articulation angle of robot, the evaluation factor is according to multigroup inverse solution and continuous arm joint rotational angle shape
Into；
The industrial robot includes industrial machine human body and bottom control module, the industrial machine human body and the bottom
Layer control module connection；
The bottom control module is used to be controlled the industrial machine human body according to the articulation angle of robot,
To cause the industrial robot to reproduce the arm motion of demonstrator.
2. the desktop industrial robot teaching system according to claim 1 based on wearable device, it is characterised in that also
Including information fusion module, described information Fusion Module respectively with the arm motion information computational module and the data encoding
Study is connected with extensive output module；
Described information Fusion Module is used to rotate the arm terminal position, the arm terminal angle and the arm joint
Angle carries out fusion treatment, to form default vector data collection；
Data encoding study and extensive output module are used for discrete vector data collection progress coded representation and extensive defeated
Go out, to obtain continuous arm terminal position, arm terminal angle and arm joint rotational angle.
3. the desktop industrial robot teaching system according to claim 2 based on wearable device, it is characterised in that institute
It is vectorial D, D={ x, y, z, α to state vector data collectionW, βW, γW, θ1, θ2, θ3, θ4, θ5, θ6, wherein x, y and z are arm end
Position in world coordinate system, αW、βWAnd γWFor posture of the arm end in world coordinate system, θ1、θ2、θ3、θ4、θ5And θ6
For the corresponding rotational angle of arm joint.
4. the desktop industrial robot teaching system according to claim 1 based on wearable device, it is characterised in that institute
The calculation formula for stating evaluation factor is as follows：
<mrow>
<mi>A</mi>
<mi>F</mi>
<mo>=</mo>
<msqrt>
<mrow>
<msup>
<msub>
<mi>&lambda;</mi>
<mn>1</mn>
</msub>
<mn>2</mn>
</msup>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>&theta;</mi>
<mrow>
<mi>R</mi>
<mn>1</mn>
</mrow>
</msub>
<mo>-</mo>
<msub>
<msup>
<mi>&theta;</mi>
<mo>,</mo>
</msup>
<mn>1</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>+</mo>
<msup>
<msub>
<mi>&lambda;</mi>
<mn>2</mn>
</msub>
<mn>2</mn>
</msup>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>&theta;</mi>
<mrow>
<mi>R</mi>
<mn>2</mn>
</mrow>
</msub>
<mo>-</mo>
<msub>
<msup>
<mi>&theta;</mi>
<mo>,</mo>
</msup>
<mn>2</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>+</mo>
<msup>
<msub>
<mi>&lambda;</mi>
<mn>3</mn>
</msub>
<mn>2</mn>
</msup>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>&theta;</mi>
<mrow>
<mi>R</mi>
<mn>3</mn>
</mrow>
</msub>
<mo>-</mo>
<msub>
<msup>
<mi>&theta;</mi>
<mo>,</mo>
</msup>
<mn>3</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>+</mo>
<msup>
<msub>
<mi>&lambda;</mi>
<mn>4</mn>
</msub>
<mn>2</mn>
</msup>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>&theta;</mi>
<mrow>
<mi>R</mi>
<mn>4</mn>
</mrow>
</msub>
<mo>-</mo>
<msub>
<msup>
<mi>&theta;</mi>
<mo>,</mo>
</msup>
<mn>4</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>+</mo>
<msup>
<msub>
<mi>&lambda;</mi>
<mn>5</mn>
</msub>
<mn>2</mn>
</msup>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>&theta;</mi>
<mrow>
<mi>R</mi>
<mn>5</mn>
</mrow>
</msub>
<mo>-</mo>
<msub>
<msup>
<mi>&theta;</mi>
<mo>,</mo>
</msup>
<mn>5</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>+</mo>
<msup>
<msub>
<mi>&lambda;</mi>
<mn>6</mn>
</msub>
<mn>2</mn>
</msup>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>&theta;</mi>
<mrow>
<mi>R</mi>
<mn>6</mn>
</mrow>
</msub>
<mo>-</mo>
<msub>
<msup>
<mi>&theta;</mi>
<mo>,</mo>
</msup>
<mn>6</mn>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
</mrow>
</msqrt>
</mrow>
Wherein, λi(i=1,2 ..., 6) is each arm joint to the position of the mechanical arm of the industrial robot and the shadow of posture
Ring the factor；
Ni={ θR1, θR2, θR3, θR4, θR5, θR6(i=1,2,3 ...)
