CN109732610A - Man-machine collaboration robot grasping system and its working method - Google Patents
Man-machine collaboration robot grasping system and its working method Download PDFInfo
- Publication number
- CN109732610A CN109732610A CN201910154643.4A CN201910154643A CN109732610A CN 109732610 A CN109732610 A CN 109732610A CN 201910154643 A CN201910154643 A CN 201910154643A CN 109732610 A CN109732610 A CN 109732610A
- Authority
- CN
- China
- Prior art keywords
- robot
- variable
- joint
- module
- man
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Manipulator (AREA)
Abstract
The invention discloses a kind of man-machine collaboration robot grasping system and its working method, man-machine collaboration robot grasping system includes Worn type inertial sensor module, camera spacing module, Gauss model EM algorithm training module, Gauss model processing module and adaptability joint of robot coordinate extraction module.Man-machine collaboration robot provided by the invention grasping system inverts without the calibration and kinematics for carrying out robotic vision system in the application scenarios of man-machine collaboration, reduces the professional skill requirement of operator.In addition, mapping can be realized without a large amount of sample in man-machine collaboration robot provided by the invention grasping system, so that robot trajectory is smoothly submissive.
Description
Technical field
The present invention relates to robotic technology field more particularly to a kind of man-machine collaboration robot grasping system and its work sides
Method.
Background technique
In industrial production and daily life, robot more and more replaces the mankind to execute various operation tasks, such as
Welding, cutting, punching press, spray painting, material processing, precise materials processing, robot have significant advantage in above-mentioned task.?
When robot grabs operation, the pose of target object changes frequent occurrence, and robot needs to be adjusted according to the posture information of object
The adaptability of crawl is realized in the movement of itself.
In the emerging industry of certain robot applications, such as the update of the main products such as mobile phone, plate, wearable device
Replacement speed is very fast, short only some months.Traditional robotic scenarios plenty of time resource will be put on production line,
This does not meet present production model, and the feature in these emerging industries is that there are many product category, volume is generally little.The mankind are negative
The process relatively high to flexibility, tactile, requirement on flexibility is blamed, then fast and accurately feature is responsible for repeatability using it for robot
Work.
More researchs at present be all placed on and the combination of vision on grabbed, but magazine observation change in order to obtain
The joint coordinates for measuring robot need to demarcate vision system, profession of traditional vision calibration method to operator
Property is more demanding, and needs to waste a large amount of time and could complete.
Summary of the invention
A kind of man-machine collaboration robot grasping system is provided to solve limitation and defect, the present invention of the existing technology,
Including data acquiring portion, data processing section, industrial robot part, the data acquiring portion includes that Worn type inertia passes
Sensor module, camera spacing module, the data processing section include Gauss model EM algorithm training module, Gauss model processing
Module, the industrial robot part include adaptability joint of robot coordinate extraction module;
Operator is arranged on hand in the Worn type inertial sensor module, grabs object for extracting the operator
Posture variable;
The camera spacing module extract body marginal point coordinate, obtains the average value of coordinate as target object central point
Coordinate (x, y);
The Gauss model EM algorithm training module forms object observation using the parameter of training data estimation Gauss model
Mapping relations between variable and joint of robot variable;
The Gauss model processing module is associated the joint of robot variable and the object observational variable, leads to
The mapping relations that training obtains in advance are crossed, corresponding joint of robot variable is predicted according to new object observational variable;
The adaptability joint of robot coordinate extraction module extracts the robot that the Gauss model processing module obtains
Joint variable.
Optionally, the Worn type inertial sensor module includes sensor node and convergence control node, the sensing
Device node is arranged at the centre of the palm, and the convergence control node is arranged at the back of the hand;
The sensor node is for acquiring posture information of operator's palm during grabbing object;
The convergence control node is used to collect the data of the sensor node, by wireless transmission method by collection
Data are sent to data processing section.
Optionally, the sensor node is three-axis gyroscope, for obtaining three axis angular velocity of rotation (g of palmx, gy,
gz);
The three-axis gyroscope is also used to carry out integral operation according to three axis angular velocity of rotations and sampling time, is sensed
The posture information of device node is indicated the posture information by way of Eulerian angles.
