CN113805696A - Machine learning method based on surface electromyographic signals and dynamic capture technology - Google Patents

Machine learning method based on surface electromyographic signals and dynamic capture technology Download PDF

Info

Publication number
CN113805696A
CN113805696A CN202111018661.3A CN202111018661A CN113805696A CN 113805696 A CN113805696 A CN 113805696A CN 202111018661 A CN202111018661 A CN 202111018661A CN 113805696 A CN113805696 A CN 113805696A
Authority
CN
China
Prior art keywords
arm
motion
signals
mechanical arm
machine learning
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
Application number
CN202111018661.3A
Other languages
Chinese (zh)
Inventor
牛福永
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jianbingjian Intelligent Technology Beijing Co ltd
Original Assignee
Jianbingjian Intelligent Technology Beijing Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Jianbingjian Intelligent Technology Beijing Co ltd filed Critical Jianbingjian Intelligent Technology Beijing Co ltd
Priority to CN202111018661.3A priority Critical patent/CN113805696A/en
Publication of CN113805696A publication Critical patent/CN113805696A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
    • G06F3/015—Input arrangements based on nervous system activity detection, e.g. brain waves [EEG] detection, electromyograms [EMG] detection, electrodermal response detection
    • 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
    • B25J9/1628—Programme controls characterised by the control loop
    • B25J9/163—Programme controls characterised by the control loop learning, adaptive, model based, rule based expert control
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/243—Classification techniques relating to the number of classes
    • G06F18/2431—Multiple classes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/02—Preprocessing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/12—Classification; Matching

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Human Computer Interaction (AREA)
  • Evolutionary Computation (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Biomedical Technology (AREA)
  • Dermatology (AREA)
  • General Health & Medical Sciences (AREA)
  • Neurology (AREA)
  • Neurosurgery (AREA)
  • Health & Medical Sciences (AREA)
  • Robotics (AREA)
  • Mechanical Engineering (AREA)
  • Manipulator (AREA)
  • Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)

Abstract

The invention relates to the technical field of machine learning based on arm electromyographic signals, in particular to a machine learning method based on surface electromyographic signals and a dynamic capture technology, which comprises the following steps: s1, acquiring myoelectric signals of an arm of an operator and motion signals of an arm joint; s2, filtering the electromyographic signals collected in the step S1 and the motion signals of the arm joints; s3, inputting the filtered electromyographic signals and the motion signals of the arm joints into a support vector machine classifier, identifying the motion direction of the arm by a direction prediction model according to the electromyographic signals, and identifying a motion trail model of the tail end of the mechanical arm by a motion prediction model according to the motion signals of the arm joints; s4, calculating the expected position of the tail end of the mechanical arm by combining the motion track model of the tail end of the mechanical arm obtained in the S3 with the motion direction of the mechanical arm; and S5, sending the expected position of the tail end of the mechanical arm calculated in the S4 to the mechanical arm controller so as to control the movement of the mechanical arm. The method has the characteristics of real-time speed tracking and quick response.