For the inverse solution of i groups under same arm terminal position；
DP={ θ '1, θ '2, θ '3, θ '4, θ '5, θ '6Be synchronization arm joint rotational angle.
5. the desktop industrial robot teaching system according to claim 1 based on wearable device, it is characterised in that institute
Stating Eulerian angles data includes the Eulerian angles Ψ at the remote shoulder joint endB, the Eulerian angles Ψ at the remote elbow joint end of forearmF, the hand
The Eulerian angles Ψ of arm endH, ΨB={ αB, βB, γB}T, ΨF={ αF, βF, γF}T, ΨH={ αH, βH, γH}T, wherein αB、βBWith
γBRespectively Eulerian angles ΨBIn nutational angle, angle of precession and angle of rotation, αF、βFAnd γFRespectively Eulerian angles ΨFIn nutating
Angle, angle of precession and angle of rotation, αH、βHAnd γHRespectively Eulerian angles ΨHIn nutational angle, angle of precession and angle of rotation.
A kind of 6. desktop industrial robot teaching method based on wearable device, it is characterised in that the teaching method uses
Desktop industrial robot teaching system based on wearable device, the teaching system include training data fetching portion, data
Process part and industrial robot, the data processing section respectively with the training data fetching portion and the industrial machine
People connects；
The training data fetching portion includes wearable device, and the wearable device includes data control node, data are sent out
Module and 6 sensor nodes are sent, 6 sensor nodes are averagely arranged on left arm and right arm, 3 sensings of every arm respectively
Device node is separately positioned on the remote shoulder joint end of upper arm, the remote elbow joint end of forearm and arm end, and the data control node is set
In human body back；
The data processing section include data reception module, arm motion information computational module, data encoding study with it is extensive
Output module, the inverse solution module of robot kinematics and optimal solution evaluation module, the arm motion information computational module respectively with
The data reception module and data encoding study are connected with extensive output module, the data encoding learn with it is extensive defeated
Go out module to be connected with the inverse solution module of the robot kinematics and the optimal solution evaluation module respectively, the optimal solution assesses mould
Block is connected with the inverse solution module of the robot kinematics；
The industrial robot includes industrial machine human body and bottom control module, the industrial machine human body and the bottom
Layer control module connection；
The teaching method includes：
The sensor node obtains movable information of the arm joint among motion process；
The data control node converges the movable information of each sensor node, and according to default Attitude estimation algorithm pair
The movable information carries out fusion treatment, to obtain Euler's angular data of sensor correspondence position；
The Eulerian angles data are transferred to the data processing section by the data transmission blocks by wireless transmission method；
The data reception module receives Euler's angular data, and the Eulerian angles data are transferred into the arm motion
Information computational module；
The arm motion information computational module calculates arm terminal position, arm terminal angle and hand according to the Eulerian angles
Shoulder joint rotational angle；
The data encoding study is with extensive output module to discrete arm terminal position, arm terminal angle and arm joint
Rotational angle carries out coded representation and extensive output, is closed with obtaining continuous arm terminal position, arm terminal angle and arm
Save rotational angle；
The inverse solution module of the robot kinematics carries out robot motion to continuous arm terminal position and arm terminal angle
Inverse solution is learned, with multigroup inverse solution corresponding to acquisition；
The optimal solution evaluation module selects one group of inverse solution to be used as machine according to the evaluation factor of minimum among multigroup inverse solution
The articulation angle of device people, the evaluation factor are formed according to multigroup inverse solution and continuous arm joint rotational angle；
The bottom control module is controlled according to the articulation angle of robot to the industrial machine human body, so that
Obtain the arm motion that the industrial robot reproduces demonstrator.
7. the desktop industrial robot teaching method according to claim 6 based on wearable device, it is characterised in that institute
Stating teaching system also includes information fusion module, described information Fusion Module respectively with the arm motion information computational module and
The data encoding study is connected with extensive output module；
The data encoding study is with extensive output module to discrete arm terminal position, arm terminal angle and arm joint
Rotational angle include before the step of coded representation and extensive output：
Described information Fusion Module is by the arm terminal position, the arm terminal angle and the arm joint rotational angle
Fusion treatment is carried out, to form default vector data collection；
The data encoding study is with extensive output module to discrete arm terminal position, arm terminal angle and arm joint
The step of rotational angle progress coded representation and extensive output, includes：
The data encoding study carries out coded representation and extensive output with extensive output module to discrete vector data collection, with
Obtain continuous arm terminal position, arm terminal angle and arm joint rotational angle.