Optionally, the industrial robot part further includes robot body and bottom controller,
Joint of robot variable is sent to the robot body by the adaptability joint of robot coordinate extraction module
And bottom controller, the bottom controller control the grasping movement that the robot body completes object.
The present invention also provides a kind of working method of man-machine collaboration robot grasping system, the man-machine collaboration robot is grabbed
Taking system includes data acquiring portion, data processing section, industrial robot part, and the data acquiring portion includes Worn type
Inertial sensor module, camera spacing module, the data processing section include Gauss model EM algorithm training module, Gaussian mode
Type processing module, the industrial robot part include adaptability joint of robot coordinate extraction module;
The working method of man-machine collaboration robot grasping system includes:
The Worn type inertial sensor module extracts the posture variable that the operator grabs object;
The camera spacing module extract body marginal point coordinate, obtains the average value of coordinate as target object central point
Coordinate (x, y);
The Gauss model EM algorithm training module forms object observation using the parameter of training data estimation Gauss model
Mapping relations between variable and joint of robot variable;
The Gauss model processing module is associated the joint of robot variable and the object observational variable, leads to
The mapping relations that training obtains in advance are crossed, corresponding joint of robot variable is predicted according to new object observational variable;
The adaptability joint of robot coordinate extraction module extracts the robot that the Gauss model processing module obtains
Joint variable.
Optionally, the Gauss model processing module carries out the joint of robot variable and the object observational variable
Association predicts corresponding joint of robot variable according to new object observational variable by the mapping relations that training obtains in advance
The step of include:
Under the probability distribution of gauss hybrid models, x ° of sample of probability of occurrence are as follows:
Wherein, m represents the total number of Gaussian Profile, p (k)-α (k) representative sample xoFrom the general of k-th Gaussian Profile
Rate, p (xo| k)=N (xo;μk;∑k) represent k-th of Gaussian Profile generation sample xoProbability, μkAnd ∑kIt respectively represents k-th
The mean vector and covariance matrix of Gaussian Profile;
New observational variable onewPosterior probability from k-th of Gaussian Profile are as follows:
p(k|onew)∝p(k)p(onew|k)
k*=argmaxp (k | onew)
By the mean vector μ of k-th of Gaussian ProfilekWith covariance matrix ∑kCarry out piecemeal:
Wherein, μrMean value for the joint variable being made of among training sample set joint of robot coordinate, μoFor training sample
The mean value for the observational variable being made of among this collection the posture information of the center point coordinate of object and sensor, KrrFor joint variable
Covariance, KooFor the covariance of observational variable, KorFor the covariance of observational variable and joint variable, KroFor joint variable with
The covariance of observational variable,
The new observational variable o of robot acquisition objectnewWith the joint angles r of robotnewAdaptable conditional probability
Distribution are as follows:
Wherein,For with new observational variable be adapted joint angles distribution mean value,For the association of Gaussian Profile
Variance matrix.
Optionally, the Worn type inertial sensor module includes sensor node and convergence control node, the sensing
Device node is arranged at the centre of the palm, and the convergence control node is arranged at the back of the hand;
The Worn type inertial sensor module extracts the step of operator grabs the posture variable of object and includes:
Posture information of sensor node acquisition operator's palm during grabbing object;
The convergence control node collects the data of the sensor node, by wireless transmission method by the data of collection
It is sent to data processing section.
The present invention have it is following the utility model has the advantages that
Among man-machine collaboration robot provided by the invention grasping system and its working method, the man-machine collaboration robot
Grasping system includes Worn type inertial sensor module, camera spacing module, Gauss model EM algorithm training module, Gauss model
Processing module, adaptability joint of robot coordinate extraction module, Worn type inertial sensor module extraction operation person grab object
Posture variable, camera spacing module extract body marginal point coordinate obtains the average value of coordinate as target object central point
Coordinate (x, y), Gauss model EM algorithm training module using training data estimation Gauss model parameter, formed object observation
Mapping relations between variable and joint of robot variable, Gauss model processing module observe joint of robot variable and object
Variable is associated, and by the mapping relations that training obtains in advance, predicts corresponding robot according to new object observational variable
Joint variable, adaptability joint of robot coordinate extraction module extract the joint of robot that Gauss model processing module obtains and become
Amount.Man-machine collaboration robot provided by the invention grasping system is in the application scenarios of man-machine collaboration, without progress robot view
The calibration of feel system and kinematics are inverted, and the professional skill requirement of operator is reduced.In addition, provided by the invention man-machine
Mapping can be realized without a large amount of sample in cooperation robot grasping system, so that robot trajectory is smoothly submissive.
