CN109968310A - A kind of mechanical arm interaction control method and system - Google Patents
A kind of mechanical arm interaction control method and system Download PDFInfo
- Publication number
- CN109968310A CN109968310A CN201910292316.5A CN201910292316A CN109968310A CN 109968310 A CN109968310 A CN 109968310A CN 201910292316 A CN201910292316 A CN 201910292316A CN 109968310 A CN109968310 A CN 109968310A
- Authority
- CN
- China
- Prior art keywords
- mechanical arm
- kinect
- coordinate system
- hand
- manpower
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J3/00—Manipulators of master-slave type, i.e. both controlling unit and controlled unit perform corresponding spatial movements
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1602—Programme controls characterised by the control system, structure, architecture
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1694—Programme controls characterised by use of sensors other than normal servo-feedback from position, speed or acceleration sensors, perception control, multi-sensor controlled systems, sensor fusion
- B25J9/1697—Vision controlled systems
Abstract
The present invention relates to a kind of mechanical arm interaction control method and systems, belong to accessible field of engineering technology.The system includes skeleton acquisition of information, computer software, mechanical arm, and human depth's image information acquisition is mainly completed by Kinect sensor.The depth of field data that computer software is obtained with the processing of bone tracer technique, establishes the 3D coordinate of 20 skeleton points of human body.The control system of mechanical arm obtains the control instruction that computer system conversion issues, and control end effector follows hand exercise, realizes manpower to the guiding in real time of robot arm end effector.The present invention constructs hand and carries out master & slave control relative to the movement mapping relations of mechanical arm pedestal relative to hip center and end effector, realizes the real-time, interactive of manpower and robotic arm.The experimental results showed that the system can be good at completing hand tracking and master & slave control task, real-time and interactivity with higher.
Description
Technical field
The invention belongs to accessible field of engineering technology, it is related to a kind of mechanical arm interaction control method and system.
Background technique
Principal and subordinate's remote operating mechanical arm control system be widely used in nuclear reactor maintenance, manned space flight, medical operating mould
In the high risks operations such as quasi- training, hand is the most common position of human body.Therefore, realize that mechanical arm follows hand exercise
The sensing capability and capacity of people can be greatly enhanced.The various movements of research hand, hand and mechanical arm
Interactive mode has important theory significance and realistic meaning.
Currently, based on data glove technology and being based on image vision there are mainly two types of existing hand tracking and positioning technologies
The gesture identification of processing.Data glove technology is more heavy due to itself and its circuit connection, it has not been convenient to use, in addition price is high
The factors such as expensive limit its application.Weinland et al. is proposed the movement of 3D modeling and HMM method identification manpower.Liu et al. benefit
Hand state is tracked and identified with the method that RFID approach is labelled at experimenter's wrist, and this method can determine test
The wrist location of person.But color image is easy to be illuminated by the light condition influence, and the motion target tracking of traditional view-based access control model
It needs to demarcate camera inside and outside parameter, but in practical situations can not Accurate Calibration.
Summary of the invention
In view of this, the purpose of the present invention is to provide a kind of based on Kinect without calibration human-computer interactive control system and
Method, Kinect are a stage body sense video cameras, and background subtracting can be carried out on each frame depth image of acquisition and obtains human body
Each joint 3D coordinate, i.e. skeleton information, and the location information without calibration acquisition manpower from skeleton information.In view of
This, the present invention obtains human hand movement information using Kinect bone tracer technique, and the movement mapping for constructing itself and mechanical arm is closed
System, is studied with regard to the master-slave control method between them.
In order to achieve the above objectives, the invention provides the following technical scheme:
A kind of mechanical arm interaction control method, method includes the following steps:
S1: human depth's image information acquisition;
S2: image preprocessing, human joint points identify and the smoothing processing of skeleton data;
S3: manpower relative to itself hip center movement and mechanical arm relative to the Mapping and Converting between base motion with
And mechanical arm Analysis of Inverse Kinematics and solve;
S4: the generation of control instruction, the transmitting-receiving of standardized instruction format and instruction.
Further, the step S1 specifically:
Using the non-contact acquisition human hand movement information of Kinect skeleton tracer technique, detailed process is: by segmentationization
Human depth's image be transmitted to distinguish human body motion capture machine learning system in, utilize the classification side of random forest
Method designs the intermediate representation of body component, and pose estimation problem is mapped as classification problem pixel-by-pixel;Pass through re-projection classifier
As a result, generate human synovial credible 3D estimation.
Further, the step S2 specifically:
First exceptional value is identified and is rejected before carrying out inverse kinematics, and to skeleton data carry out noise reduction and
Smoothing processing;
It is assumed that the time-shifting step-length of smooth trajectory Processing Algorithm is N, i is current location information in planned position information
Weight, then as k >=N, the planned position at k moment is calculated by following formula and is obtained:
As k < N, the planned position at k moment are as follows:
During solving positional increment, the increment of adjacent moment manpower position is smoothed.
Further, the step S3 specifically:
Cartesian space mapped mode is taken, i.e., using the pose of hand as the input of robot arm end effector pose;
Control mode selection on using increment type control, by the motion pose increment of manpower be added to current robot end execute
On the pose of device, realize that hand controls the guiding in real time of mechanical arm;
The location information of manpower and hip center be under Kinect coordinate system, to establish above-mentioned movement mapping relations,
Its coordinate is transformed under robot coordinate system, concrete methods of realizing is as follows:
S31: position vector of the right hand at Kinect coordinate system { K } is extracted from bone information stream SkeletonFrameKP1=[x1,y1,z1]TAnd position vector of the hip center at { K }KP2=[x2,y2,z2]T;Obtain Kinect coordinate system
Position of { K } the operator right hand relative to itself hip centerKP:
KP=KP1-KP2 (3)
S32: bone tracking coordinate system is all in Kinect camera coordinate system, and origin is infrared camera center, and Z axis is
Infrared camera optical axis, X-direction are horizontal direction, and Y-axis is vertical direction;To obtain seat of the mechanical arm tail end relative to pedestal
MarkBP, demand go out the spin matrix that Kinect camera coordinate system { K } is transformed into robot coordinate system { B }Kinect camera shooting
Head coordinate system { K } rotates -90 ° by the right-hand rule about the z axis, rotates -90 ° further around X-axis, obtains spin matrix
S33: coordinate of the mechanical arm tail end relative to pedestal is finally obtainedBP are as follows:
Using inverse transformation method, i.e., multiply one or several inverse-transform matrixs before transformation matrix, by comparing equation both sides
Corresponding element achievees the purpose that solve Inverse Kinematics Solution;As described above, obtaining mechanical arm tail end in basis coordinates system by formula (5)
Under position vectorBP=[x, y, z]T, homogeneous transformation based on D-H parameters Mechanical arm ending coordinates system relative to base coordinate system
Matrix are as follows:
Wherein, r11=c1c234c5+s1s5;r12=-c1c234c5+s1s5;r13=-c1s234;r21=s1c234c5-c1s5;r22
=-s1c234c5-c1s5;r23=-s1s234;r31=-s234c5;r32=s234c5;r33=-c234;px=c1(a2c2+a3c23-
d5s234);py=s1(a2c2+a3c23-d5s234);pz=-a2s2-a3s23+d1-d5c234.
S in formula234=sin (θ2+θ3+θ4)、c234=cos (θ2+θ3+θ4), other situations are similarly;
Kinematical equation are as follows:
The expression formula for finding out each joint angle of mechanical arm by algebraic approach by (6), (7) is as follows:
θ1=A tan2 (px,py) (8)
θ2=A tan2 (s2,c2) (9)
θ5=A tan2 (r11py-r21px,r12py-r22px) (12)
Wherein
M=(r13px+r23py+r33(pz-d1);
All joint angles of five degree-of-freedom manipulator are found out using algebraic approach.
Mechanical arm intersection control routine based on the method, the system include Kinect sensor, computer software system
System and mechanical arm;
Kinect sensor is used for the acquisition of human depth's image information;
The depth of field data that computer software is obtained with the processing of bone tracer technique, establishes 20 skeleton points of human body
3D coordinate;
The control system of mechanical arm obtains the control instruction that computer system conversion issues, and control end effector is with conveniently
Portion's movement realizes manpower to the guiding in real time of robot arm end effector.
The beneficial effects of the present invention are:
The present invention is a kind of novel man-machine interaction mode, handles depth of field number using Kinect sensor bone tracer technique
According to hand position is obtained, building hand, which is mapped relative to hip center and end effector relative to the movement of mechanical arm pedestal, to be closed
System carries out master & slave control, realizes the real-time, interactive of manpower and robotic arm.And it is directed to hand jitter elimination and outlier processing, it proposes
A kind of rolling average smooth trajectory algorithm based on positional increment.The experimental results showed that the system can be good at completing hand
Tracking and master & slave control task, real-time and interactivity with higher.
Other advantages, target and feature of the invention will be illustrated in the following description to a certain extent, and
And to a certain extent, based on will be apparent to those skilled in the art to investigating hereafter, Huo Zheke
To be instructed from the practice of the present invention.Target of the invention and other advantages can be realized by following specification and
It obtains.
Detailed description of the invention
To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention is made below in conjunction with attached drawing excellent
The detailed description of choosing, in which:
Fig. 1 is master & slave control system framework of the present invention;
Fig. 2 is present system flow chart;
Fig. 3 is that smoothing processing realizes process;
Fig. 4 is link rod coordinate system.
Specific embodiment
Illustrate embodiments of the present invention below by way of specific specific example, those skilled in the art can be by this specification
Other advantages and efficacy of the present invention can be easily understood for disclosed content.The present invention can also pass through in addition different specific realities
The mode of applying is embodied or practiced, the various details in this specification can also based on different viewpoints and application, without departing from
Various modifications or alterations are carried out under spirit of the invention.It should be noted that diagram provided in following embodiment is only to show
Meaning mode illustrates basic conception of the invention, and in the absence of conflict, the feature in following embodiment and embodiment can phase
Mutually combination.
Wherein, the drawings are for illustrative purposes only and are merely schematic diagrams, rather than pictorial diagram, should not be understood as to this
The limitation of invention;Embodiment in order to better illustrate the present invention, the certain components of attached drawing have omission, zoom in or out, not
Represent the size of actual product;It will be understood by those skilled in the art that certain known features and its explanation may be omitted and be in attached drawing
It is understood that.
The same or similar label correspond to the same or similar components in the attached drawing of the embodiment of the present invention;It is retouched in of the invention
In stating, it is to be understood that if there is the orientation or positional relationship of the instructions such as term " on ", "lower", "left", "right", "front", "rear"
To be based on the orientation or positional relationship shown in the drawings, be merely for convenience of description of the present invention and simplification of the description, rather than indicate or
It implies that signified device or element must have a particular orientation, be constructed and operated in a specific orientation, therefore is described in attached drawing
The term of positional relationship only for illustration, is not considered as limiting the invention, for the ordinary skill of this field
For personnel, the concrete meaning of above-mentioned term can be understood as the case may be.
It is a kind of soft without calibration human-computer interactive control system, including skeleton acquisition of information, computer based on Kinect
Part system, mechanical arm, human depth's image information acquisition are mainly completed by Kinect sensor.Computer software uses bone
The depth of field data that the processing of bone tracer technique obtains, establishes the 3D coordinate of 20 skeleton points of human body.The control system of mechanical arm obtains
The control instruction that computer system conversion issues, control end effector follow hand exercise, realize manpower to mechanical arm tail end
The guiding in real time of actuator.
1, system framework
Whole system can be divided into three parts, i.e. skeleton acquisition of information, computer software, mechanical arm.System
Block diagram is illustrated in fig. 1 shown below, and it is as shown in Figure 2 that system runs block diagram.
Human depth's image information acquisition is mainly completed by Kinect sensor.Computer software is tracked with bone
The depth of field data that technical treatment obtains, establishes the 3D coordinate of 20 skeleton points of human body.The control system of mechanical arm obtains computer
The control instruction that system conversion issues, control end effector follow hand exercise, realize manpower to robot arm end effector
Guiding in real time.
2, hand information and master-slave control method are obtained
2.1 Kinect brief introductions
Kinect is a equipment that can obtain in real time color image and depth image data of Microsoft's exploitation, together
When support real-time whole body and half body bone tracing mode, and can identify a series of movement.It by RGB color camera,
RF transmitter and infrared C OMS video camera (IR) three parts composition.
2.2 human testings and the acquisition of hand information
In master & slave control system of the present invention, a key technology is exactly the accurate acquisition of manpower pose.Moving target at present
The common method of detection and tracking has: frame differential method, optical flow method and background subtraction.The shortcomings that optical flow method is to need repeatedly repeatedly
In generation, can just restrain, computationally intensive, it is difficult to meet requirement of real-time;The problem of frame difference method is the region meeting of before and after frames movement overlapping
There is empty generation, detection effect is inaccurate, and the selection of threshold value is affected to result, and threshold value is excessive to will form cavity, threshold
It is worth too small, noise can be generated.Traditional background subtraction is by establishing background model, by current frame image and background model phase
Subtract, can determine position and the shape information of target.However color image is led vulnerable to illumination variation, the influence of complex background
It causes target tracking difficult and divides inaccurate.Background phase is carried out on the depth image for there are some scholars to generate using Kinect
Subtract, has obtained human depth's image of segmentationization, obtained good effect.For this purpose, author proposes a kind of utilization Kinect people
The non-contact method for obtaining human hand movement information of body bone tracer technique, detailed process is: by human depth's image of segmentationization
It is transmitted in the motion capture machine learning system for distinguishing human body, devises body group using the classification method of random forest
Pose estimation problem is mapped as classification problem pixel-by-pixel by the intermediate representation of part.By re-projection classifier, (body component is estimated
Meter) as a result, generate human synovial credible 3D estimation.
The present invention saves complicated camera calibration process, obtains hand position method energy relative to by camera calibration
Enough obtain the higher hand position information of precision.
2.3 coordinate data smoothing processings
Due to the problems such as performance of Kinect hardware is unstable, operator's movement is not coherent enough, estimated by the above method
The relative position of skeletal joint point may change very greatly between frames, and exceptional value may be contained in skeleton data sequence.Cause
This, has to that first exceptional value is identified and rejected before carrying out inverse kinematics, and carries out noise reduction to skeleton data
And smoothing processing.
The principle of rolling average smooth trajectory algorithm is by the manpower location information at current time and preceding N-1 sampling period
Master & slave control location information of the average value of interior manpower location information as current time, and according to time series with step-length N
The collected location information of Kinect is elapsed item by item, finally obtains desired positional value during entire master & slave control,
It realizes to the control from mechanical arm terminal position.The algorithm can filter out the shadow periodically changed to position curve smoothing
It rings, the realization process of smooth trajectory processing is as shown in Figure 3.
Rolling average smooth trajectory Processing Algorithm also has while eliminating manpower periodic jitter and other random disturbances
The desired position operation information of operator may be filtered out.In order to while eliminating hand tremor, reservation operations person as far as possible
Desired location information, the present invention improve current location information after smoothing processing in position on the basis of gliding smoothing algorithm
Shared weight.It is assumed that the time-shifting step-length of smooth trajectory Processing Algorithm is N, i is that current location information is believed in planned position
Weight in breath, then as k >=N, the planned position at k moment can be calculated by following formula to be obtained:
As k < N, the planned position at k moment are as follows:
In collected manpower space motion location information, in addition to human hand movement is not coherent enough and shakes the dry of generation
It disturbs, also the interference of the exceptional value comprising randomness.Since the present invention is the master-slave control strategy based on positional increment, author
During solving positional increment, be not directly to collected hand position information carry out smooth trajectory Processing Algorithm, and
It is to be smoothed to the increment of adjacent moment manpower position.It can preferably be filtered using rolling average algorithm using this mode
Except the erratic variation of collected hand position information.It is verified, using the rolling average smooth trajectory based on positional increment
Algorithm, which carries out processing, can eliminate most of shake and the interference of exceptional value during hand exercise.