8. the desktop industrial robot teaching method according to claim 6 based on wearable device, it is characterised in that institute
Data encoding study is stated with extensive output module to discrete arm terminal position, arm terminal angle and arm joint angle of rotation
The step of degree progress coded representation and extensive output, includes：
The data encoding study is with extensive output module using gauss hybrid models to discrete arm terminal position, arm end
Hold posture and arm joint rotational angle to carry out coded representation, with realize discrete arm terminal position, arm terminal angle and
The representative learning of arm joint rotational angle；
The data encoding study is with extensive output module using Gaussian Mixture regression model to the arm end position after coding
Put, arm terminal angle and arm joint rotational angle carry out data reconstruction and extensive output, to obtain continuous arm end
Position, arm terminal angle and arm joint rotational angle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710707674.9A CN107363813B (en) | 2017-08-17 | 2017-08-17 | Desktop industrial robot teaching system and method based on wearable equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710707674.9A CN107363813B (en) | 2017-08-17 | 2017-08-17 | Desktop industrial robot teaching system and method based on wearable equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107363813A true CN107363813A (en) | 2017-11-21 |
| CN107363813B CN107363813B (en) | 2020-08-04 |
Family
ID=60309162
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710707674.9A Active CN107363813B (en) | 2017-08-17 | 2017-08-17 | Desktop industrial robot teaching system and method based on wearable equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107363813B (en) |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108214495A (en) * | 2018-03-21 | 2018-06-29 | 北京无远弗届科技有限公司 | A kind of industrial robot teaching system and method |
| CN108427282A (en) * | 2018-03-30 | 2018-08-21 | 华中科技大学 | A kind of solution of Inverse Kinematics method based on learning from instruction |
| CN108500966A (en) * | 2018-03-30 | 2018-09-07 | 阜阳师范学院 | A kind of novel radio mechanical arm and control method |
| CN109732610A (en) * | 2019-03-01 | 2019-05-10 | 北京航空航天大学 | Man-machine collaboration robot grasping system and its working method |
| CN109799771A (en) * | 2018-12-29 | 2019-05-24 | 深圳市越疆科技有限公司 | A kind of control system of industrial robot, method and device |
| CN110815188A (en) * | 2019-11-19 | 2020-02-21 | 福州大学 | System and method for teaching industrial robot based on limb movement |
| CN110919638A (en) * | 2019-11-15 | 2020-03-27 | 华中科技大学 | 3+4 new-configuration double-arm cooperative robot machining system and method |
| CN110919639A (en) * | 2018-09-19 | 2020-03-27 | 华东至正工业自动化(常熟)有限公司 | Articulated teaching arm and teaching method based on same |
| CN111390941A (en) * | 2020-05-08 | 2020-07-10 | 中国船舶科学研究中心 | Command input device for underwater mechanical arm control and control method |
| CN111730601A (en) * | 2020-07-20 | 2020-10-02 | 季华实验室 | Wearable demonstrator demonstration control method and device and electronic equipment |
| CN112549025A (en) * | 2020-11-27 | 2021-03-26 | 北京工业大学 | Coordination diagram double-arm cooperative control method based on fusion of human body kinematic constraints |
| CN112659115A (en) * | 2019-10-15 | 2021-04-16 | 大族激光科技产业集团股份有限公司 | Industrial robot control method, device, equipment and storage medium |
| CN112706150A (en) * | 2020-12-17 | 2021-04-27 | 武汉大学 | Wearable external limb auxiliary grabbing device and control method |
| CN112959330A (en) * | 2021-02-02 | 2021-06-15 | 浙江大学 | Robot double-arm motion man-machine corresponding device and method based on master-slave dynamic motion elements |
| CN113180738A (en) * | 2021-05-25 | 2021-07-30 | 山东大学 | Virtual probe for remote cardiac ultrasound and diagnosis and treatment system |
| CN113386128A (en) * | 2021-05-11 | 2021-09-14 | 华南理工大学 | Body potential interaction method for multi-degree-of-freedom robot |