Detailed description of the invention
Fig. 1 is the structural schematic diagram for the man-machine collaboration robot grasping system that the embodiment of the present invention one provides.
Fig. 2 is the flow chart for the man-machine collaboration robot grasping system that the embodiment of the present invention one provides.
Fig. 3 is the pixel seat that the man-machine collaboration robot grasping system that the embodiment of the present invention one provides obtains object central point
Mark is intended to.
Fig. 4 is the man-machine collaboration robot grasping system that provides of the embodiment of the present invention one according to the prediction pair of object observational variable
The joint of robot variable answered.
Specific embodiment
To make those skilled in the art more fully understand technical solution of the present invention, the present invention is mentioned with reference to the accompanying drawing
The man-machine collaboration robot grasping system and its working method of confession are described in detail.
Embodiment one
Fig. 1 is the structural schematic diagram for the man-machine collaboration robot grasping system that the embodiment of the present invention one provides.Such as Fig. 1 institute
Show, man-machine collaboration robot provided in this embodiment grasping system include Worn type inertial sensor module, camera spacing module,
Gauss model EM algorithm training module, Gauss model processing module, adaptability joint of robot coordinate extraction module, Worn type are used
Property sensor module extraction operation person grab object posture variable, camera spacing module extract body marginal point coordinate, obtain
Coordinate (x, y) of the average value of coordinate as target object central point, Gauss model EM algorithm training module use training data
It estimates the parameter of Gauss model, forms the mapping relations between object observational variable and joint of robot variable, at Gauss model
Reason module is associated joint of robot variable and object observational variable, by training obtained mapping relations in advance, according to
New object observational variable predicts that corresponding joint of robot variable, adaptability joint of robot coordinate extraction module extract Gauss
The joint of robot variable that model processing modules obtain.Man-machine collaboration robot provided in this embodiment grasping system is in man-machine association
In the application scenarios of work, inverts without the calibration and kinematics for carrying out robotic vision system, reduce the special of operator
Industry skill requirement.It is reflected in addition, man-machine collaboration robot provided in this embodiment grasping system can be realized without a large amount of sample
It penetrates, so that robot trajectory is smoothly submissive.
Fig. 2 is the flow chart for the man-machine collaboration robot grasping system that the embodiment of the present invention one provides.As shown in Fig. 2, institute
Stating man-machine collaboration robot grasping system includes Worn type inertial sensor module, camera spacing module, Gauss model processing mould
Block, adaptability joint of robot coordinate extraction module, are broadly divided into data acquiring portion, data processing section and industrial machine
People part.Data acquiring portion includes Worn type inertial sensor module and camera spacing module, and data processing section includes number
According to reception, Gaussian process processing, training data sample, industrial robot includes joint of robot extraction module and industrial robot
Ontology and its bottom controller.
Worn type inertial sensor module provided in this embodiment includes 1 sensor node and 1 convergence control node.
Sensor node is for acquiring posture information of operator's palm during grabbing object block.Sensor node is mainly three axis tops
Spiral shell instrument, for acquiring three axis angular velocity of rotation (g of palmx, gy, gz).Convergence control node is used to converge the number of sensor node
According to handling above-mentioned data, send it to data processing section by wireless transmission method.
The sensor node is arranged at the palm centre of the palm, and each sensor node is made of 1 three-axis gyroscope.Three axis
Gyroscope is used to acquire three axis angular velocity of rotation (g of each jointx, gy, gz).Control node distribution is converged to be arranged in the back of the hand
On, for collecting angular velocity data.Gyro sensor measures carrier along the angular speed of X, Y and Z coordinate system direction respectively, then ties
The conjunction sampling time is integrated, and the posture information of sensor is calculated, and be indicated using the form of Eulerian angles, finally by data
It is packaged by radioing to data processing section.
Fig. 3 is the pixel seat that the man-machine collaboration robot grasping system that the embodiment of the present invention one provides obtains object central point
Mark is intended to.As shown in figure 3, working region is the rectangular area of workbench center 400mm × 350mm.The present embodiment uses picture
Plain (pixel) indicates that it is about 0.4mm that a pixel, which represents actual range, and wherein x pixel coordinate from 200 to 1200, sit by y pixel
Mark is from 200 to 1075.The object block of crawl is the small cubic block of yellow, and the present embodiment is carried out by the coordinate to color lump marginal point
Average calculating operation obtains average value, can obtain the pixel coordinate x and y of object central point.
Data processing section provided in this embodiment includes data receiver part, Gauss model EM algorithm training module and height
This model processing modules.The data for receiving camera spacing and sensor is sent are responsible in data receiver part, then pass it to height
This processing module.Training data is that object block is placed in working region by people, so that it is reached crawl by dragging robot
Position records the value in each joint of handgrip pose and robot at this time.
The present embodiment executes the data of 20 acquirements as sample set at random.Using the EM algorithm pair of gauss hybrid models
Sample is trained, while sample being sorted out, and every class sample corresponds to different working regions.The sight of the present embodiment building object
Survey the mapping between variable and joint of robot variable.After initialization model parameter, pass through changing for E-step and M-step
In generation, constantly updates the parameter of model, until convergence.Man-machine collaboration robot provided in this embodiment grasping system is in man-machine collaboration
Application scenarios in, without carry out robotic vision system calibration and kinematics invert, reduce the profession of operator
Skill requirement.In addition, mapping can be realized without a large amount of sample in man-machine collaboration robot provided in this embodiment grasping system,
So that robot trajectory is smoothly submissive.
E-step: i-th of sample (x is calculatedi) the weight r from k-th of Gaussian Profileik。
M-step: the parameter of each cluster is updated.
The present embodiment is classified as m class during with EM algorithm training pattern, by training sample, and every one kind sample obeys one
A Gaussian Profile, a region of corresponding working space.Assuming that Different categories of samples is obeyed respectively with μ1μ2…μmFor mean value, with ∑1
∑2... ∑ m is multiple Gaussian Profiles of covariance matrix:
Wherein, hk(k=1,2 ..., m) represents the training sample for obeying k-th of Gaussian Profile, is the one of training sample set X
A subset.
Wherein,Represent i-th of sample for being under the jurisdiction of k-th of Gaussian Profile, nkIt represents
The sample size of k-th of Gaussian Profile (corresponding k-th of zonule).
In the present embodiment, robot needs to adjust the pose for oneself deacclimatizing object: f:o → r when executing task
Wherein, o is the observational variable of object, following oiIt is same;R is joint of robot corresponding with observational variable change
Amount, following riIt is same;F is the mapping from observational variable to joint variable.
In the present embodiment, it is assumed that X={ x1,x2,…,xnIt is the training sample set that is obtained by teaching, wherein xi=[ri,
oi]TFor the individualized training sample (vector) of joint variable and observational variable composition.Machine man-hour, first from training sample set
Learn in X, mapping function f is obtained, when obtaining new observational variable onewWhen, corresponding joint is obtained by mapping function f
Variable rnew。
Fig. 4 is the man-machine collaboration robot grasping system that provides of the embodiment of the present invention one according to the prediction pair of object observational variable
The joint of robot variable answered.As shown in figure 4, selecting suitable model to describe mapping function f, to the crawl of robot adaptability
Learn extremely important.The present embodiment is modeled using gauss hybrid models, constructs the observational variable and joint of robot of object
Mapping between variable.The present embodiment uses EM algorithm training pattern, training sample is divided into several classes, every one kind sample is obeyed
One Gaussian Profile, a corresponding region.Gauss hybrid models are the linear superpositions of multiple Gaussian Profiles.The present embodiment is using high
The modeling of this mixed model, the probability distribution of training sample is described using multiple Gaussian processes.In the probability point of gauss hybrid models
It plants, x ° of sample of probability of occurrence are as follows:
Wherein, m represents the total number of Gaussian Profile, p (k)-α (k) representative sample xoFrom the general of k-th Gaussian Profile
Rate, p (xo| k)=N (xo;μk;∑k) represent k-th of Gaussian Profile generation sample xoProbability, μkAnd ∑kIt respectively represents k-th
The mean vector and covariance matrix of Gaussian Profile.
