Summary of the invention
In view of the problems of the existing technology, the present invention provides a kind of wearable smart machine controlling parties that auxiliary is carried
Method.
The invention is realized in this way a kind of control method for the wearable smart machine that auxiliary is carried, including:
Step 1: multi-channel surface myoelectric signal when index finger, middle finger, nameless activity is obtained, using Hidden Markov
Model obtains the wavelet coefficient of finger electromyography signal to surface electromyogram signal wavelet decomposition, which is utilized greatest hope
Algorithm training, using gauss hybrid models, it is assumed that all wavelet coefficients are same distributions and have identical in same scale
State-transition matrix, the parameters of Hidden Markov Model are obtained by EM algorithm;
Step 2: obtain index finger, middle finger, nameless corresponding Hidden Markov Model parameter after, with removal
The wavelet coefficient of noise reconstructs the multi-channel surface myoelectric signal after being filtered;
Step 3: obtaining myoelectricity integral strength and myoelectrical activity space by the characteristic parameter of multi-channel surface myoelectric signal
The value indicative matrix decomposition is individual factor matrix Z and action mode matrix X, action mode by the multidimensional characteristic value matrix of distribution
Input of the matrix X as pattern recognition classifier device indicates eigenvalue matrix y using symmetrical bilinear modelk=zTWkx;
In formula:zTWhat is indicated is individual factor part, and what x was indicated is action mode part, and Wk belongs to bilinear model
Coefficient matrix;
Defined feature value matrix
In formula:That indicate is u-th of subject
Execute multidimensional characteristic value matrix when m movement n-th;
Step 4: obtaining eigenvalue matrix y of the new user under some movement, the action mode square of bilinear model is utilized
Battle array mean value and coefficient matrix, calculate new individual subscriber factor matrix:
Z=[[WX] [WX]CV]+ycv
By the surface electromyogram signal eigenvalue matrix y under different action modes, obtaining action mode matrix part x is:
X '=[[WCVz]CV]+y′
Step 5: the musculus extensor digitorum entirety myoelectricity when index finger extracted with flexible electrode array, middle finger, nameless activity is believed
Number establish surface myoelectric amplitude, myoelectrical activity spatial distribution characteristic matrix bilinear model, carry out the knowledge of finger force level
Not;
Step 6: obtain arm support portion faces electromyography signal, choose the absolute average A of electromyography signal, variance S,
Three characteristic parameters of average frequency constitute the input vector of direction of error Propagation Neural NetworkIt is obtained after training
Neural network parameter matrix W1And W2, use W1And W2Calculate the estimated value F of arm support position muscular exertion sizee;
Step 7: input variable and output variable are blurred, arm support position muscular exertion size FeAccording to numerical value
Fuzzy language is set as several grades, the corresponding practical grip size F in arm support position by sizehFuzzy language also set
For several grades, for output variable, motor speed S fuzzy language is set as several grades;
Step 8: the movement of arm support booster parts is triggered according to the grade of arm support position muscular exertion size, it should
Passive control of the action signal triggering finger booster parts of arm support booster parts to finger;
Step 9: obtain lumbar surface electromyography signal, calculates waist muscle and exert oneself the estimated value of size, arm support power-assisted
Component actuation signal triggers the movement of waist booster parts;
Step 10: leg booster parts are using single-degree-of-freedom exoskeleton system by wearer and leg power-assisted driving unit one
It rises and ectoskeleton kinetic moment is provided, the output torque of motor is obtained according to ectoskeleton self information, and the output torque of motor is Ta=
(1-α-1)G′(q)。
Further, after step 6 obtains arm support portion faces electromyography signal, using Hidden Markov Model to surface
Electromyography signal wavelet decomposition obtains the wavelet coefficient of finger electromyography signal;
By the wavelet coefficient using EM algorithm training, using gauss hybrid models, it is assumed that the institute in same scale
Wavelet coefficient is same distribution and has identical state-transition matrix;
The parameters of Hidden Markov Model are obtained by EM algorithm, obtain corresponding Hidden Markov Model
After parameter, the arm support portion faces electromyography signal after being filtered is reconstructed with the wavelet coefficient of removal noise.