| CN113618710A (en) * | 2021-07-21 | 2021-11-09 | 慧灵科技（深圳）有限公司 | Dragging teaching method and device and dragging teaching robot |
| CN113618717A (en) * | 2021-09-07 | 2021-11-09 | 浙江理工大学 | Human upper limb skill action sensing learning device and method |
| CN114631687A (en) * | 2022-03-07 | 2022-06-17 | 深圳市瑞必达科技有限公司 | Control method for performing knocking protection and inclination protection on lifting table without calculating Euler angle |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090132088A1 (en) * | 2007-04-24 | 2009-05-21 | Tairob Ltd. | Transfer of knowledge from a human skilled worker to an expert machine - the learning process |
| CN102735260A (en) * | 2012-06-18 | 2012-10-17 | 航天东方红卫星有限公司 | Determination method of star sensor on-orbit measurement errors |
| CN103895022A (en) * | 2014-03-17 | 2014-07-02 | 东南大学 | Wearable type somatosensory control mechanical arm |
| CN104267815A (en) * | 2014-09-25 | 2015-01-07 | 黑龙江节点动画有限公司 | Motion capturing system and method based on inertia sensor technology |
| CN106064378A (en) * | 2016-06-07 | 2016-11-02 | 南方科技大学 | Control method and device for unmanned aerial vehicle mechanical arm |
-
2017
- 2017-08-17 CN CN201710707674.9A patent/CN107363813B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090132088A1 (en) * | 2007-04-24 | 2009-05-21 | Tairob Ltd. | Transfer of knowledge from a human skilled worker to an expert machine - the learning process |
| CN102735260A (en) * | 2012-06-18 | 2012-10-17 | 航天东方红卫星有限公司 | Determination method of star sensor on-orbit measurement errors |
| CN103895022A (en) * | 2014-03-17 | 2014-07-02 | 东南大学 | Wearable type somatosensory control mechanical arm |
| CN104267815A (en) * | 2014-09-25 | 2015-01-07 | 黑龙江节点动画有限公司 | Motion capturing system and method based on inertia sensor technology |
| CN106064378A (en) * | 2016-06-07 | 2016-11-02 | 南方科技大学 | Control method and device for unmanned aerial vehicle mechanical arm |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108214495A (en) * | 2018-03-21 | 2018-06-29 | 北京无远弗届科技有限公司 | A kind of industrial robot teaching system and method |
| CN108427282A (en) * | 2018-03-30 | 2018-08-21 | 华中科技大学 | A kind of solution of Inverse Kinematics method based on learning from instruction |
| CN108500966A (en) * | 2018-03-30 | 2018-09-07 | 阜阳师范学院 | A kind of novel radio mechanical arm and control method |
| CN110919639A (en) * | 2018-09-19 | 2020-03-27 | 华东至正工业自动化(常熟)有限公司 | Articulated teaching arm and teaching method based on same |
| CN109799771B (en) * | 2018-12-29 | 2021-01-05 | 深圳市越疆科技有限公司 | Control system, method and device of industrial robot |
| CN109799771A (en) * | 2018-12-29 | 2019-05-24 | 深圳市越疆科技有限公司 | A kind of control system of industrial robot, method and device |
| CN109732610A (en) * | 2019-03-01 | 2019-05-10 | 北京航空航天大学 | Man-machine collaboration robot grasping system and its working method |
| CN112659115A (en) * | 2019-10-15 | 2021-04-16 | 大族激光科技产业集团股份有限公司 | Industrial robot control method, device, equipment and storage medium |
| CN110919638A (en) * | 2019-11-15 | 2020-03-27 | 华中科技大学 | 3+4 new-configuration double-arm cooperative robot machining system and method |
| CN110815188A (en) * | 2019-11-19 | 2020-02-21 | 福州大学 | System and method for teaching industrial robot based on limb movement |
| CN111390941A (en) * | 2020-05-08 | 2020-07-10 | 中国船舶科学研究中心 | Command input device for underwater mechanical arm control and control method |
| CN111390941B (en) * | 2020-05-08 | 2021-08-10 | 中国船舶科学研究中心 | Command input device for underwater mechanical arm control and control method |
| CN111730601A (en) * | 2020-07-20 | 2020-10-02 | 季华实验室 | Wearable demonstrator demonstration control method and device and electronic equipment |
| CN112549025A (en) * | 2020-11-27 | 2021-03-26 | 北京工业大学 | Coordination diagram double-arm cooperative control method based on fusion of human body kinematic constraints |
| CN112549025B (en) * | 2020-11-27 | 2022-06-21 | 北京工业大学 | Coordination diagram double-arm cooperative control method based on fusion of human body kinematic constraints |
| CN112706150B (en) * | 2020-12-17 | 2022-06-07 | 武汉大学 | Wearable external limb auxiliary grabbing device and control method |