The present embodiment carries out model training, the parameter of learning model using EM algorithm.With EM algorithm learning model parameter
During, training sample is clustered, sample set is divided into m subset, each sample set obeys Gauss point
Cloth corresponds to a region in working space.
Machine man-hour obtains the observational variable o of object by sensor and cameranew, calculate the observational variable and come from
Each Gaussian Profile posterior probability p (k | onew), it chooses the corresponding Gaussian process recurrence of maximum a posteriori probability and acquires robot pass
Coordinate is saved, driving robot, which realizes, grabs the adaptability of object.
New observational variable onewPosterior probability from k-th of Gaussian Profile are as follows:
Assuming that the corresponding posterior probability of k-th of Gaussian Profile is maximum, then the corresponding Gaussian process of k-th of Gaussian Profile is chosen
Regression forecasting robot joint angles.By the mean vector μ of k-th of Gaussian ProfilekWith covariance matrix ∑kCarry out piecemeal:
Wherein, μrMean value for the joint variable being made of among training sample set joint of robot coordinate, μoFor training sample
The mean value for the observational variable being made of among this collection the posture information of the center point coordinate of object and sensor, KrrFor joint variable
Covariance, KooFor the covariance of observational variable, KorFor the covariance of observational variable and joint variable, KroFor joint variable with
The covariance of observational variable, and
The new observational variable o of robot acquisition objectnewWith the joint angles r of robotnewAdaptable conditional probability
Distribution are as follows:
Wherein,For with new observational variable be adapted joint angles distribution mean value, corresponding to Gaussian Profile
Maximum probability density;For the covariance matrix of Gaussian Profile, the uncertainty of prediction result represent.By driving machine
Each joint of people reachesRobot can be made to complete grasping manipulation with a possibility that maximum.
In the case where not needing progress vision system calibration and kinematics is inverted, gauss hybrid models are directly to multiple areas
Joint variable in domain is associated with observational variable, allows the robot to be adapted therewith according to new observational variable prediction
Joint variable.
Industrial robot part provided in this embodiment includes joint extraction module, robot body and bottom controller,
Treated that joint coordinates extract is sent to robot body and controller by Gauss for joint extraction module, and robot can
To smoothly complete the crawl of object block.
Man-machine collaboration robot provided in this embodiment grasping system includes Worn type inertial sensor module, camera spacing
Module, Gauss model EM algorithm training module, Gauss model processing module, adaptability joint of robot coordinate extraction module are worn
The posture variable that formula inertial sensor module extraction operation person grabs object is worn, camera spacing module extract body marginal point is sat
Mark, obtains coordinate (x, y) of the average value as target object central point of coordinate, Gauss model EM algorithm training module use is shown
The parameter of data estimation Gauss model is taught, the mapping relations between object observational variable and joint of robot variable, Gauss are formed
Model processing modules are associated joint of robot variable and object observational variable, are closed by the mapping that training obtains in advance
System predicts corresponding joint of robot variable, adaptability joint of robot coordinate extraction module according to new object observational variable
Extract the joint of robot variable that Gauss model processing module obtains.Man-machine collaboration robot provided in this embodiment grasping system
In the application scenarios of man-machine collaboration, inverts without the calibration and kinematics for carrying out robotic vision system, reduce operation
The professional skill requirement of personnel.In addition, man-machine collaboration robot provided in this embodiment grasping system is without a large amount of sample
Mapping can be achieved, so that robot trajectory is smoothly submissive.
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, however the present invention is not limited thereto.For those skilled in the art, essence of the invention is not being departed from
In the case where mind and essence, various changes and modifications can be made therein, these variations and modifications are also considered as protection scope of the present invention.
Claims (7)
1. a kind of man-machine collaboration robot grasping system, which is characterized in that including data acquiring portion, data processing section, work
Industry robot part, the data acquiring portion include Worn type inertial sensor module, camera spacing module, at the data
Reason part includes Gauss model EM algorithm training module, Gauss model processing module, and the industrial robot part includes adapting to
Property joint of robot coordinate extraction module;
Operator is arranged on hand in the Worn type inertial sensor module, and the appearance of object is grabbed for extracting the operator
State variable;
The camera spacing module extract body marginal point coordinate, obtains seat of the average value of coordinate as target object central point
It marks (x, y);
The Gauss model EM algorithm training module forms object observational variable using the parameter of training data estimation Gauss model
With the mapping relations between joint of robot variable;
The Gauss model processing module is associated the joint of robot variable and the object observational variable, by pre-
The first mapping relations that training obtains predict corresponding joint of robot variable according to new object observational variable;
The adaptability joint of robot coordinate extraction module extracts the joint of robot that the Gauss model processing module obtains
Variable.
2. man-machine collaboration robot according to claim 1 grasping system, which is characterized in that the Worn type inertia sensing
Device module includes sensor node and convergence control node, and the sensor node is arranged at the centre of the palm, the convergence control section
Point is arranged at the back of the hand;
The sensor node is for acquiring posture information of operator's palm during grabbing object;
The convergence control node is used to collect the data of the sensor node, by wireless transmission method by the data of collection
It is sent to data processing section.
3. man-machine collaboration robot according to claim 2 grasping system, which is characterized in that the sensor node is three
Axis gyroscope, for obtaining three axis angular velocity of rotation (g of palmx, gy, gz);
The three-axis gyroscope is also used to carry out integral operation according to three axis angular velocity of rotations and sampling time, obtains sensor section
The posture information of point, is indicated the posture information by way of Eulerian angles.
4. man-machine collaboration robot according to claim 1 grasping system, which is characterized in that the industrial robot part
It further include robot body and bottom controller,
Joint of robot variable is sent to the robot body and bottom by the adaptability joint of robot coordinate extraction module
Layer controller, the bottom controller control the grasping movement that the robot body completes object.
5. a kind of working method of man-machine collaboration robot grasping system, which is characterized in that the man-machine collaboration robot crawl
System includes data acquiring portion, data processing section, industrial robot part, and the data acquiring portion includes that Worn type is used
Property sensor module, camera spacing module, the data processing section includes Gauss model EM algorithm training module, Gauss model
Processing module, the industrial robot part include adaptability joint of robot coordinate extraction module;
The working method of man-machine collaboration robot grasping system includes:
The Worn type inertial sensor module extracts the posture variable that the operator grabs object;
The camera spacing module extract body marginal point coordinate, obtains seat of the average value of coordinate as target object central point
It marks (x, y);
The Gauss model EM algorithm training module forms object observational variable using the parameter of training data estimation Gauss model
With the mapping relations between joint of robot variable;
The Gauss model processing module is associated the joint of robot variable and the object observational variable, by pre-
The first mapping relations that training obtains predict corresponding joint of robot variable according to new object observational variable;
The adaptability joint of robot coordinate extraction module extracts the joint of robot that the Gauss model processing module obtains
Variable.
6. the working method of man-machine collaboration robot according to claim 5 grasping system, which is characterized in that the Gauss
Model processing modules are associated the joint of robot variable and the object observational variable, are obtained by training in advance
Mapping relations, the step of predicting corresponding joint of robot variable according to new object observational variable include:
Under the probability distribution of gauss hybrid models, x ° of sample of probability of occurrence are as follows:
Wherein, m represents the total number of Gaussian Profile, p (k)-α (k) representative sample xoFrom the probability of k-th of Gaussian Profile, p
(xo| k)=N (xo;μk;∑k) represent k-th of Gaussian Profile generation sample xoProbability, μkAnd ∑kRespectively represent k-th of Gauss
The mean vector and covariance matrix of distribution;
New observational variable onewPosterior probability from k-th of Gaussian Profile are as follows:
By the mean vector μ of k-th of Gaussian ProfilekWith covariance matrix ∑kCarry out piecemeal:
Wherein, μrMean value for the joint variable being made of among training sample set joint of robot coordinate, μoFor training sample set
Among the mean value of observational variable that is made of the posture information of the center point coordinate of object and sensor, KrrFor the association of joint variable
Variance, KooFor the covariance of observational variable, KorFor the covariance of observational variable and joint variable, KroFor joint variable and observation
The covariance of variable,
The new observational variable o of robot acquisition objectnewWith the joint angles r of robotnewAdaptable conditional probability distribution
Are as follows:
Wherein,For with new observational variable be adapted joint angles distribution mean value,For the covariance of Gaussian Profile
Matrix.
7. the working method of man-machine collaboration robot according to claim 5 grasping system, which is characterized in that the wearing
Formula inertial sensor module includes sensor node and convergence control node, and the sensor node is arranged at the centre of the palm, described
Control node is converged to be arranged at the back of the hand;
The Worn type inertial sensor module extracts the step of operator grabs the posture variable of object and includes:
Posture information of sensor node acquisition operator's palm during grabbing object;
The convergence control node collects the data of the sensor node, is sent the data of collection by wireless transmission method
To data processing section.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910154643.4A CN109732610A (en) | 2019-03-01 | 2019-03-01 | Man-machine collaboration robot grasping system and its working method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910154643.4A CN109732610A (en) | 2019-03-01 | 2019-03-01 | Man-machine collaboration robot grasping system and its working method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109732610A true CN109732610A (en) | 2019-05-10 |
Family
ID=66368946
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910154643.4A Pending CN109732610A (en) | 2019-03-01 | 2019-03-01 | Man-machine collaboration robot grasping system and its working method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109732610A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111230873A (en) * | 2020-01-31 | 2020-06-05 | 武汉大学 | Teaching learning-based collaborative handling control system and method |
CN111251277A (en) * | 2020-01-31 | 2020-06-09 | 武汉大学 | Human-computer collaboration tool submission system and method based on teaching learning |
CN111376269A (en) * | 2020-03-04 | 2020-07-07 | 北京海益同展信息科技有限公司 | Object grabbing method and device, storage medium and electronic equipment |
CN111424380A (en) * | 2020-03-31 | 2020-07-17 | 山东大学 | Robot sewing system and method based on skill learning and generalization |
CN111775153A (en) * | 2020-07-17 | 2020-10-16 | 中国科学院宁波材料技术与工程研究所 | Heavy-load robot calibration method |
CN112388628A (en) * | 2019-08-13 | 2021-02-23 | 罗伯特·博世有限公司 | Apparatus and method for training a gaussian process regression model |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103895022A (en) * | 2014-03-17 | 2014-07-02 | 东南大学 | Wearable type somatosensory control mechanical arm |
CN107016342A (en) * | 2017-03-06 | 2017-08-04 | 武汉拓扑图智能科技有限公司 | A kind of action identification method and system |
CN107363813A (en) * | 2017-08-17 | 2017-11-21 | 北京航空航天大学 | A kind of desktop industrial robot teaching system and method based on wearable device |
KR101948558B1 (en) * | 2017-09-28 | 2019-02-18 | 김종태 | Hand-operated programmable modular robot |
CN109382828A (en) * | 2018-10-30 | 2019-02-26 | 武汉大学 | A kind of Robot Peg-in-Hole assembly system and method based on learning from instruction |