Master-slave control method of 2.4 hands to slave manipulator arm
The movement of manpower must be mapped with the movement of mechanical arm when hand controls mechanical arm.It is transported in master & slave control system
There are two types of dynamic mapped mode is usual, one is joint space mapping, another kind is cartesian space mapping.Operation of the present invention task
It is controlled primarily directed to mechanical arm tail end pose, therefore system takes cartesian space mapped mode, it may be assumed that by the position of hand
Input of the appearance as robot arm end effector pose.It is controlled in control mode selection using increment type, by the movement of manpower
Pose increment be added to current robot end effector pose on, realize that hand controls the guiding in real time of mechanical arm.
The location information of manpower and hip center is under Kinect coordinate system, in order to which the movement mapping for establishing above-mentioned is closed
Its coordinate, need to be transformed under robot coordinate system, concrete methods of realizing is as follows by system:
S31: position vector of the right hand at Kinect coordinate system { K } is extracted from bone information stream SkeletonFrameKP1=[x1,y1,z1]TAnd position vector of the hip center at { K }KP2=[x2,y2,z2]T.Therefore without to camera into
Rower can be obtained by position of Kinect coordinate system { K } the operator right hand relative to itself hip center surelyKP:
KP=KP1-KP2 (3)
S32: as shown in Figure 1, bone tracking coordinate system is all in Kinect camera coordinate system, origin is infrared camera
Center, Z axis are infrared camera optical axis, and X-direction is horizontal direction, and Y-axis is vertical direction.Mechanical arm tail end phase in order to obtain
For the coordinate of pedestalBP, demand go out the spin matrix that Kinect camera coordinate system { K } is transformed into robot coordinate system { B }Kinect camera coordinate system { K } rotates -90 ° by the right-hand rule about the z axis, rotates -90 ° further around X-axis, spin moment can be obtained
Battle array
S33: coordinate of the mechanical arm tail end relative to pedestal is finally obtainedBP are as follows:
3, mechanical arm inverse kinematics
The link rod coordinate system for the five degree-of-freedom manipulator that the present invention uses is as shown in figure 4, D-H parameter is as shown in table 1:
The D-H parameter list of 1 mechanical arm of table
In cartesian space, robotic arm Inverse Kinematic Problem is posture and the position of known end opposite base coordinate system
It sets, robotic arm geometric parameter, determines the process of its each joint variable.The present invention uses inverse transformation method, i.e., multiplies before transformation matrix
One or several inverse-transform matrixs achieve the purpose that solve Inverse Kinematics Solution by comparing equation both sides corresponding element.As above
It is described, by position vector of formula (5) the available mechanical arm tail end under basis coordinates systemBP=[x, y, z]T, obtained according to table 1
It arrives, the homogeneous transform matrix based on D-H parameters Mechanical arm ending coordinates system relative to base coordinate system are as follows:
Wherein, r11=c1c234c5+s1s5;r12=-c1c234c5+s1s5;r13=-c1s234;r21=s1c234c5-c1s5;r22
=-s1c234c5-c1s5;r23=-s1s234;r31=-s234c5;r32=s234c5;r33=-c234;px=c1(a2c2+a3c23-
d5s234);py=s1(a2c2+a3c23-d5s234);pz=-a2s2-a3s23+d1-d5c234.
S in formula234=sin (θ2+θ3+θ4)、c234=cos (θ2+θ3+θ4), other situations are similar.
Kinematical equation are as follows:
The expression formula that each joint angle of mechanical arm can be found out by algebraic approach by (6), (7) is as follows:
θ1=A tan2 (px,py) (8)
θ2=A tan2 (s2,c2) (9)
θ5=A tan2 (r11py-r21px,r12py-r22px) (12)
Wherein
M=(r13px+r23py+r33(pz-d1)。
In this way, just finding out all joint angles of five degree-of-freedom manipulator using algebraic approach.Since the certain joints of mechanical arm are deposited
In symmetrical angle, the case where above method acquires some pose of robot arm end effector inverse there may be multiple groups solution.So
And joint motions are limited by displacement range, so the corresponding pose of certain inverse solutions is impossible.To machine
When tool arm carries out practical control, the present invention takes multiple groups against solution using joint limit and the angle from optimal path
House.
Finally, it is stated that the above examples are only used to illustrate the technical scheme of the present invention and are not limiting, although referring to compared with
Good embodiment describes the invention in detail, those skilled in the art should understand that, it can be to skill of the invention
Art scheme is modified or replaced equivalently, and without departing from the objective and range of the technical program, should all be covered in the present invention
Scope of the claims in.
Claims (5)
1. a kind of mechanical arm interaction control method, it is characterised in that: method includes the following steps:
S1: human depth's image information acquisition;
S2: image preprocessing, human joint points identify and the smoothing processing of skeleton data;
S3: movement and mechanical arm of the manpower relative to itself hip center are relative to the Mapping and Converting and machine between base motion
Tool arm Analysis of Inverse Kinematics and solution;
S4: the generation of control instruction, the transmitting-receiving of standardized instruction format and instruction.
2. a kind of mechanical arm interaction control method according to claim 1, it is characterised in that: the step S1 specifically:
Using the non-contact acquisition human hand movement information of Kinect skeleton tracer technique, detailed process is: by the people of segmentationization
Body depth image is transmitted in the motion capture machine learning system for distinguishing human body, is set using the classification method of random forest
The intermediate representation for counting body component, is mapped as classification problem pixel-by-pixel for pose estimation problem;Pass through the knot of re-projection classifier
Fruit generates the credible 3D estimation of human synovial.
3. a kind of mechanical arm interaction control method according to claim 1, it is characterised in that: the step S2 specifically:
First exceptional value is identified and rejected before carrying out inverse kinematics, and to skeleton data progress noise reduction and smoothly
Processing;
It is assumed that the time-shifting step-length of smooth trajectory Processing Algorithm is N, i is power of the current location information in planned position information
Weight, then as k >=N, the planned position at k moment is calculated by following formula to be obtained:
As k < N, the planned position at k moment are as follows:
During solving positional increment, the increment of adjacent moment manpower position is smoothed.
4. a kind of mechanical arm interaction control method according to claim 1, it is characterised in that: the step S3 specifically:
Cartesian space mapped mode is taken, i.e., using the pose of hand as the input of robot arm end effector pose;It is controlling
Mode processed selects to control using increment type, and the motion pose increment of manpower is added to the end effector of current robot
On pose, realize that hand controls the guiding in real time of mechanical arm;
The location information of manpower and hip center is under Kinect coordinate system, to establish above-mentioned movement mapping relations, by it
Coordinate is transformed under robot coordinate system, and concrete methods of realizing is as follows:
S31: position vector of the right hand at Kinect coordinate system { K } is extracted from bone information stream SkeletonFrameKP1=
[x1,y1,z1]TAnd position vector of the hip center at { K }KP2=[x2,y2,z2]T;It obtains under Kinect coordinate system { K }
Position of operator's right hand relative to itself hip centerKP:
KP=KP1-KP2 (3)
S32: bone tracking coordinate system is all in Kinect camera coordinate system, and origin is infrared camera center, and Z axis is infrared
Camera optical axis, X-direction are horizontal direction, and Y-axis is vertical direction;To obtain coordinate of the mechanical arm tail end relative to pedestalBP, demand go out the spin matrix that Kinect camera coordinate system { K } is transformed into robot coordinate system { B }Kinect camera
Coordinate system { K } rotates -90 ° by the right-hand rule about the z axis, rotates -90 ° further around X-axis, obtains spin matrix
S33: coordinate of the mechanical arm tail end relative to pedestal is finally obtainedBP are as follows:
Using inverse transformation method, i.e., multiply one or several inverse-transform matrixs before transformation matrix, it is corresponding by comparing equation both sides
Element achievees the purpose that solve Inverse Kinematics Solution;As described above, obtaining mechanical arm tail end under basis coordinates system by formula (5)
Position vectorBP=[x, y, z]T, homogeneous transform matrix based on D-H parameters Mechanical arm ending coordinates system relative to base coordinate system
Are as follows:
Wherein, r11=c1c234c5+s1s5;r12=-c1c234c5+s1s5;r13=-c1s234;r21=s1c234c5-c1s5;r22=-
s1c234c5-c1s5;r23=-s1s234;r31=-s234c5;r32=s234c5;r33=-c234;px=c1(a2c2+a3c23-d5s234);py
=s1(a2c2+a3c23-d5s234);pz=-a2s2-a3s23+d1-d5c234.