Further, the absolute average of electromyography signal
The variance of electromyography signalAverage frequency
In formula, xijFor the numerical values recited of j-th of sampled point in i-th of timeslice,For signal in i-th of timeslice
Average value, fjFor Frequency point discrete on i-th of timeslice power spectrum, P (fj) it is discrete point in frequency fjCorresponding power, each
There is n sampled point in timeslice.
Further, lumbar surface electromyography signal is obtained, the absolute average, variance, average frequency three of electromyography signal are chosen
A characteristic parameter constitutes the input vector of direction of error Propagation Neural Network, show that neural network parameter matrix calculates hand after training
The estimated value of arm support zone muscular exertion size;
Input variable and output variable are blurred, waist muscle size of exerting oneself sets fuzzy language according to numerical values recited
Fuzzy language for several grades, the corresponding practical grip size of waist is also set to several grades, for output variable, motor
Revolving speed fuzzy language is set as several grades.
Text of the invention will be applied in the de-noising filtering processing of electromyography signal based on wavelet domain concealed Markov model method,
Using Hidden Markov Model is constructed the characteristics of the aggregation and duration of wavelet coefficient between adjacent scale, estimated using Bayes
Meter obtains the wavelet coefficient of actual signal, effectively removes noise by signal reconstruction, by utilizing Short Time Fourier Transform
Thought carries out timeslice segmentation to signal, and has chosen several representative electromyography signals to the signal analysis in timeslice
Parameter, and processing is analyzed it with the neural network after training, and then estimate corresponding muscular exertion size.Simultaneously will
The resulting power haptic signal of myoelectricity force-touch sensor and muscular exertion high low signal input fuzzy controller, and driving motor turns
Dynamic speed triggers arm support booster parts to control grip size, using the grade of arm support position muscular exertion size
Movement, passive control of the action signal triggering finger booster parts of the arm support booster parts to finger, arm support help
Power component actuation signal triggers the movement of waist booster parts, realizes that the action logic of booster parts is more scientific and reasonable, can be bigger
The power-assisted effect of booster parts is realized in degree.
Specific embodiment
In order to further understand the content, features and effects of the present invention, the following examples are hereby given, and cooperate attached drawing
Detailed description are as follows.
Structure of the invention is explained in detail with reference to the accompanying drawing.
A kind of control method for the wearable smart machine that auxiliary is carried, including:
Multi-channel surface myoelectric signal when S101, acquisition index finger, middle finger, nameless activity, using Hidden Markov mould
Type obtains the wavelet coefficient of finger electromyography signal to surface electromyogram signal wavelet decomposition, which is calculated using greatest hope
Method training, using gauss hybrid models, it is assumed that all wavelet coefficients are same distributions and have identical in same scale
State-transition matrix obtains the parameters of Hidden Markov Model by EM algorithm;
S102, obtain index finger, middle finger, nameless corresponding Hidden Markov Model parameter after, made an uproar with removal
The wavelet coefficient of sound reconstructs the multi-channel surface myoelectric signal after being filtered;
As shown in Fig. 2, entire electromyography signal filtering includes wavelet decomposition, training Hidden Markov Model (the maximum phase
Hope value-based algorithm), four part of Bayesian Estimation and wavelet reconstruction, this method does not need any free parameter undetermined, has fine
Adaptivity, the noise in electromyography signal can be effective filtered out and remain the detailed information in signal.