Description

Machine learning method based on surface electromyographic signals and dynamic capture technology
Technical Field
The invention relates to the technical field of machine learning based on arm electromyographic signals, in particular to a machine learning method based on surface electromyographic signals and a dynamic capture technology.
Background
Existing machine learning techniques use a similar principle to a joystick, specifically, setting a starting point for the user's arm, and then if the arm is moved, the arm will start moving in the same direction at the same speed. If the user moves the arm in the opposite direction, the robotic arm will move in the same direction at a constant speed. The user can only control the direction of movement by a few variables because the speed is pre-programmed.
In view of the above, a machine learning method is needed to solve the problem that the existing robot arm can only move in the same direction at a constant speed.
Disclosure of Invention
The invention aims to provide a machine learning method based on a surface electromyogram signal and a motion capture technology, which realizes teleoperation control of a mechanical arm by combining the electromyogram signal and a motion capture signal with a machine learning method, and has the characteristics of real-time speed tracking and quick response.
The purpose of the invention is realized by the following technical scheme.
A machine learning method based on surface electromyography signals and a dynamic capture technology comprises the following steps:
s1, acquiring myoelectric signals of an arm of an operator and motion signals of an arm joint;
s2, filtering the electromyographic signals collected in the step S1 and the motion signals of the arm joints;
s3, inputting the filtered electromyographic signals and the motion signals of the arm joints into a support vector machine classifier, identifying the motion direction of the arm by a direction prediction model according to the electromyographic signals, and identifying a motion trail model of the tail end of the mechanical arm by a motion prediction model according to the motion signals of the arm joints;
s4, calculating the expected position of the tail end of the mechanical arm by combining the motion track model of the tail end of the mechanical arm obtained in the S3 with the motion direction of the mechanical arm;
and S5, sending the expected position of the tail end of the mechanical arm calculated in the S4 to a mechanical arm controller, and controlling the movement of the mechanical arm by the mechanical arm controller.
Preferably, in step S1, the myoelectric and motion capture sensor worn on the arm of the operator captures the arm motion of the operator in real time, and obtains the myoelectric signal of the arm and the motion signal of the arm joint of the operator.
Preferably, in step S2, the myoelectric signal and the motion signal of the arm joint are filtered by a kalman filter.
Preferably, in step S3, the support vector machine classifier obtains a direction prediction model by training electromyographic signal sample data, and obtains a motion prediction model by training motion signal sample data of an arm joint.
Preferably, in step S3, when the support vector machine classifier is used for training electromyographic signal sample data, the classification of multiple classes of electromyographic signals is realized by the following formula,
Figure RE-GDA0003346532900000021
s.t.yi((w.xi)+b)≥1-ξi,i＝1,...,l
ξi＞0,i＝1,...,l
wherein x isiIs a feature input vector, w is a weight vector, b is an offset;
combined use of radial basis kernel function (RBF):
K(x,y)＝exp(-γ||x-y||2),γ＞0
wherein x is a support vector, y is a vector to be classified, | | x-y | | calculation2Calculating a two-norm distance;
and forming a grid variable by using a grid search optimization method and a penalty factor C and a kernel function radius gamma, performing cross validation to calculate the accuracy, and testing according to the calculation and training results.
Preferably, the kernel function radius γ is 200, and the penalty factor C is 0.15.
Preferably, in step S3, the motion prediction model recognizing the motion trail model of the end of the mechanical arm according to the motion signal of the arm joint includes the following sub-steps:
s31, calculating the position increment of the flange at the tail end of the mechanical arm according to the motion signal of the arm joint by the motion prediction model;