S in formula234=sin (θ2+θ3+θ4)、c234=cos (θ2+θ3+θ4), other situations are similarly;
Kinematical equation are as follows:
The expression formula for finding out each joint angle of mechanical arm by algebraic approach by (6), (7) is as follows:
θ1=A tan2 (px,py) (8)
θ2=A tan2 (s2,c2) (9)
θ5=A tan2 (r11py-r21px,r12py-r22px) (12)
Wherein
M=(r13px+r23py+r33(pz-d1);
All joint angles of five degree-of-freedom manipulator are found out using algebraic approach.
5. the mechanical arm intersection control routine based on any one of Claims 1 to 4 the method, it is characterised in that: the system
Including Kinect sensor, computer software and mechanical arm;
Kinect sensor is used for the acquisition of human depth's image information;
The depth of field data that computer software is obtained with the processing of bone tracer technique, the 3D for establishing 20 skeleton points of human body are sat
Mark;
The control system of mechanical arm obtains the control instruction that computer system conversion issues, and control end effector follows hand to transport
It is dynamic, realize manpower to the guiding in real time of robot arm end effector.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910292316.5A CN109968310A (en) | 2019-04-12 | 2019-04-12 | A kind of mechanical arm interaction control method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910292316.5A CN109968310A (en) | 2019-04-12 | 2019-04-12 | A kind of mechanical arm interaction control method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109968310A true CN109968310A (en) | 2019-07-05 |
Family
ID=67084334
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910292316.5A Pending CN109968310A (en) | 2019-04-12 | 2019-04-12 | A kind of mechanical arm interaction control method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109968310A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110480634A (en) * | 2019-08-08 | 2019-11-22 | 北京科技大学 | A kind of arm guided-moving control method for manipulator motion control |
CN110509281A (en) * | 2019-09-16 | 2019-11-29 | 中国计量大学 | The apparatus and method of pose identification and crawl based on binocular vision |
CN110877335A (en) * | 2019-09-27 | 2020-03-13 | 华南理工大学 | Self-adaptive unmarked mechanical arm track tracking method based on hybrid filter |
CN111589098A (en) * | 2020-04-21 | 2020-08-28 | 哈尔滨拓博科技有限公司 | Follow-up gesture control method and system for doll with stepping crown block |
CN113070877A (en) * | 2021-03-24 | 2021-07-06 | 浙江大学 | Variable attitude mapping method for seven-axis mechanical arm visual teaching |
CN113742992A (en) * | 2021-04-19 | 2021-12-03 | 广东工业大学 | Master-slave control method based on deep learning and application |
WO2022141160A1 (en) * | 2020-12-30 | 2022-07-07 | 诺创智能医疗科技(杭州)有限公司 | Master-slave mapping method for parallel platform, and mechanical arm system and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103399637A (en) * | 2013-07-31 | 2013-11-20 | 西北师范大学 | Man-computer interaction method for intelligent human skeleton tracking control robot on basis of kinect |
CN104570731A (en) * | 2014-12-04 | 2015-04-29 | 重庆邮电大学 | Uncalibrated human-computer interaction control system and method based on Kinect |
CN107253192A (en) * | 2017-05-24 | 2017-10-17 | 湖北众与和智能装备科技有限公司 | It is a kind of based on Kinect without demarcation human-computer interactive control system and method |
-
2019
- 2019-04-12 CN CN201910292316.5A patent/CN109968310A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103399637A (en) * | 2013-07-31 | 2013-11-20 | 西北师范大学 | Man-computer interaction method for intelligent human skeleton tracking control robot on basis of kinect |
CN104570731A (en) * | 2014-12-04 | 2015-04-29 | 重庆邮电大学 | Uncalibrated human-computer interaction control system and method based on Kinect |
CN107253192A (en) * | 2017-05-24 | 2017-10-17 | 湖北众与和智能装备科技有限公司 | It is a kind of based on Kinect without demarcation human-computer interactive control system and method |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110480634A (en) * | 2019-08-08 | 2019-11-22 | 北京科技大学 | A kind of arm guided-moving control method for manipulator motion control |
CN110509281A (en) * | 2019-09-16 | 2019-11-29 | 中国计量大学 | The apparatus and method of pose identification and crawl based on binocular vision |
CN110877335A (en) * | 2019-09-27 | 2020-03-13 | 华南理工大学 | Self-adaptive unmarked mechanical arm track tracking method based on hybrid filter |
CN111589098A (en) * | 2020-04-21 | 2020-08-28 | 哈尔滨拓博科技有限公司 | Follow-up gesture control method and system for doll with stepping crown block |
WO2022141160A1 (en) * | 2020-12-30 | 2022-07-07 | 诺创智能医疗科技(杭州)有限公司 | Master-slave mapping method for parallel platform, and mechanical arm system and storage medium |
CN113070877A (en) * | 2021-03-24 | 2021-07-06 | 浙江大学 | Variable attitude mapping method for seven-axis mechanical arm visual teaching |
CN113070877B (en) * | 2021-03-24 | 2022-04-15 | 浙江大学 | Variable attitude mapping method for seven-axis mechanical arm visual teaching |
CN113742992A (en) * | 2021-04-19 | 2021-12-03 | 广东工业大学 | Master-slave control method based on deep learning and application |
CN113742992B (en) * | 2021-04-19 | 2024-03-01 | 广东工业大学 | Master-slave control method based on deep learning and application |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109968310A (en) | A kind of mechanical arm interaction control method and system | |
CN107253192A (en) | It is a kind of based on Kinect without demarcation human-computer interactive control system and method | |
US20210205986A1 (en) | Teleoperating Of Robots With Tasks By Mapping To Human Operator Pose | |
CN110480634B (en) | Arm guide motion control method for mechanical arm motion control | |
CN103271784B (en) | Man-machine interactive manipulator control system and method based on binocular vision | |
CN108972494A (en) | A kind of Apery manipulator crawl control system and its data processing method | |
CN104570731A (en) | Uncalibrated human-computer interaction control system and method based on Kinect | |
Zhang | Application of optical motion capture technology in power safety entitative simulation training system | |
Li et al. | Development of kinect based teleoperation of nao robot | |
CN109807887A (en) | Flexible arm Intellisense and control method and system based on deep neural network | |
Zhang et al. | A real-time upper-body robot imitation system | |
CN113829343A (en) | Real-time multi-task multi-person man-machine interaction system based on environment perception | |
Zhang et al. | Multi‐target detection and grasping control for humanoid robot NAO | |
Kofman et al. | Robot-manipulator teleoperation by markerless vision-based hand-arm tracking | |
Gulde et al. | RoPose: CNN-based 2D pose estimation of industrial robots | |
Khalil et al. | Visual monitoring of surface deformations on objects manipulated with a robotic hand | |
Zhou et al. | Imitating tool-based garment folding from a single visual observation using hand-object graph dynamics | |
Li et al. | A dexterous hand-arm teleoperation system based on hand pose estimation and active vision | |
Molet et al. | An architecture for immersive evaluation of complex human tasks | |
Bai et al. | Kinect-based hand tracking for first-person-perspective robotic arm teleoperation | |
Infantino et al. | Visual control of a robotic hand | |
Sharma et al. | Vision-driven tele-operation for robot arm manipulation | |
Jayasurya et al. | Gesture controlled AI-robot using Kinect | |
Hong et al. | Interactive humanoid robot arm imitation system using human upper limb motion tracking | |
CN112790760A (en) | Three-dimensional motion attitude capturing method, device, processing equipment and system |
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: 20190705 |