| CN112706150A (en) * | 2020-12-17 | 2021-04-27 | 武汉大学 | Wearable external limb auxiliary grabbing device and control method |
| CN112959330A (en) * | 2021-02-02 | 2021-06-15 | 浙江大学 | Robot double-arm motion man-machine corresponding device and method based on master-slave dynamic motion elements |
| CN113386128A (en) * | 2021-05-11 | 2021-09-14 | 华南理工大学 | Body potential interaction method for multi-degree-of-freedom robot |
| CN113386128B (en) * | 2021-05-11 | 2022-06-10 | 华南理工大学 | Body potential interaction method for multi-degree-of-freedom robot |
| CN113180738A (en) * | 2021-05-25 | 2021-07-30 | 山东大学 | Virtual probe for remote cardiac ultrasound and diagnosis and treatment system |
| CN113618710A (en) * | 2021-07-21 | 2021-11-09 | 慧灵科技（深圳）有限公司 | Dragging teaching method and device and dragging teaching robot |
| CN113618717A (en) * | 2021-09-07 | 2021-11-09 | 浙江理工大学 | Human upper limb skill action sensing learning device and method |
| CN114631687A (en) * | 2022-03-07 | 2022-06-17 | 深圳市瑞必达科技有限公司 | Control method for performing knocking protection and inclination protection on lifting table without calculating Euler angle |
| CN114631687B (en) * | 2022-03-07 | 2023-05-12 | 深圳市瑞必达科技有限公司 | Control method for knocking protection and tilting protection by lifting table without calculating Euler angle |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107363813B (en) | 2020-08-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107363813A (en) | A kind of desktop industrial robot teaching system and method based on wearable device | |
| CN108549322B (en) | Pose synchronization method and device for arc track motion of robot | |
| CN105291138B (en) | It is a kind of to strengthen the visual feedback platform of virtual reality immersion sense | |
| CN105014677B (en) | Vision Mechanical arm control method based on Camshift visual tracking and D-H modeling algorithm | |
| CN109895101A (en) | A kind of articulated manipulator inverse kinematics numerical value unique solution acquiring method | |
| CN106313049B (en) | A kind of apery mechanical arm motion sensing control system and control method | |
| CN107953331B (en) | human body posture mapping method applied to humanoid robot action simulation | |
| CN106055522A (en) | Minimum base attitude disturbance track planning method for redundant space manipulator | |
| CN106938470A (en) | A kind of device and method of Robot Force control teaching learning by imitation | |
| CN102375416B (en) | Human type robot kicking action information processing method based on rapid search tree | |
| CN103085069A (en) | Novel robot kinematics modeling method | |
| CN107818318B (en) | Humanoid robot simulation similarity evaluation method | |
| CN106607910A (en) | Robot real-time simulation method | |
| CN108656116A (en) | Serial manipulator kinematic calibration method based on dimensionality reduction MCPC models | |
| CN115464659B (en) | Mechanical arm grabbing control method based on visual information deep reinforcement learning DDPG algorithm | |
| CN107791248A (en) | Control method based on the six degree of freedom serial manipulator for being unsatisfactory for pipper criterions | |
| CN106903680B (en) | Snake-shaped robot based on elastic envelope climbs safely control method | |
| CN111260649A (en) | Close-range mechanical arm sensing and calibrating method | |
| CN105459116B (en) | A kind of teleoperation of robot device and method based on magnetometer | |
| CN104678766A (en) | Method for solving optimal batting configuration for flight ball body operation of humanoid manipulator | |
| Natale et al. | Learning precise 3d reaching in a humanoid robot | |
| Rosado et al. | Reproduction of human arm movements using Kinect-based motion capture data | |
| Gäbert et al. | Generation of human-like arm motions using sampling-based motion planning | |
| CN109693235A (en) | A kind of Prosthetic Hand vision tracking device and its control method | |
| Kondo | Inverse kinematics of a human arm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |