CN107363813B - Desktop industrial robot teaching system and method based on wearable equipment - Google Patents
Desktop industrial robot teaching system and method based on wearable equipment Download PDFInfo
- Publication number
- CN107363813B CN107363813B CN201710707674.9A CN201710707674A CN107363813B CN 107363813 B CN107363813 B CN 107363813B CN 201710707674 A CN201710707674 A CN 201710707674A CN 107363813 B CN107363813 B CN 107363813B
- Authority
- CN
- China
- Prior art keywords
- arm
- module
- data
- tail end
- teaching
- 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.)
- Active
Links
Images
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
Abstract
The invention discloses a desktop industrial robot teaching system and method based on wearable equipment. The teaching system obtains the Euler angles of the positions of all sensors on the arms of the human body, calculates and obtains the positions of the tail ends of the arms and the rotation angles of all joints, and obtains continuous motion tracks of the tail ends of the arms and the rotation angles of the corresponding joints after coding, representing and generalizing the discrete motion data. The robot is subjected to kinematic inverse solution according to the tail end position, so that multiple groups of inverse solutions are obtained, the multiple groups of inverse solutions are evaluated according to the rotation angles of the corresponding joints, and the inverse solution closest to the arm posture is selected as the joint angle of the robot. The teaching system provided by the invention can be used for teaching an industrial robot only by normal manual operation without teaching a teaching box or a trailing mechanical arm, so that the teaching efficiency is improved.
Description
Technical Field
The invention relates to the technical field of automation, in particular to a desktop industrial robot teaching system and method based on wearable equipment.
Background
Various six-degree-of-freedom mechanical arms are widely applied to the current automatic production system due to the flexibility of motion and the reasonability of mechanism design. The main task of the robot is to replace human beings to perform manual operations with repeatability, poor environment and high risk, and the premise for completing the operations is to give instructions to the robot in advance and to specify the specific contents of actions and operations to be completed by the robot, and the process is to teach the robot. Teaching playback is a programming method commonly used by robots, and an operator needs to repeatedly adjust the operation parameters of the robot at each teaching point by using a teaching box. And after the whole teaching process is finished, the robot repeatedly operates according to the recorded data. The teaching of the hand grip is also a form of teaching reproducing system, the operator demonstrates the operation track of the robot by operating the control handle installed at the tail end of the robot, and the robot runs according to the previously taught track through the stored data during operation. The existing teaching and reproducing system has higher requirement on the operating skill of an operator, and has the disadvantages of complicated teaching process, time consumption and low efficiency. In order to improve the teaching speed, most mechanical arm control systems provide teaching boxes convenient to operate, but the teaching efficiency of the six-degree-of-freedom mechanical arm cannot be improved fundamentally in the mode.
Disclosure of Invention
In order to solve the above problems, the present invention provides a desktop industrial robot teaching system and method based on wearable device, which at least partially solve the above technical problems.
Therefore, the invention provides a desktop industrial robot teaching system based on wearable equipment, which comprises a teaching data acquisition part, a data processing part and an industrial robot, wherein the data processing part is respectively connected with the teaching data acquisition part and the industrial robot;
the teaching data acquisition part comprises wearable equipment, the wearable equipment comprises data control nodes, a data sending module and 6 sensor nodes, the 6 sensor nodes are respectively and averagely arranged on the left arm and the right arm, 3 sensor nodes of each arm are respectively arranged at the far shoulder joint end of the upper arm, the far elbow joint end of the forearm and the tail end of the arm, and the data control nodes are arranged on the back of a human body;
the sensor node is used for acquiring motion information of the arm joint in a motion process;
the data control node is used for converging the motion information of each sensor node and carrying out fusion processing on the motion information according to a preset attitude estimation algorithm so as to obtain Euler angle data of the corresponding position of the sensor;
the data sending module is used for transmitting the Euler angle data to the data processing part in a wireless transmission mode;
the data processing part comprises a data receiving module, an arm motion information calculating module, a data coding learning and generalization output module, a robot kinematics inverse solution module and an optimal solution evaluating module, wherein the arm motion information calculating module is respectively connected with the data receiving module and the data coding learning and generalization output module, the data coding learning and generalization output module is respectively connected with the robot kinematics inverse solution module and the optimal solution evaluating module, and the optimal solution evaluating module is connected with the robot kinematics inverse solution module;
the data receiving module is used for receiving the Euler angle data and transmitting the Euler angle data to the arm movement information calculating module;
the arm motion information calculation module is used for calculating an arm tail end position, an arm tail end posture and an arm joint rotation angle according to the Euler angle;
the data coding learning and generalization output module is used for carrying out coding representation and generalization output on discrete arm tail end positions, arm tail end postures and arm joint rotation angles so as to obtain continuous arm tail end positions, arm tail end postures and arm joint rotation angles;
the robot kinematics inverse solution module is used for performing robot kinematics inverse solution on continuous arm tail end positions and arm tail end postures to obtain corresponding multiple groups of inverse solutions;
the optimal solution evaluation module is used for selecting one inverse solution from the multiple groups of inverse solutions as the joint rotation angle of the robot according to the minimum evaluation factor, and the evaluation factor is formed according to the multiple groups of inverse solutions and the continuous arm joint rotation angle;
the industrial robot comprises an industrial robot body and a bottom layer control module, wherein the industrial robot body is connected with the bottom layer control module;
the bottom layer control module is used for controlling the industrial robot body according to the joint rotation angle of the robot, so that the industrial robot reproduces arm movement of a demonstrator.
Optionally, the system further comprises an information fusion module, wherein the information fusion module is respectively connected with the arm movement information calculation module and the data coding learning and generalization output module;
the information fusion module is used for carrying out fusion processing on the arm tail end position, the arm tail end posture and the arm joint rotation angle to form a preset vector data set;
and the data coding learning and generalization output module is used for carrying out coding representation and generalization output on the discrete vector data set so as to obtain continuous arm tail end positions, arm tail end postures and arm joint rotation angles.
Optionally, the vector data set is a vector D, D ═ x, y, z, αW,βW,γW,θ1,θ2,θ3,θ4,θ5,θ6Where x, y and z are the positions of the ends of the arms in the world coordinate system, αW、βWAnd gammaWIs the pose of the end of the arm in the world coordinate system, θ1、θ2、θ3、θ4、θ5And theta6The corresponding rotation angle of the arm joint.
Optionally, the calculation formula of the evaluation factor is as follows:
wherein λ isi(i ═ 1, 2, ·, 6) is an influence factor of each arm joint on the position and posture of a mechanical arm of the industrial robot;
Ni={θR1,θR2,θR3,θR4,θR5,θR6the (i is 1, 2, 3. cndot.) is the i group inverse solution at the end position of the same arm;
DP={θ’1,θ’2,θ’3,θ’4,θ’5,θ’6the angle of rotation of the arm joint at the same time is used as the mean value.
Optionally, the euler angle data comprises the euler angle Ψ of the distal shoulder joint endBThe Euler angle psi of the distal elbow joint end of the forearmFEuler angle psi of the tip of the armH,ΨB={αB,βB,γB}T,ΨF={αF,βF,γF}T,ΨH={αH,βH,γH}TWherein αB、βBAnd gammaBAre respectively Euler angle psiBNutation, precession and autorotation angles of (iii), αF、βFAnd gammaFAre respectively Euler angle psiFNutation, precession and autorotation angles of (iii), αH、βHAnd gammaHAre respectively Euler angle psiHNutation angle, precession angle and spin angle.
The invention also provides a desktop industrial robot teaching method based on wearable equipment, which adopts a desktop industrial robot teaching system based on wearable equipment, wherein the teaching system comprises a teaching data acquisition part, a data processing part and an industrial robot, and the data processing part is respectively connected with the teaching data acquisition part and the industrial robot;
the teaching data acquisition part comprises wearable equipment, the wearable equipment comprises data control nodes, a data sending module and 6 sensor nodes, the 6 sensor nodes are respectively and averagely arranged on the left arm and the right arm, 3 sensor nodes of each arm are respectively arranged at the far shoulder joint end of the upper arm, the far elbow joint end of the forearm and the tail end of the arm, and the data control nodes are arranged on the back of a human body;
the data processing part comprises a data receiving module, an arm motion information calculating module, a data coding learning and generalization output module, a robot kinematics inverse solution module and an optimal solution evaluating module, wherein the arm motion information calculating module is respectively connected with the data receiving module and the data coding learning and generalization output module, the data coding learning and generalization output module is respectively connected with the robot kinematics inverse solution module and the optimal solution evaluating module, and the optimal solution evaluating module is connected with the robot kinematics inverse solution module;
the industrial robot comprises an industrial robot body and a bottom layer control module, wherein the industrial robot body is connected with the bottom layer control module;
the teaching method comprises the following steps:
the sensor node acquires motion information of the arm joint in a motion process;
the data control node gathers motion information of each sensor node, and fusion processing is carried out on the motion information according to a preset attitude estimation algorithm so as to obtain Euler angle data of the corresponding position of the sensor;
the data sending module transmits the Euler angle data to the data processing part in a wireless transmission mode;
the data receiving module receives the Euler angle data and transmits the Euler angle data to the arm movement information calculating module;
the arm motion information calculation module calculates the position of the tail end of the arm, the posture of the tail end of the arm and the rotation angle of the arm joint according to the Euler angle;
the data coding learning and generalization output module is used for carrying out coding representation and generalization output on discrete arm tail end positions, arm tail end postures and arm joint rotation angles so as to obtain continuous arm tail end positions, arm tail end postures and arm joint rotation angles;
the robot kinematics inverse solution module carries out robot kinematics inverse solution on the continuous arm tail end positions and arm tail end postures so as to obtain a plurality of corresponding inverse solutions;
the optimal solution evaluation module selects one inverse solution from the multiple inverse solutions as a joint rotation angle of the robot according to a minimum evaluation factor, and the evaluation factor is formed according to the multiple inverse solutions and continuous arm joint rotation angles;
the bottom layer control module controls the industrial robot body according to the joint rotation angle of the robot, so that the industrial robot reproduces arm movement of a demonstrator.
Optionally, the teaching system further includes an information fusion module, and the information fusion module is respectively connected to the arm movement information calculation module and the data coding learning and generalization output module;
the data coding learning and generalization output module carries out coding representation and generalization output on discrete arm tail end positions, arm tail end postures and arm joint rotation angles, and the steps comprise:
the information fusion module performs fusion processing on the arm tail end position, the arm tail end posture and the arm joint rotation angle to form a preset vector data set;
the data coding learning and generalization output module carries out coding representation and generalization output on discrete arm tail end positions, arm tail end postures and arm joint rotation angles, and the steps comprise:
and the data coding learning and generalization output module is used for carrying out coding representation and generalization output on the discrete vector data set so as to obtain continuous arm tail end positions, arm tail end postures and arm joint rotation angles.
Optionally, the step of performing coded representation and generalized output on the discrete arm end position, the arm end posture and the arm joint rotation angle by the data coding learning and generalized output module includes:
the data coding learning and generalization output module is used for carrying out coding representation on the discrete arm tail end position, the discrete arm tail end posture and the discrete arm joint rotation angle by utilizing a Gaussian mixture model so as to realize the characterization learning of the discrete arm tail end position, the discrete arm tail end posture and the discrete arm joint rotation angle;
and the data coding learning and generalization output module performs data reconstruction and generalization output on the coded arm tail end position, arm tail end posture and arm joint rotation angle by using a Gaussian mixture regression model so as to obtain continuous arm tail end position, arm tail end posture and arm joint rotation angle.
The invention has the following beneficial effects:
the invention provides a desktop industrial robot teaching system and method based on wearable equipment. The teaching system obtains the Euler angles of the positions of all sensors on the human arm in the teaching process, calculates and obtains the positions of the tail ends of the arms and the rotation angles of all joints, and obtains continuous tail end motion tracks and the rotation angles of the corresponding joints after encoding, representing and generalizing the discrete motion data. The robot is subjected to kinematic inverse solution according to the tail end position, so that multiple groups of inverse solutions are obtained, the multiple groups of inverse solutions are evaluated according to the rotation angles of the corresponding joints, and the inverse solution closest to the arm posture is selected as the joint angle of the robot. The teaching system provided by the invention realizes that a teaching person can complete teaching of the industrial robot only according to normal manual operation without teaching a teaching box or dragging a mechanical arm, thereby improving the teaching efficiency. Teaching process of the teaching system is simple and convenient, does not have higher operating skill requirement to the teaching personnel, and only wearable equipment needs to be worn. The teaching system enables the gesture in the motion process of the robot to be close to the gesture of the motion of the human arm, so that the motion process of the human arm can be restored to the maximum extent, the motion is smooth and stable, the motion precision is controllable, and the teaching effect is good.
Drawings
Fig. 1 is a schematic structural diagram of a desktop industrial robot teaching system based on a wearable device according to an embodiment of the present invention;
fig. 2 is a schematic information transfer diagram of a wearable device-based desktop industrial robot teaching system according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a human arm simplified to a three-link seven-degree-of-freedom model according to an embodiment of the present invention;
fig. 4 is a schematic flow chart of a wearable device-based desktop industrial robot teaching method according to a second embodiment of the present invention.
Detailed Description
In order to make those skilled in the art better understand the technical solution of the present invention, the following describes in detail a wearable device based desktop industrial robot teaching system and method provided by the present invention with reference to the accompanying drawings.
Example one
Fig. 1 is a schematic structural diagram of a wearable device-based desktop industrial robot teaching system according to an embodiment of the present invention, and fig. 2 is a schematic information transfer diagram of the wearable device-based desktop industrial robot teaching system according to an embodiment of the present invention. As shown in fig. 1 and fig. 2, the present embodiment provides a desktop industrial robot teaching system based on a wearable device, which includes a teaching data acquisition section, a data processing section, and an industrial robot, wherein the data processing section is connected to the teaching data acquisition section and the industrial robot, respectively. The teaching system obtains the Euler angles of the positions of all sensors on the human arm in the teaching process, calculates and obtains the positions of the tail ends of the arms and the rotation angles of all joints, and obtains continuous tail end motion tracks and the rotation angles of the corresponding joints after encoding, representing and generalizing the discrete motion data. In this embodiment, a kinematic inverse solution is performed on the robot according to the end position, so as to obtain multiple sets of inverse solutions, the multiple sets of inverse solutions are evaluated according to the rotation angles of the corresponding joints, and a set of inverse solutions closest to the arm posture is selected as the joint angle of the robot.
The teaching data acquisition part that this embodiment provided includes wearable equipment, wearable equipment includes data control node, data transmission module and 6 sensor nodes, and 6 sensor nodes are average to be set up respectively at left arm and right arm, and 3 sensor nodes of every arm set up respectively at shoulder joint end far away from the upper arm, elbow joint end far away from the forearm and arm terminal, data control node sets up at human back. The sensor nodes are used for acquiring motion information of arm joints in a motion process, the data control nodes are used for converging the motion information of each sensor node and carrying out fusion processing on the motion information according to a preset attitude estimation algorithm so as to obtain Euler angle data of the corresponding position of the sensor, and the data sending module is used for transmitting the Euler angle data to the data processing part in a wireless transmission mode. The teaching system that this embodiment provided has realized that the demonstrator need not the teach box or drags the arm and demonstrates, only need just can accomplish the teaching to industrial robot according to normal manual operation to teaching efficiency has been improved. The teaching process of the teaching system provided by the embodiment is simple and convenient, does not have higher operation skill requirement to the teaching personnel, and only wearable equipment needs to be worn. The teaching system provided by the embodiment enables the gesture in the robot motion process to be close to the gesture of the human arm motion, so that the motion process of the human arm can be restored to the maximum extent, the motion is smooth and stable, the motion precision is controllable, and the teaching effect is good.
The data control node provided by this embodiment is a convergence control node. Wearing equipment based on inertial sensor includes 6 sensor nodes and 1 aggregation control node. The sensor node collects motion information of the arm joint in the motion process, mainly comprises a three-axis accelerometer, a three-axis gyroscope and a three-axis geomagnetic instrument, and is respectively used for collecting three-axis acceleration (a) of the arm jointx,ay,az) Three-axis angular velocity (g)x,gy,gz) And three-axis magnetic field strength (m)x,my,mz). The aggregation control node aggregates the data of each sensor node, processes the data and sends the data to the data processing part in a wireless transmission mode.
The 6 sensor nodes provided by the embodiment are distributed on the left arm and the right arm, and the left arm and the right arm are respectively 3. The 3 sensor nodes of each arm are respectively distributed at the far shoulder joint end of the upper arm, the far elbow joint end of the forearm and the palm center. The palm center provided by the embodiment is the end of the arm. Each sensor node consists of 1 triaxial accelerometer, 1 triaxial gyroscope and 1 triaxial geomagnetic instrument. The three-axis accelerometer is used to acquire the three-axis acceleration (a) at each jointx,ay,az) A three-axis gyroscope is used to acquire the three-axis angular velocities (g) of rotation at each jointx,gy,gz) The three-axis magnetometer is used for acquiring the three-axis magnetic field intensity (m) at each jointx,my,mz). The convergence control nodes are distributed on the back of the human body and used for converging the acceleration, the angular velocity and the data of the magnetometer of each joint and carrying out fusion processing on the data. TopThe spirometer sensor respectively measures angular speeds of the carrier along the directions of X, Y and a Z coordinate system, and then the integral is carried out by combining with sampling time, the posture of the sensor is calculated and is expressed by an Euler angle form. In the embodiment, the attitude information is corrected and jointly resolved by combining the acceleration sensor and the magnetic sensor, then the sensor data is subjected to fusion calculation based on the attitude estimation algorithm of Kalman filtering, and then the data is packaged and wirelessly transmitted to the data processing part.
The data processing part that this embodiment provided includes that data receiving module, arm motion information calculation module, data code study and generalization output module, robot kinematics inverse solution module and optimal solution evaluation module, arm motion information calculation module respectively with data receiving module with data code study is connected with generalization output module, data code study and generalization output module respectively with robot kinematics inverse solution module with the optimal solution evaluation module is connected, the optimal solution evaluation module with robot kinematics inverse solution module is connected. The data receiving module is used for receiving the Euler angle data and transmitting the Euler angle data to the arm motion information calculating module, the arm motion information calculating module is used for calculating an arm tail end position, an arm tail end posture and an arm joint rotation angle according to the Euler angle, the data coding learning and generalization output module is used for coding, representing and generalizing the discrete arm tail end position, the arm tail end posture and the arm joint rotation angle so as to obtain continuous arm tail end position, arm tail end posture and arm joint rotation angle, the robot kinematics inverse solution module is used for carrying out robot kinematics inverse solution on the continuous arm tail end position and arm tail end posture so as to obtain corresponding multiple groups of inverse solutions, the optimal solution evaluating module is used for selecting one group of inverse solutions from the multiple groups of inverse solutions according to the minimum evaluation factor as the joint rotation angle of the robot, the evaluation factor is formed from the plurality of sets of inverse solutions and successive arm joint rotation angles. The teaching system that this embodiment provided has realized that the demonstrator need not the teach box or drags the arm and demonstrates, only need just can accomplish the teaching to industrial robot according to normal manual operation to teaching efficiency has been improved. The teaching process of the teaching system provided by the embodiment is simple and convenient, does not have higher operation skill requirement to the teaching personnel, and only wearable equipment needs to be worn. The teaching system provided by the embodiment enables the gesture in the robot motion process to be close to the gesture of the human arm motion, so that the motion process of the human arm can be restored to the maximum extent, the motion is smooth and stable, the motion precision is controllable, and the teaching effect is good.
In this embodiment, the data receiving module receives the teaching data sent from the viewing angle data acquiring module, and then transmits the teaching data to the arm joint rotation angle calculating module. The calculation content of the arm movement information calculation module is divided into two parts: the arm end position and the arm joint angle. During the teaching process, the pose of the end of the arm is important teaching data for teaching the end effector of the industrial robot. In this embodiment, the end of the arm is the palm center. The accurate teaching of the pose of the end effector is an important prerequisite for the industrial robot to complete the operation task. And according to the characteristics of the teaching data acquired by the wearable device, the arm end pose calculation submodule calculates the position and the posture of the arm respectively. The data collected by the inertial sensor is the euler angles of the positions of the inertial sensor relative to the world coordinate system, and the data collected by the wearable device provided by the embodiment is the euler angles of the elbow joint, the wrist joint and the palm center relative to the world coordinate system. Therefore, the pose of the end of the arm can be derived from the data of the inertial sensor at the palm. The position of the palm center needs to be obtained by combining the data calculation of the inertial sensors at the elbow joint, the wrist joint and the palm center.
The present embodiment establishes the following coordinate system: the demonstrator lifts the arms horizontally and with the palm facing upwards. A world coordinate system WCS is established by taking the shoulder joint as a coordinate origin, the X axis is forward along the arm, the Z axis is vertically upward, the Y axis is determined by a right-hand rule, and the world coordinate system WCS does not move along with the arm. In the embodiment, a coordinate system S, a coordinate system E and a coordinate system W are respectively established at the shoulder joint, the elbow joint and the wrist joint: the coordinate system S takes the shoulder joint as the origin of coordinates, and the forward direction along the arm is the X axis and the Z axisVertically upwards, determining a Y axis by adopting a right-hand rule, wherein a coordinate system S is fixedly connected to the large arm and moves along with the large arm, namely the coordinate system S is superposed with a world coordinate system WCS at an initial position; a coordinate system E takes the elbow joint as the origin of coordinates, the forward direction along the arm direction is an X axis, a Z axis is vertical upward, a Y axis is determined by adopting a right hand rule, and the coordinate system E is fixedly connected to the forearm and moves along with the forearm; the coordinate system W takes the wrist joint as the origin of coordinates, the forward direction along the palm direction is an X axis, the Z axis is vertical upward, the Y axis is determined by adopting a right hand rule, and the coordinate system W is fixedly connected with the palm and moves along with the palm. Thus, the data acquired by the inertial sensor at the elbow joint (distal shoulder joint of the upper arm) is the euler angle Ψ of the coordinate system S relative to the world coordinate system WCSS={αS,βS,γS}T(ii) a The data acquired by the inertial sensor at the wrist joint (distal forearm elbow) is the euler angle Ψ of the coordinate system E relative to the world coordinate system WCSE={αE,βE,γE}T(ii) a The data collected by the inertial sensor at the palm center is the euler angle Ψ of the coordinate system W relative to the world coordinate system WCSW={αW,βW,γW}T. The length of the big arm of the human arm is r1The length of the forearm being r2The length from the sensor at the palm to the wrist joint is taken as the calculated length r of the palm3。
In the embodiment, when calculating the position of the end of the arm, the arm is regarded as a group of vectors connected end to end in the world coordinate system WCS:whereinIn order to represent the vector of the large arm,in order to represent the vector of the forearm,is a vector representing the palm. Thus, the position of the end of the armThe position is the coordinate of the sum vector of the 3 vectors. As can be seen from the calculation of the vectors, the sum vector of a group of vectors connected end to end can be equivalent to the sum vector of a group of vectors sharing a starting point, and the coordinates of the sum vector of 3 vectors can be obtained by projecting the 3 vectors in the direction X, Y, Z respectively. Namely, it is
In this embodiment, the euler angle Ψ of the coordinate system S relative to the world coordinate system WCSS={αS,βS,γS}TEuler angle Ψ of coordinate system E relative to world coordinate system WCSE={αE,βE,γE}TW phase of coordinate systemEuler angle Ψ for the world coordinate System WCSW={αW,βW,γW}T. From the euler angle transformation between the two coordinate systems, the rotation matrix R thereof can be obtained, the rotation order being in terms of Z-X-Z:
this embodiment will be psiS={αS,βS,γS}T、ΨE={αE,βE,γE}TAnd ΨW={αW,βW,γW}TSubstituting the above formula can obtain:
rotation matrix of coordinate system S relative to world coordinate system WCSRotation matrix of coordinate system E relative to world coordinate system WCSRotation matrix of coordinate system W relative to world coordinate system WCSTherefore, the position matrix of the arm end in the world coordinate system WCS is:
in this embodiment, the posture of the end of the arm is the palm posture, and since the sensor is fixed on the palm of the hand, the sensor and the palm have the same rotation state, i.e., the euler angle calculated from the sensor data can be used to represent the posture of the palm. Therefore, the posture matrix of the palm center isFor the convenience of subsequent module processing, it is still passed down in the form of Euler angles, i.e. ΨW={αW,βW,γW}T。
Fig. 3 is a schematic diagram of a human arm simplified into a three-link seven-degree-of-freedom model according to an embodiment of the present invention. As shown in fig. 3, the shoulder joint, the elbow joint and the wrist joint of the human arm have 3 degrees of freedom, 2 degrees of freedom and 2 degrees of freedom, respectively, and the human arm can be simplified into a three-link seven-degree-of-freedom model according to the motion characteristics of the human arm and the theory of rigid body hypothesis. The arm joint rotation angle calculation module receives the teaching data transmitted by the data receiving module and calculates the rotation angles of 7 joints of the arm. The embodiment measures the Euler angles psi of the large arm, the small arm and the palm respectively through the motion data obtained by the wearable deviceS,ΨE,ΨW,ΨS={αS,βS,γS}T,ΨE={αE,βE,γE}T,ΨW={αW,βW,γW}TWherein αB,βB,γBRespectively nutation angle, precession angle and self-rotation angle of Euler angle of large arm αF,βF,γFRespectively nutation angle, precession angle and self-rotation angle in Euler angle of small arm αH,βH,γHRespectively, nutation angle, precession angle and rotation angle in palm euler angle. In order to calculate the joint angle of the arm of the human body, the present embodiment adopts a joint space nonsingular mapping algorithm based on quaternion, and the calculation steps are as follows:
(1) the rotation angle of the shoulder joint is: theta1=βS,θ2=αS,θ3=γS
(2) The remaining 4 joint angles of the arm were obtained by calculating the euler angle of the lower arm relative to the upper arm and the euler angle of the palm relative to the lower arm. This embodiment converts the measured euler angles into quaternions Q of the upper arm, lower arm and wrist of the human armS,QE,QWTo avoid singularities. Euler angle psirTo quaternion QrThe conversion formula of (1) is as follows:
wherein q is1r,q2r,q3r,q4rIs a quaternion Qrαr,βr,γrAre respectively Euler angle psirNutation angle, precession angle and spin angle.
(3) Suppose the quaternion of the small arm relative to the large arm is QESThe quaternion of the palm with respect to the forearm being QWEThen Q isE=QESQS,QW=QWEQEI.e. relative quaternion is QES=QEQS -1,QWE=QWQE -1
(4) Relative quaternion QESAnd QWEConversion to the relative euler angle ΨES={αES,βES,γES}T,ΨWE={αWE,βWE,γWE}TWherein ΨESIs the Euler angle of the small arm relative to the large arm, αES,βES,γESRespectively corresponding nutation angle, precession angle and self-rotation angle; ΨWEIs the Euler angle of the palm relative to the forearm, αWE,βWE,γWERespectively the nutation angle, the precession angle and the rotation angle corresponding to the nutation angle, the precession angle and the rotation angle. Quaternion QSTo the Euler angle psiSThe conversion formula of (1) is as follows:
(5) angle of rotation of elbow joint theta4=αES,θ5=γES
(6) Rotation angle of wrist joint is theta6=αWE,θ7=βWE
Finally, the embodiment transmits the calculated 7 joint angles to the information fusion module.
The true bookIn the embodiment, the arm end position calculation submodule and the arm joint angle calculation submodule respectively calculate the arm end position, the arm end posture and the rotation angle of each joint of the arm. The two data are obtained by different calculation methods from the same group of sensor data, and are arm movement data at the same time. Meanwhile, the human arm has 7 degrees of freedom, but the industrial robot for motion reproduction has only 6 degrees of freedom, and only 6 degrees of freedom are needed in the evaluation of the optimal solution, so that 1 degree of freedom of the human arm needs to be abandoned. Giving up the last degree of freedom at the wrist, θ, taking into account the degree of influence on the posture of the human arm7. Discarding theta7Does not affect the attitude accuracy of the industrial robot because of theta7It is only used in the optimal solution evaluation phase. Meanwhile, in order to facilitate the processing of the subsequent modules, the two groups of data are merged into one group of data, and the vector D is used to represent:
D={x,y,z,αW,βW,γW,θ1,θ2,θ3,θ4,θ5,θ6}
wherein x, y, z and αW,βW,γWRespectively representing the position and attitude of the arm's tip in the world coordinate system WCS, theta1,θ2,θ3,θ4,θ5,θ6The corresponding joint angle of the arm. And all motion information in the whole teaching process is transmitted to a subsequent module in a data set form and is recorded as a data set { D }.
In this embodiment, the teaching data is acquired at a certain time interval, and the acquired data is discrete. In the subsequent motion reproduction process, the pose adopted by the robot kinematics inverse solution module is not necessarily a point acquired in the teaching process. Therefore, it is necessary to generate continuous arm movement information using discrete arm movement information. In this embodiment, the motion information obtained above, i.e., the vectors D at different times, are encoded by using the GMM gaussian mixture model, so as to implement characterization learning, and then data reconstruction and generalization output are performed by using the GMR gaussian mixture regression model, so as to obtain the continuous arm end positions and the rotation angles of the joints. Given any time, the present embodiment can obtain corresponding arm movement data, denoted by D'.
Suppose the ith data is di={Di,TiAnd f, i is {1, 2, 3, ·, N }, wherein N is the number of times of data acquired by the teaching data acquisition module in the process of one arm movement, and DiAnd Ti is the time value of the arm movement information acquired at the ith time. If each data point di obeys the following probability distribution:
where p (k) is the prior probability and p (di | k) is the conditional probability distribution, obeying a gaussian distribution. Therefore, the entire teaching data set can be expressed by the gaussian mixture model, and K is the number of gaussian distributions constituting the gaussian mixture model. The formula of prior probability and conditional probability distribution is as follows:
p(k)=πk(10)
where D is the dimension of the GMM encoding the teaching data. Thus, the parameter that the Gaussian mixture model needs to determine is { π }k,μk,∑kDenotes the prior probability, expectation and variance, respectively, of the kth component. The parameters of the GMM are estimated using a maximum expectation algorithm, the EM algorithm, and parameter learning is performed by finding a maximum likelihood estimate of the parameters in the probabilistic model. Time value T of teaching dataiUsed as a query point, the Di' of the corresponding motion information is estimated using the GMR. It is known that p (di | k) satisfies the Gaussian distribution, { D }k,Tk}~N(μk,∑k) Wherein, in the step (A),
μk={μD,k,μT,k} (12)
at a given TkUnder the conditions of (D)kAlso satisfies the Gaussian distribution, i.e., Dk|Tk~N(μ’D,k,∑’D,k). Wherein
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 is clear from the above equation, the mean value μ 'of the Gaussian mixture model of K Gaussian components'DAnd variance ∑'DWherein
From Dk|Tk~N(μ’D,k,∑’D,k) Distribution estimation conditions expect E (D | T), i.e., μ'DFor the reconstructed spatial value corresponding to T, the generalized data point is thus D ' ═ { D ', T }, which is not included in the arm motion information provided by the preceding modules, but which encapsulates all essential features of arm motion, under covariance constraint ∑ 'DContinuous arm movement information can be generated.
According to the requirements of the motion reproduction technology for the motion of the industrial robot, at corresponding time intervals, T is usediThe query points are queried for arm movement information D'. The subsequent modules provided by this embodiment have different uses for the arm end pose and the arm joint rotation angle in the arm motion information vector D'. Accurate meter for arm end positionTherefore, the data representation learning and generalization output module is used for correlating the data x ', y ', z ', α ' with the pose of the end arm 'W、β’W、γ’WTransmitting the data to a robot kinematics inverse solution module, and converting the data theta 'related to the arm joint rotation angle'1、θ’2、θ’3、θ’4、θ’5,θ’6And transmitting the data to an optimal solution evaluation module. The teaching system provided by the embodiment enables the gesture in the robot motion process to be close to the gesture of the human arm motion, so that the motion process of the human arm can be restored to the maximum extent, the motion is smooth and stable, the motion precision is controllable, and the teaching effect is good.
The inverse solution module for robot kinematics provided in this embodiment obtains arm end pose data x ', y', z ', α' W, β 'transmitted by the data representation learning and generalization output module'W、γ’WThen, the Euler angle is α'W,β’W,γ’WConverted into rotation matrix according to the methodAnd then combining the positions x, y and z of the tail ends of the arms to form a pose matrix in the robot learningWherein P ═ x ', y ', z ']'. The kinematic inverse solution of the robot is used to obtain the corresponding rotation angle of each joint of the robot, and the vector N is used as the resultiIs represented by Ni={θR1,θR2,θR3,θR4,θR5,θR6And i is the number of inverse solutions.
For the same end position, the inverse kinematics solution module of the robot will obtain multiple sets of solutions. The optimal solution is selected from a plurality of groups of solutions, and the common method is to select the possible solution which is closest to the value at the moment of the joint angle from the large joint as the only solution. The method ensures that the motion amount of the industrial robot is minimum and the motion is more stable, but the similarity between the gesture of the industrial robot in the motion process and the gesture of the teaching arm cannot be ensured.
In this example, Ni={θR1,θR2,θR3,θR4,θR5,θR6And (i ═ 1, 2, 3 ·) is the inverse solution of the i group at the same terminal position, DP={θ’1,θ’2,θ’3,θ’4,θ’5,θ’6The angle of the arm joint at the same moment is defined as follows:
wherein λiAnd (i & ltgt1, 2 & gth & lt6 & gt) is an influence factor of each joint on the pose of the mechanical arm. In the movement of human arm, the swing of arm has far more influence on the arm position than the twisting of arm, which only has great influence on the arm posture. Similarly, in the motion process of the industrial robot, the effect of the rotation of different joints on the position of the mechanical arm is greatly different. The rotational effect of some joints is equivalent to the swinging of the human arm. Aiming at a specific mechanical arm, the influence of different joints on the pose of the mechanical arm is analyzed according to specific structural characteristics, and similar to a human arm, corresponding influence factors are given to the different joints of the mechanical arm, and the larger the influence of the joints on the pose of the mechanical arm is, the larger the influence factors are.
The larger the numerical value of the evaluation factor of each group of solutions is, the larger the difference between the pose of the mechanical arm corresponding to the group of solutions and the pose of the human arm is. Conversely, the smaller the numerical value of the evaluation factor is, the closer the pose of the mechanical arm corresponding to the solution is to the pose of the human arm. Therefore, the optimal solution evaluation module selects one solution with the minimum evaluation factor from the multiple sets of inverse solutions, so that the posture of the robot in the motion process is close to the posture of the motion of the human arm, the motion process of the human arm can be restored to the maximum extent, the motion is smooth and stable, the motion precision is controllable, and the teaching effect is good.
The industrial robot that this embodiment provided includes industrial robot body and bottom control module, the industrial robot body with bottom control module connects, bottom control module is used for controlling according to the joint turned angle of robot the industrial robot body to make industrial robot reproduces teach person's arm motion. The industrial robot body that this embodiment provided and bottom control module thereof are as the executive component that teaching system's motion reappeared, bottom control module receives a set of solution that optimal solution evaluation module selected, with this set of solution as the turned angle of each joint to control the robot body, make the robot body reappear demonstrator's arm motion process, teaching process is simple and convenient, does not have higher operating skill requirement to the demonstrator, only need dress wearable equipment can.
The desktop industrial robot teaching system based on the wearable equipment comprises a teaching data acquisition part, a data processing part and an industrial robot. The teaching system obtains the Euler angles of the positions of all sensors on the human arm in the teaching process, calculates and obtains the positions of the tail ends of the arms and the rotation angles of all joints, and obtains continuous tail end motion tracks and the rotation angles of the corresponding joints after encoding, representing and generalizing the discrete motion data. In this embodiment, a kinematic inverse solution is performed on the robot according to the end position, so as to obtain multiple sets of inverse solutions, the multiple sets of inverse solutions are evaluated according to the rotation angles of the corresponding joints, and a set of inverse solutions closest to the arm posture is selected as the joint angle of the robot. The teaching system that this embodiment provided has realized that the demonstrator need not the teach box or drags the arm and demonstrates, only need just can accomplish the teaching to industrial robot according to normal manual operation to teaching efficiency has been improved. The teaching process of the teaching system provided by the embodiment is simple and convenient, does not have higher operation skill requirement to the teaching personnel, and only wearable equipment needs to be worn. The teaching system provided by the embodiment enables the gesture in the robot motion process to be close to the gesture of the human arm motion, so that the motion process of the human arm can be restored to the maximum extent, the motion is smooth and stable, the motion precision is controllable, and the teaching effect is good.
Example two
The embodiment provides a desktop industrial robot teaching method based on wearable equipment, the teaching method adopts a desktop industrial robot teaching system based on the wearable equipment, the teaching system comprises a teaching data acquisition part, a data processing part and an industrial robot, the data processing part is respectively connected with the teaching data acquisition part and the industrial robot, the teaching data acquisition part comprises the wearable equipment, the wearable equipment comprises a data control node, a data sending module and 6 sensor nodes, the 6 sensor nodes are respectively and averagely arranged on a left arm and a right arm, the 3 sensor nodes of each arm are respectively arranged on an upper arm far shoulder joint end, a forearm far elbow joint end and an arm tail end, the data control node is arranged on the back of a human body, the data processing part comprises a data receiving module, a data processing module and a data processing module, The robot kinematics inverse solution system comprises an arm movement information calculation module, a data coding learning and generalization output module, a robot kinematics inverse solution module and an optimal solution evaluation module, wherein the arm movement information calculation module is respectively connected with a data receiving module and the data coding learning and generalization output module, the data coding learning and generalization output module is respectively connected with the robot kinematics inverse solution module and the optimal solution evaluation module, the optimal solution evaluation module is connected with the robot kinematics inverse solution module, an industrial robot comprises an industrial robot body and a bottom layer control module, and the industrial robot body is connected with the bottom layer control module. The teaching system that this embodiment provided has realized that the demonstrator need not the teach box or drags the arm and demonstrates, only need just can accomplish the teaching to industrial robot according to normal manual operation to teaching efficiency has been improved. The teaching process of the teaching system provided by the embodiment is simple and convenient, does not have higher operation skill requirement to the teaching personnel, and only wearable equipment needs to be worn.
Fig. 4 is a schematic flow chart of a wearable device-based desktop industrial robot teaching method according to a second embodiment of the present invention. As shown in fig. 4, the teaching method includes: the sensor node acquires motion information of the arm joint in a motion process; the data control node gathers motion information of each sensor node, and fusion processing is carried out on the motion information according to a preset attitude estimation algorithm so as to obtain Euler angle data of the corresponding position of the sensor; the data sending module transmits the Euler angle data to the data processing part in a wireless transmission mode; the data receiving module receives the Euler angle data and transmits the Euler angle data to the arm movement information calculating module; the arm motion information calculation module calculates the position of the tail end of the arm, the posture of the tail end of the arm and the rotation angle of the arm joint according to the Euler angle; the data coding learning and generalization output module is used for carrying out coding representation and generalization output on discrete arm tail end positions, arm tail end postures and arm joint rotation angles so as to obtain continuous arm tail end positions, arm tail end postures and arm joint rotation angles; the robot kinematics inverse solution module carries out robot kinematics inverse solution on the continuous arm tail end positions and arm tail end postures so as to obtain a plurality of corresponding inverse solutions; the optimal solution evaluation module selects one inverse solution from the multiple inverse solutions as a joint rotation angle of the robot according to a minimum evaluation factor, and the evaluation factor is formed according to the multiple inverse solutions and continuous arm joint rotation angles; the bottom layer control module controls the industrial robot body according to the joint rotation angle of the robot, so that the industrial robot reproduces arm movement of a demonstrator. For a specific implementation process of the teaching method, please refer to the specific description of the teaching process of the teaching system in the embodiment, which is not described herein again.
In the teaching method of the desktop industrial robot based on the wearable device, the teaching system comprises a teaching data acquisition part, a data processing part and an industrial robot. The teaching system obtains the Euler angles of the positions of all sensors on the human arm in the teaching process, calculates and obtains the positions of the tail ends of the arms and the rotation angles of all joints, and obtains continuous tail end motion tracks and the rotation angles of the corresponding joints after encoding, representing and generalizing the discrete motion data. In this embodiment, a kinematic inverse solution is performed on the robot according to the end position, so as to obtain multiple sets of inverse solutions, the multiple sets of inverse solutions are evaluated according to the rotation angles of the corresponding joints, and a set of inverse solutions closest to the arm posture is selected as the joint angle of the robot. The teaching system that this embodiment provided has realized that the demonstrator need not the teach box or drags the arm and demonstrates, only need just can accomplish the teaching to industrial robot according to normal manual operation to teaching efficiency has been improved. The teaching process of the teaching system provided by the embodiment is simple and convenient, does not have higher operation skill requirement to the teaching personnel, and only wearable equipment needs to be worn. The teaching system provided by the embodiment enables the gesture in the robot motion process to be close to the gesture of the human arm motion, so that the motion process of the human arm can be restored to the maximum extent, the motion is smooth and stable, the motion precision is controllable, and the teaching effect is good.
It will be understood that the above embodiments are merely exemplary embodiments taken to illustrate the principles of the present invention, which is not limited thereto. It will be apparent to those skilled in the art that various modifications and improvements can be made without departing from the spirit and substance of the invention, and these modifications and improvements are also considered to be within the scope of the invention.
Claims (6)
1. A desktop industrial robot teaching system based on wearable equipment is characterized by comprising a teaching data acquisition part, a data processing part and an industrial robot, wherein the data processing part is respectively connected with the teaching data acquisition part and the industrial robot;
the teaching data acquisition part comprises wearable equipment, the wearable equipment comprises data control nodes, a data sending module and 6 sensor nodes, the 6 sensor nodes are respectively and averagely arranged on the left arm and the right arm, 3 sensor nodes of each arm are respectively arranged at the far shoulder joint end of the upper arm, the far elbow joint end of the forearm and the tail end of the arm, and the data control nodes are arranged on the back of a human body;
the sensor node is used for acquiring motion information of the arm joint in a motion process;
the data control node is used for converging the motion information of each sensor node and carrying out fusion processing on the motion information according to a preset attitude estimation algorithm so as to obtain Euler angle data of the corresponding position of the sensor;
the data sending module is used for transmitting the Euler angle data to the data processing part in a wireless transmission mode;
the data processing part comprises a data receiving module, an arm motion information calculating module, a data coding learning and generalization output module, a robot kinematics inverse solution module and an optimal solution evaluating module, wherein the arm motion information calculating module is respectively connected with the data receiving module and the data coding learning and generalization output module, the data coding learning and generalization output module is respectively connected with the robot kinematics inverse solution module and the optimal solution evaluating module, and the optimal solution evaluating module is connected with the robot kinematics inverse solution module;
the data receiving module is used for receiving the Euler angle data and transmitting the Euler angle data to the arm movement information calculating module;
the arm motion information calculation module is used for calculating an arm tail end position, an arm tail end posture and an arm joint rotation angle according to the Euler angle;
the data coding learning and generalization output module is used for carrying out coding representation and generalization output on discrete arm tail end positions, arm tail end postures and arm joint rotation angles so as to obtain continuous arm tail end positions, arm tail end postures and arm joint rotation angles;
the robot kinematics inverse solution module is used for performing robot kinematics inverse solution on continuous arm tail end positions and arm tail end postures to obtain corresponding multiple groups of inverse solutions;
the optimal solution evaluation module is used for selecting one inverse solution from the multiple groups of inverse solutions as the joint rotation angle of the robot according to the minimum evaluation factor, and the evaluation factor is formed according to the multiple groups of inverse solutions and the continuous arm joint rotation angle;
the industrial robot comprises an industrial robot body and a bottom layer control module, wherein the industrial robot body is connected with the bottom layer control module;
the bottom layer control module is used for controlling the industrial robot body according to the joint rotation angle of the robot so that the industrial robot reproduces arm movement of a demonstrator;
the desktop industrial robot teaching system based on the wearable equipment further comprises an information fusion module, and the information fusion module is respectively connected with the arm movement information calculation module and the data coding learning and generalization output module;
the information fusion module is used for carrying out fusion processing on the arm tail end position, the arm tail end posture and the arm joint rotation angle to form a preset vector data set;
and the data coding learning and generalization output module is used for carrying out coding representation and generalization output on the discrete vector data set so as to obtain continuous arm tail end positions, arm tail end postures and arm joint rotation angles.
2. The wearable device-based desktop industrial robot teaching system of claim 1 wherein the vector data set is a vector D, D ═ x, y, z, αW,βW,γW,θ1,θ2,θ3,θ4,θ5,θ6Where x, y and z are the positions of the ends of the arms in the world coordinate system, αW、βWAnd gammaWIs the pose of the end of the arm in the world coordinate system, θ1、θ2、θ3、θ4、θ5And theta6The corresponding rotation angle of the arm joint.
3. The wearable device based desktop industrial robot teaching system of claim 1 wherein the evaluation factor is calculated as follows:
wherein λ isi(i ═ 1, 2, ·, 6) is an influence factor of each arm joint on the position and posture of a mechanical arm of the industrial robot;
Ni={θR1,θR2,θR3,θR4,θR5,θR6the (i is 1, 2, 3. cndot.) is the i group inverse solution at the end position of the same arm;
DP={θ’1,θ’2,θ’3,θ’4,θ’5,θ’6the angle of rotation of the arm joint at the same time is used as the mean value.
4. The wearable device based desktop industrial robot teach system of claim 1 wherein the euler angle data comprises the euler angle Ψ of the distal shoulder joint endBThe Euler angle psi of the distal elbow joint end of the forearmFEuler angle psi of the tip of the armH,ΨB={αB,βB,γB}T,ΨF={αF,βF,γF}T,ΨH={αH,βH,γH}TWherein αB、βBAnd gammaBAre respectively Euler angle psiBNutation, precession and autorotation angles of (iii), αF、βFAnd gammaFAre respectively Euler angle psiFNutation, precession and autorotation angles of (iii), αH、βHAnd gammaHAre respectively Euler angle psiHNutation angle, precession angle and spin angle.