and S32, the motion prediction model is used for obtaining a motion trail model of the tail end of the mechanical arm by superposing the position increment of the flange at the tail end of the mechanical arm according to the motion direction of the mechanical arm.
Preferably, in step S32, the mathematical model of the motion signal of the arm joint mapped to the position increment data of the arm end flange is as follows:
Figure RE-GDA0003346532900000031
wherein q is the operator arm joint motion angle measured by myoelectricity and motion capture sensors in real time, q0For the starting position of the arm of the operator,
Figure RE-GDA0003346532900000032
for the movement speed of the arm joints of the operator,
Figure RE-GDA0003346532900000033
j (q) is the jacobian matrix of the arm, the velocity of the arm end flange.
Preferably, in step S4, the formula for calculating the desired position of the end of the robot arm is:
Figure RE-GDA0003346532900000041
where x is the desired movement of the end flange of the robot arm, x0Is the starting position of the robot arm, vrThe movement direction of the arm is identified according to the electromyographic signals,
Figure RE-GDA0003346532900000042
the position increment of the end flange of the mechanical arm.
The invention has the beneficial effects that:
the invention provides a machine learning method based on surface electromyogram signals and a dynamic capture technology, which comprises the following steps: s1, acquiring myoelectric signals of an arm of an operator and motion signals of an arm joint; s2, filtering the electromyographic signals collected in the step S1 and the motion signals of the arm joints; s3, inputting the filtered electromyographic signals and the motion signals of the arm joints into a support vector machine classifier, identifying the motion direction of the arm by a direction prediction model according to the electromyographic signals, and identifying a motion trail model of the tail end of the mechanical arm by a motion prediction model according to the motion signals of the arm joints; s4, calculating the expected position of the tail end of the mechanical arm by combining the motion track model of the tail end of the mechanical arm obtained in the S3 with the motion direction of the mechanical arm; and S5, sending the expected position of the tail end of the mechanical arm calculated in the S4 to a mechanical arm controller, and controlling the movement of the mechanical arm by the mechanical arm controller. The method realizes the teleoperation control of the mechanical arm by combining the electromyographic signal and the motion capture signal with a machine learning method, and has the characteristics of real-time speed tracking and quick response.
Drawings
FIG. 1 is a flow chart of a machine learning method based on a surface electromyogram signal and a kinetic capture technology provided by the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
As shown in fig. 1, the machine learning method based on the surface electromyogram signal and the kinetic capture technology provided by the invention comprises the following steps:
and S1, acquiring myoelectric signals of the arm of the operator and motion signals of the arm joint.
As a preferred scheme, the real-time capture of the arm movement of an operator is realized through the myoelectric and motion capture sensors worn on the arm of the operator, and the myoelectric signal of the arm of the operator and the motion signal of the arm joint are obtained.
And S2, performing filtering processing on the myoelectric signals acquired in the step S1 and the motion signals of the arm joints.
As a preferred scheme, a Kalman filter is adopted to carry out filtering processing on the electromyographic signals and the motion signals of the arm joints, and noise and interference are removed. The existing wavelet transform, notch filtering and low-pass filtering are not good for electromyographic signals containing a large amount of aperiodic Gaussian noise and white noise. Therefore, the scheme provides myoelectricity and motion capture signal processing based on Kalman filtering according to the defects of the prior art, and the Kalman filter can well estimate the motion state of the arm of a person and effectively remove Gaussian noise and white noise.
S3, inputting the filtered electromyographic signals and the motion signals of the arm joints into a support vector machine classifier, identifying the motion direction of the arm by a direction prediction model according to the electromyographic signals, and identifying a motion trail model of the tail end of the mechanical arm by a motion prediction model according to the motion signals of the arm joints.
And S4, calculating the expected position of the tail end of the mechanical arm by combining the motion track model of the tail end of the mechanical arm obtained in the S3 with the motion direction of the mechanical arm.
And S5, sending the expected position of the tail end of the mechanical arm calculated in the step S4 to a mechanical arm controller, and controlling the motion of the mechanical arm by the mechanical arm controller to realize real-time tracking control on the motion of the human arm.
In this embodiment, as a preferable scheme, in the step S3, the support vector machine classifier obtains a direction prediction model by training electromyographic signal sample data, and obtains a motion prediction model by training motion signal sample data of an arm joint. When an operator wears the electro-myoelectricity and motion capture sensors on the arms, the direction prediction model can predict the motion direction of the arms in real time, and the motion prediction model can predict the motion of the arms in real time.
The machine learning method adopted by the scheme is a statistical-based small sample machine learning method, can solve the problem of nonlinear sample classification, can process a multi-feature high-latitude sample data set, has no local minimum problem, and has the advantage of strong generalization capability.
In this embodiment, as a preferable scheme, in step S3, when the support vector machine classifier is used to train the electromyographic signal sample data, the classification of multiple classes of electromyographic signals is realized by the following formula,
Figure RE-GDA0003346532900000061
s.t.yi((w.xi)+b)≥1-ξi,i＝1,...,l
ξi＞0,i＝1,...,l
wherein x isiIs a feature input vector, w is a weight vector, b is an offset;
combined use of radial basis kernel function (RBF):
K(x,y)＝exp(-γ||x-y||2),γ＞0
wherein x is a support vector, y is a vector to be classified, | | x-y | | calculation2Calculating a two-norm distance;
and forming a grid variable by using a grid search optimization method and a penalty factor C and a kernel function radius gamma, performing cross validation to calculate the accuracy, and testing according to the calculation and training results. The kernel function radius gamma is 200, and the penalty factor C is 0.15, so that the recognition rate is high.
In this embodiment, as a preferable scheme, in the step S3, the recognizing, by the motion prediction model, the motion trail model of the end of the mechanical arm according to the motion signal of the arm joint includes the following sub-steps:
s31, calculating the position increment of the flange at the tail end of the mechanical arm according to the motion signal of the arm joint by the motion prediction model;
and S32, the motion prediction model is used for obtaining a motion trail model of the tail end of the mechanical arm by superposing the position increment of the flange at the tail end of the mechanical arm according to the motion direction of the mechanical arm.
In this embodiment, as a preferable scheme, in step S32, the mathematical model of the motion signal of the arm joint mapped to the position increment data of the end flange of the arm is as follows:
Figure RE-GDA0003346532900000071
wherein q is the operator arm joint motion angle measured by myoelectricity and motion capture sensors in real time, q0For the starting position of the arm of the operator,
Figure RE-GDA0003346532900000072
for the movement speed of the arm joints of the operator,
Figure RE-GDA0003346532900000073
j (q) is the jacobian matrix of the arm, the velocity of the arm end flange.
In this embodiment, as a preferable scheme, in step S4, the formula for calculating the desired position of the end of the mechanical arm is as follows:
Figure RE-GDA0003346532900000074
where x is the desired movement of the end flange of the robot arm, x0Is the starting position of the robot arm, vrThe movement direction of the arm is identified according to the electromyographic signals,
Figure RE-GDA0003346532900000081
the position increment of the end flange of the mechanical arm.
The above are only typical examples of the present invention, and besides, the present invention may have other embodiments, and all the technical solutions formed by equivalent substitutions or equivalent changes are within the scope of the present invention as claimed.