S103, myoelectricity integral strength and myoelectrical activity space point are obtained by the characteristic parameter of multi-channel surface myoelectric signal
The value indicative matrix decomposition is individual factor matrix Z and action mode matrix X, action mode square by the multidimensional characteristic value matrix of cloth
Input of the battle array X as pattern recognition classifier device, indicates eigenvalue matrix using symmetrical bilinear model:
yk=zTWkx;
In formula:zTWhat is indicated is individual factor part, and what x was indicated is action mode part, and Wk belongs to bilinear model
Coefficient matrix;
Defined feature value matrix
In formula:That indicate is u-th of subject
Execute multidimensional characteristic value matrix when m movement n-th;
S104, eigenvalue matrix y of the new user under some movement is obtained, utilizes the action mode matrix of bilinear model
Mean value and coefficient matrix calculate new individual subscriber factor matrix:
Z=[[WX] [WX]CV]+ycv
By the surface electromyogram signal eigenvalue matrix y under different action modes, obtaining action mode matrix part x is:
X '=[[WCVz]CV]+y′
S105, the musculus extensor digitorum entirety electromyography signal with the index finger of flexible electrode array extraction, middle finger, the third finger when movable
Establish surface myoelectric amplitude, myoelectrical activity spatial distribution characteristic matrix bilinear model, carry out the identification of finger force level;
The steady section sEMG signal subsection by filtering processing is calculated into root mean square (time window H=256 in MATLAB
Sampled point, each time window do not overlap), it is segmented myoelectrical activity intensity of the average value as current record channel of root mean square, with hand
Refer to that the percentage of maximal voluntary contractile force amount indicates;
S106, arm support portion faces electromyography signal is obtained, chooses the absolute average A of electromyography signal, variance S, puts down
Equal three characteristic parameters of frequency constitute the input vector of direction of error Propagation Neural NetworkIt must be spellbound after training
Through network paramter matrix W1And W2, use W1And W2Calculate the estimated value Fe of arm support position muscular exertion size;
The absolute average of electromyography signal
The variance of electromyography signalAverage frequency
In formula, xijFor the numerical values recited of j-th of sampled point in i-th of timeslice,For signal in i-th of timeslice
Average value, fjFor Frequency point discrete on i-th of timeslice power spectrum, P (fj) it is discrete point in frequency fjCorresponding power, each
There is n sampled point in timeslice.
After obtaining arm support portion faces electromyography signal, using Hidden Markov Model to the small wavelength-division of surface electromyogram signal
Solve the wavelet coefficient of finger electromyography signal;By the wavelet coefficient using EM algorithm training, using Gaussian Mixture mould
Type, it is assumed that all wavelet coefficients are same distributions and have identical state-transition matrix in same scale;By the maximum phase
Algorithm is hoped to obtain the parameters of Hidden Markov Model, after obtaining the parameter of corresponding Hidden Markov Model, with removal
The wavelet coefficient of noise reconstructs the arm support portion faces electromyography signal after being filtered.
Estimate that the firmly degree of measured's arm, first design are real by the neural network of training error backpropagation
It tests, myoelectricity acquisition electrode is fitted on the musculus flexor carpi ulnaris position of measured, while finger pressing when measured's wrist flexion being allowed to survey
Power device can measure the firmly size F of measured's wrist muscle while acquiring electromyography signal in this way.
S107, input variable and output variable are blurred, arm support position muscular exertion size FeIt is big according to numerical value
It is small that fuzzy language is set as several grades, the corresponding practical grip size F in arm support positionhFuzzy language be also set to
Several grades, for output variable, motor speed S fuzzy language is set as several grades;
The output variable of fuzzy controller is that the speed S of arm closure namely arm support booster parts act motor
The locked-rotor torque of revolving speed, motor speed and motor is directly proportional, so realizing arm grip indirectly by the closing speed of arm
Control.
Input variable and output variable are blurred, wherein being directed to input variable, muscular exertion size FeFuzzy language is set
It is set to attonity, small, smaller, larger, 5 grades big;
The practical grip size F of armhFuzzy language be defined as it is 4 grades small, smaller, larger, big;
For output variable, motor speed S fuzzy language is set as quickly opening, middling speed is opened, open at a slow speed, attonity, at a slow speed
It closes, middling speed is closed, quickly 7 grades of pass, positive value indicate that motor rotates forward, i.e. the steering of arm closure, negative value expression motor reversal, i.e. arm
The steering of opening.