-
2019
- 2019-03-01 CN CN201910154643.4A patent/CN109732610A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103895022A (en) * | 2014-03-17 | 2014-07-02 | 东南大学 | Wearable type somatosensory control mechanical arm |
CN107016342A (en) * | 2017-03-06 | 2017-08-04 | 武汉拓扑图智能科技有限公司 | A kind of action identification method and system |
CN107363813A (en) * | 2017-08-17 | 2017-11-21 | 北京航空航天大学 | A kind of desktop industrial robot teaching system and method based on wearable device |
KR101948558B1 (en) * | 2017-09-28 | 2019-02-18 | 김종태 | Hand-operated programmable modular robot |
CN109382828A (en) * | 2018-10-30 | 2019-02-26 | 武汉大学 | A kind of Robot Peg-in-Hole assembly system and method based on learning from instruction |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112388628A (en) * | 2019-08-13 | 2021-02-23 | 罗伯特·博世有限公司 | Apparatus and method for training a gaussian process regression model |
CN111230873A (en) * | 2020-01-31 | 2020-06-05 | 武汉大学 | Teaching learning-based collaborative handling control system and method |
CN111251277A (en) * | 2020-01-31 | 2020-06-09 | 武汉大学 | Human-computer collaboration tool submission system and method based on teaching learning |
CN111251277B (en) * | 2020-01-31 | 2021-09-03 | 武汉大学 | Human-computer collaboration tool submission system and method based on teaching learning |
CN111230873B (en) * | 2020-01-31 | 2022-02-01 | 武汉大学 | Teaching learning-based collaborative handling control system and method |
CN111376269A (en) * | 2020-03-04 | 2020-07-07 | 北京海益同展信息科技有限公司 | Object grabbing method and device, storage medium and electronic equipment |
CN111376269B (en) * | 2020-03-04 | 2021-11-09 | 北京海益同展信息科技有限公司 | Object grabbing method and device, storage medium and electronic equipment |
CN111424380A (en) * | 2020-03-31 | 2020-07-17 | 山东大学 | Robot sewing system and method based on skill learning and generalization |
CN111775153A (en) * | 2020-07-17 | 2020-10-16 | 中国科学院宁波材料技术与工程研究所 | Heavy-load robot calibration method |
CN111775153B (en) * | 2020-07-17 | 2022-08-26 | 中国科学院宁波材料技术与工程研究所 | Heavy-load robot calibration method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109732610A (en) | Man-machine collaboration robot grasping system and its working method | |
CN108972494B (en) | Humanoid manipulator grabbing control system and data processing method thereof | |
Li et al. | Asymmetric bimanual control of dual-arm exoskeletons for human-cooperative manipulations | |
CN106346485B (en) | The Non-contact control method of bionic mechanical hand based on the study of human hand movement posture | |
Morales et al. | Integrated grasp planning and visual object localization for a humanoid robot with five-fingered hands | |
CN108563995B (en) | Human computer cooperation system gesture identification control method based on deep learning | |
CN106737664A (en) | Sort the Delta robot control methods and system of multiclass workpiece | |
Alonso et al. | Current research trends in robot grasping and bin picking | |
CN107160364A (en) | A kind of industrial robot teaching system and method based on machine vision | |
CN110298886B (en) | Dexterous hand grabbing planning method based on four-stage convolutional neural network | |
Kabir et al. | Automated planning for robotic cleaning using multiple setups and oscillatory tool motions | |
CN104156068B (en) | Virtual maintenance interaction operation method based on virtual hand interaction feature layer model | |
CN103112007A (en) | Human-machine interaction method based on mixing sensor | |
CN108858193A (en) | A kind of mechanical arm grasping means and system | |
Vianello et al. | Human posture prediction during physical human-robot interaction | |
CN109968310A (en) | A kind of mechanical arm interaction control method and system | |
CN106514667A (en) | Human-computer cooperation system based on Kinect skeletal tracking and uncalibrated visual servo | |
CN108628260A (en) | Multi items Tool set equipment based on robot and automatic assembling technique | |
Omrčen et al. | Autonomous acquisition of pushing actions to support object grasping with a humanoid robot | |
Li et al. | Neural learning and kalman filtering enhanced teaching by demonstration for a baxter robot | |
Nadon et al. | Automatic selection of grasping points for shape control of non-rigid objects | |
Abad et al. | Fuzzy logic-controlled 6-DOF robotic arm color-based sorter with machine vision feedback | |
Arsenic | Developmental learning on a humanoid robot | |
CN113681565A (en) | Man-machine cooperation method and device for realizing article transfer between robots | |
CN108051001A (en) | A kind of robot movement control method, system and inertia sensing control device |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190510 |