Claims (9)

1. A machine learning method based on surface electromyogram signals and a dynamic capture technology is characterized by comprising the following steps:
s1, acquiring myoelectric signals of an arm of an operator and motion signals of an arm joint;
s2, filtering the electromyographic signals collected in the step S1 and the motion signals of the arm joints;
s3, inputting the filtered electromyographic signals and the motion signals of the arm joints into a support vector machine classifier, identifying the motion direction of the arm by a direction prediction model according to the electromyographic signals, and identifying a motion trail model of the tail end of the mechanical arm by a motion prediction model according to the motion signals of the arm joints;
s4, calculating the expected position of the tail end of the mechanical arm by combining the motion track model of the tail end of the mechanical arm obtained in the S3 with the motion direction of the mechanical arm;
and S5, sending the expected position of the tail end of the mechanical arm calculated in the S4 to a mechanical arm controller, and controlling the movement of the mechanical arm by the mechanical arm controller.
2. The method for machine learning based on surface electromyography and kinetic capture technology of claim 1, wherein in step S1, real-time capture of arm movement of the operator is realized by electromyography and motion capture sensors worn on the arm of the operator, and electromyography signals of the arm and motion signals of the arm joint of the operator are obtained.
3. The method for machine learning based on surface electromyography and kinetic capture technology of claim 1, wherein in step S2, a kalman filter is used to filter the electromyography signals and the motion signals of arm joints.
4. The method for machine learning based on surface electromyography and kinetic capture technology of claim 1, wherein in step S3, the support vector machine classifier obtains a direction prediction model by training electromyography sample data, and obtains a motion prediction model by training motion signal sample data of an arm joint.
5. The method for machine learning based on surface electromyography and kinetic capture technology of claim 4, wherein in step S3, when the SVM classifier trains the electromyography sample data, the classification of multiple classes of electromyography signals is achieved by the following formula,
Figure RE-FDA0003346532890000021
s.t.yi((w.xi)+b)≥1-ξi,i＝1,...,l
ξi＞0,i＝1,...,l
wherein x isiIs a feature input vector, w is a weight vector, b is an offset;
combined use of radial basis kernel function (RBF):
K(x,y)＝exp(-γ||x-y||2),γ＞0
wherein x is a support vector, y is a vector to be classified, | | x-y | | calculation2Calculating a two-norm distance;
and forming a grid variable by using a grid search optimization method and a penalty factor C and a kernel function radius gamma, performing cross validation to calculate the accuracy, and testing according to the calculation and training results.
6. The method for machine learning based on surface electromyography and kinetic trapping technology of claim 5, wherein the kernel radius γ is 200 and the penalty factor C is 0.15.
7. The surface electromyography signal and kinetic capture technology-based machine learning method of claim 1, wherein the step S3, the recognition of the mechanical arm tip motion trajectory model by the motion prediction model according to the motion signal of the arm joint comprises the following sub-steps:
s31, calculating the position increment of the flange at the tail end of the mechanical arm according to the motion signal of the arm joint by the motion prediction model;
and S32, the motion prediction model is used for obtaining a motion trail model of the tail end of the mechanical arm by superposing the position increment of the flange at the tail end of the mechanical arm according to the motion direction of the mechanical arm.
8. The surface electromyography and kinetic capture technique-based machine learning method of claim 7, wherein in step S32, the mathematical model of the mapping of the motion signals of the arm joints to the position increment data of the end flange of the mechanical arm is as follows:
Figure RE-FDA0003346532890000031
wherein q is the operator arm joint motion angle measured by myoelectricity and motion capture sensors in real time, q0For the starting position of the arm of the operator,
Figure RE-FDA0003346532890000032
for the movement speed of the arm joints of the operator,
Figure RE-FDA0003346532890000033
j (q) is the jacobian matrix of the arm, the velocity of the arm end flange.
9. The surface electromyography and kinetic capture technology-based machine learning method of claim 8, wherein in step S4, the formula for calculating the desired position of the mechanical arm tip is:
Figure RE-FDA0003346532890000034
where x is the desired movement of the end flange of the robot arm, x0Is the starting position of the robot arm, vrThe movement direction of the arm is identified according to the electromyographic signals,
Figure RE-FDA0003346532890000035
the position increment of the end flange of the mechanical arm.
CN202111018661.3A 2021-09-01 2021-09-01 Machine learning method based on surface electromyographic signals and dynamic capture technology Pending CN113805696A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111018661.3A CN113805696A (en) 2021-09-01 2021-09-01 Machine learning method based on surface electromyographic signals and dynamic capture technology

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111018661.3A CN113805696A (en) 2021-09-01 2021-09-01 Machine learning method based on surface electromyographic signals and dynamic capture technology

Publications (1)

Publication Number Publication Date
CN113805696A true CN113805696A (en) 2021-12-17

Family

ID=78894534

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111018661.3A Pending CN113805696A (en) 2021-09-01 2021-09-01 Machine learning method based on surface electromyographic signals and dynamic capture technology

Country Status (1)

Country Link
CN (1) CN113805696A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115990876A (en) * 2022-11-22 2023-04-21 燕山大学 Simulation mechanical arm control method based on multichannel electromyographic signals

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103190905A (en) * 2013-04-01 2013-07-10 武汉理工大学 Multi-channel surface electromyography signal collection system based on wireless fidelity (Wi-Fi) and processing method thereof
CN107553499A (en) * 2017-10-23 2018-01-09 上海交通大学 Natural the gesture motion control system and method for a kind of Multi-shaft mechanical arm
CN109375585A (en) * 2018-09-28 2019-02-22 湖南希法工程机械有限公司 Arm support tail end position control method and system and engineering machinery comprising the system
CN110238852A (en) * 2019-06-17 2019-09-17 湖南大学 A kind of robots arm's hand remote operating control method and system
CN111890363A (en) * 2020-07-27 2020-11-06 四川大学 Mechanical arm motion planning method based on rapid self-adaptive gradient neural network algorithm
NL2027179A (en) * 2020-03-05 2021-04-06 Nan Kai Univ Flexible-arm-oriented multi-modal human-machine interaction control method