S108, the movement of arm support booster parts, the hand are triggered according to the grade of arm support position muscular exertion size
Arm supports passive control of the action signal triggering finger booster parts of booster parts to finger;
For example, triggering finger booster parts movement when the practical grip of arm is larger can be set;
S109, lumbar surface electromyography signal is obtained, calculates waist muscle and exerts oneself the estimated value of size, arm support power-assisted portion
Part action signal triggers the movement of waist booster parts;
For example, triggering waist booster parts movement when the practical grip of arm is smaller can be set;
Lumbar surface electromyography signal is obtained, using processing method identical with arm support portion faces electromyography signal, choosing
Three absolute average of taking electromyographic signal, variance, average frequency characteristic parameters constitute the defeated of direction of error Propagation Neural Network
Incoming vector show that neural network parameter matrix calculates the estimated value of arm support position muscular exertion size after training;
Input variable and output variable are blurred, waist muscle size of exerting oneself sets fuzzy language according to numerical values recited
Fuzzy language for several grades, the corresponding practical grip size of waist is also set to several grades, for output variable, motor
Revolving speed fuzzy language is set as several grades.
Compared with existing control mode, the present invention is by following advantages:
The practical grip that arm myoelectricity is measured is small and practical grip that finger myoelectricity is measured is big, illustrates that this thing only has finger dynamic
Work, arm and attonity or movement very little, further explanation does not need arm at this time too big grip, so arm powered portion
Part does not need to act, once and the practical grip measured of practical grip and finger myoelectricity that arm myoelectricity is measured is mostly big, explanation
Arm powered component needs to act, once arm powered component needs to act, then waist certainty stress, excites waist power-assisted at this time
The action logic of component actuation, booster parts is more scientific and reasonable, can realize the power-assisted effect of booster parts to a greater extent.
S110, leg booster parts using single-degree-of-freedom exoskeleton system by wearer and leg power-assisted driving unit together
Ectoskeleton kinetic moment is provided, the output torque of motor is obtained according to ectoskeleton self information, and the output torque of motor is Ta=(1-
α-1) G ' (q), α is the angle between thigh and vertical direction in formula.
In other drive forces, when only wearer provides torque, T=T at this timehw, q=G (T), sensitivity system
Number is
Text of the invention will be applied in the de-noising filtering processing of electromyography signal based on wavelet domain concealed Markov model method,
Using Hidden Markov Model is constructed the characteristics of the aggregation and duration of wavelet coefficient between adjacent scale, estimated using Bayes
Meter obtains the wavelet coefficient of actual signal, effectively removes noise by signal reconstruction, by utilizing Short Time Fourier Transform
Thought carries out timeslice segmentation to signal, and has chosen several representative electromyography signals to the signal analysis in timeslice
Parameter, and processing is analyzed it with the neural network after training, and then estimate corresponding muscular exertion size.Simultaneously will
The resulting power haptic signal of myoelectricity force-touch sensor and muscular exertion high low signal input fuzzy controller, and driving motor turns
Dynamic speed triggers arm support booster parts to control grip size, using the grade of arm support position muscular exertion size
Movement, passive control of the action signal triggering finger booster parts of the arm support booster parts to finger, arm support help
Power component actuation signal triggers the movement of waist booster parts, realizes that the action logic of booster parts is more scientific and reasonable, can be bigger
The power-assisted effect of booster parts is realized in degree.
The above is only the preferred embodiments of the present invention, and is not intended to limit the present invention in any form,
Any simple modification made to the above embodiment according to the technical essence of the invention, equivalent variations and modification, belong to
In the range of technical solution of the present invention.