5. A desktop industrial robot teaching method based on wearable equipment is characterized in that a desktop industrial robot teaching system based on the wearable equipment is adopted in the teaching method, the teaching system comprises a teaching data acquisition part, a data processing part and an industrial robot, and the data processing part is respectively connected with the teaching data acquisition part and the industrial robot;
the teaching data acquisition part comprises wearable equipment, the wearable equipment comprises data control nodes, a data sending module and 6 sensor nodes, the 6 sensor nodes are respectively and averagely arranged on the left arm and the right arm, 3 sensor nodes of each arm are respectively arranged at the far shoulder joint end of the upper arm, the far elbow joint end of the forearm and the tail end of the arm, and the data control nodes are arranged on the back of a human body;
the data processing part comprises a data receiving module, an arm motion information calculating module, a data coding learning and generalization output module, a robot kinematics inverse solution module and an optimal solution evaluating module, wherein the arm motion information calculating module is respectively connected with the data receiving module and the data coding learning and generalization output module, the data coding learning and generalization output module is respectively connected with the robot kinematics inverse solution module and the optimal solution evaluating module, and the optimal solution evaluating module is connected with the robot kinematics inverse solution module;
the industrial robot comprises an industrial robot body and a bottom layer control module, wherein the industrial robot body is connected with the bottom layer control module;
the teaching method comprises the following steps:
the sensor node acquires motion information of the arm joint in a motion process;
the data control node gathers motion information of each sensor node, and fusion processing is carried out on the motion information according to a preset attitude estimation algorithm so as to obtain Euler angle data of the corresponding position of the sensor;
the data sending module transmits the Euler angle data to the data processing part in a wireless transmission mode;
the data receiving module receives the Euler angle data and transmits the Euler angle data to the arm movement information calculating module;
the arm motion information calculation module calculates the position of the tail end of the arm, the posture of the tail end of the arm and the rotation angle of the arm joint according to the Euler angle;
the data coding learning and generalization output module is used for carrying out coding representation and generalization output on discrete arm tail end positions, arm tail end postures and arm joint rotation angles so as to obtain continuous arm tail end positions, arm tail end postures and arm joint rotation angles;
the robot kinematics inverse solution module carries out robot kinematics inverse solution on the continuous arm tail end positions and arm tail end postures so as to obtain a plurality of corresponding inverse solutions;
the optimal solution evaluation module selects one inverse solution from the multiple inverse solutions as a joint rotation angle of the robot according to a minimum evaluation factor, and the evaluation factor is formed according to the multiple inverse solutions and continuous arm joint rotation angles;
the bottom layer control module controls the industrial robot body according to the joint rotation angle of the robot, so that the industrial robot reproduces arm movement of a demonstrator;
the teaching system also comprises an information fusion module which is respectively connected with the arm movement information calculation module and the data coding learning and generalization output module;
the data coding learning and generalization output module carries out coding representation and generalization output on discrete arm tail end positions, arm tail end postures and arm joint rotation angles, and the steps comprise:
the information fusion module performs fusion processing on the arm tail end position, the arm tail end posture and the arm joint rotation angle to form a preset vector data set;
the data coding learning and generalization output module carries out coding representation and generalization output on discrete arm tail end positions, arm tail end postures and arm joint rotation angles, and the steps comprise:
and the data coding learning and generalization output module is used for carrying out coding representation and generalization output on the discrete vector data set so as to obtain continuous arm tail end positions, arm tail end postures and arm joint rotation angles.
6. The wearable device based desktop industrial robot teaching method of claim 5 wherein the data encoding learning and generalization output module encoding and generalizing discrete arm end positions, arm end poses and arm joint rotation angles comprises:
the data coding learning and generalization output module is used for carrying out coding representation on the discrete arm tail end position, the discrete arm tail end posture and the discrete arm joint rotation angle by utilizing a Gaussian mixture model so as to realize the characterization learning of the discrete arm tail end position, the discrete arm tail end posture and the discrete arm joint rotation angle;
and the data coding learning and generalization output module performs data reconstruction and generalization output on the coded arm tail end position, arm tail end posture and arm joint rotation angle by using a Gaussian mixture regression model so as to obtain continuous arm tail end position, arm tail end posture and arm joint rotation 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 CN107363813A (en) | 2017-11-21 |
CN107363813B true 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) |
Families Citing this family (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 |
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 |
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 |
CN112659115B (en) * | 2019-10-15 | 2022-12-09 | 深圳市大族机器人有限公司 | Industrial robot control method, device, equipment and storage medium |
CN110919638B (en) * | 2019-11-15 | 2021-11-02 | 华中科技大学 | 3+ 4-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 |
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 |
CN111730601B (en) * | 2020-07-20 | 2021-01-01 | 季华实验室 | Wearable demonstrator demonstration control method and device and electronic equipment |
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 |
CN112959330B (en) * | 2021-02-02 | 2022-05-17 | 浙江大学 | Robot double-arm motion man-machine corresponding device and method based on master-slave dynamic motion elements |
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 |
CN113618710B (en) * | 2021-07-21 | 2023-03-24 | 慧灵科技(深圳)有限公司 | Dragging teaching method and device and dragging teaching robot |
CN113618717B (en) * | 2021-09-07 | 2023-01-24 | 浙江理工大学 | Human upper limb skill action sensing learning device and method |
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 |
Family Cites Families (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 |
CN102735260B (en) * | 2012-06-18 | 2015-02-11 | 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 |
CN104267815B (en) * | 2014-09-25 | 2018-03-27 | 黑龙江节点动画有限公司 | Motion capture system and method based on inertia sensing technology |
CN106064378A (en) * | 2016-06-07 | 2016-11-02 | 南方科技大学 | The control method of a kind of unmanned plane mechanical arm and device |
-
2017
- 2017-08-17 CN CN201710707674.9A patent/CN107363813B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN107363813A (en) | 2017-11-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107363813B (en) | Desktop industrial robot teaching system and method based on wearable equipment | |
CN106625666B (en) | The control method and device of redundant mechanical arm | |
CN108549322B (en) | Pose synchronization method and device for arc track motion of robot | |
US20180020978A1 (en) | System and method for measuring the movements of articulated rigid bodies | |
Fang et al. | A robotic hand-arm teleoperation system using human arm/hand with a novel data glove | |
US8467904B2 (en) | Reconstruction, retargetting, tracking, and estimation of pose of articulated systems | |
Almetwally et al. | Real-time tele-operation and tele-walking of humanoid Robot Nao using Kinect Depth Camera | |
JP2022542241A (en) | Systems and methods for augmenting visual output from robotic devices | |
CN109079794B (en) | Robot control and teaching method based on human body posture following | |
US20120072022A1 (en) | Robot and control method thereof | |
Xu et al. | Development of a hybrid motion capture method using MYO armband with application to teleoperation | |
CN110986939A (en) | Visual inertial odometer method based on IMU pre-integration | |
JP2018024082A (en) | Multiaxial motion control device, robot arm system, method of controlling movement of robot arm system, and method of controlling movement of multiaxial motion driving device | |
CN112207835B (en) | Method for realizing double-arm cooperative work task based on teaching learning | |
CN107696033B (en) | Space manipulator trajectory rolling planning method based on visual measurement | |
CN108098780A (en) | A kind of new robot apery kinematic system | |
US11845006B2 (en) | Skeleton model updating apparatus, skeleton model updating method, and program | |
Zhang et al. | A real-time upper-body robot imitation system | |
CN110609621A (en) | Posture calibration method and human motion capture system based on micro-sensor | |
Guilamo et al. | Manipulability optimization for trajectory generation | |
Biess et al. | Simulating discrete and rhythmic multi-joint human arm movements by optimization of nonlinear performance indices | |
Hunt et al. | Predictive trajectory estimation during rehabilitative tasks in augmented reality using inertial sensors | |
US20200320719A1 (en) | Determining a kinematic sequence | |
CN114417738A (en) | Sparse IMU real-time human body motion capture and joint stress prediction method and system | |
CN111563346B (en) | Arm kinematics modeling method based on Gaussian process learning |
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 |