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103190905A (en) * 2013-04-01 2013-07-10 武汉理工大学 Multi-channel surface electromyography signal collection system based on wireless fidelity (Wi-Fi) and processing method thereof
CN107553499A (en) * 2017-10-23 2018-01-09 上海交通大学 Natural the gesture motion control system and method for a kind of Multi-shaft mechanical arm
CN109375585A (en) * 2018-09-28 2019-02-22 湖南希法工程机械有限公司 Arm support tail end position control method and system and engineering machinery comprising the system
CN110238852A (en) * 2019-06-17 2019-09-17 湖南大学 A kind of robots arm's hand remote operating control method and system
NL2027179A (en) * 2020-03-05 2021-04-06 Nan Kai Univ Flexible-arm-oriented multi-modal human-machine interaction control method
CN111890363A (en) * 2020-07-27 2020-11-06 四川大学 Mechanical arm motion planning method based on rapid self-adaptive gradient neural network algorithm

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115990876A (en) * 2022-11-22 2023-04-21 燕山大学 Simulation mechanical arm control method based on multichannel electromyographic signals

Similar Documents

Publication Publication Date Title
DE102014101735A1 (en) METHOD AND DEVICE FOR CONTROLLING AN ROBOT DEVICE FROM SURGERY ON THE BODY
CN110605721A (en) Mechanical arm dragging teaching method based on terminal six-dimensional force sensor
Artemiadis et al. EMG-based teleoperation of a robot arm in planar catching movements using ARMAX model and trajectory monitoring techniques
CN109702740B (en) Robot compliance control method, device, equipment and storage medium
CN112276944A (en) Man-machine cooperation system control method based on intention recognition
CN111249005A (en) Puncture surgical robot compliance control system
WO2019202900A1 (en) Behavior estimation device, behavior estimation method, and behavior estimation program
CN113805696A (en) Machine learning method based on surface electromyographic signals and dynamic capture technology
CN111399640A (en) Multi-mode man-machine interaction control method for flexible arm
Dwivedi et al. A shared control framework for robotic telemanipulation combining electromyography based motion estimation and compliance control
CN114770461B (en) Mobile robot based on monocular vision and automatic grabbing method thereof
JP2558599B2 (en) Robot controller
CN116236328A (en) Visual-based intelligent artificial limb system capable of realizing natural grabbing
CN113478462B (en) Method and system for controlling intention assimilation of upper limb exoskeleton robot based on surface electromyogram signal
CN112214109B (en) Myoelectricity and gesture data-based composite control method, device and system
Zhao et al. POSITIONING AND GRABBING TECHNOLOGY OF INDUSTRIAL ROBOT BASED ON VISION.
CN109885159B (en) State space electromyography model construction method based on forward dynamics and Hill model
CN111358659B (en) Robot power-assisted control method and system and lower limb rehabilitation robot
CN110103226B (en) Auxiliary robot control method and system
CN109676583B (en) Deep learning visual acquisition method based on target posture, learning system and storage medium
Chen et al. Hand tracking accuracy enhancement by data fusion using leap motion and myo armband
Carlevarino et al. An incremental growing neural network and its application to robot control
Zhou et al. Visual servo control system of 2-DOF parallel robot
CN114714358A (en) Method and system for teleoperation of mechanical arm based on gesture protocol
Shanshan IMPLEMENTATION OF INTELLIGENT ROBOT CONTROL ALGORITHM BASED ON VISUAL SERVO CONTROL.

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