CN103876867A - Fuzzy estimation method for initial article grabbing reference force of hand prosthesis - Google Patents
Fuzzy estimation method for initial article grabbing reference force of hand prosthesis Download PDFInfo
- Publication number
- CN103876867A CN103876867A CN201310329353.1A CN201310329353A CN103876867A CN 103876867 A CN103876867 A CN 103876867A CN 201310329353 A CN201310329353 A CN 201310329353A CN 103876867 A CN103876867 A CN 103876867A
- Authority
- CN
- China
- Prior art keywords
- contact force
- prosthetic hand
- gradient
- fuzzy
- max
- 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.)
- Granted
Links
Images
Abstract
A fuzzy estimation method for initial article grabbing reference force of hand prosthesis includes the steps: 1) assuming that the hand prosthesis grabs articles different in softness at a certain speed before touching; 2) measuring contact force by the aid of a force sensor mounted at a knuckle when a finger of the hand prosthesis contacts with the articles; 3) obtaining a touch force gradient by differential of the contact force F within a short time when the finger contacts with the articles; 4) subjecting the contact force and the contact force gradient to normalization, fuzzification, fuzzy reasoning and defuzzification to obtain a fuzzy estimation value, wherein a specific rule fuzzy rule base is designed; 7) obtaining an expected grabbing force by means of coordinate inversion transformation of the fuzzy estimation value; 8) taking the maximum expected force calculated within a short time (50-300ms) after the hand prosthesis contacts with the articles as the initial article grabbing reference force of the hand prosthesis. The fuzzy estimation method has the advantages that estimation of the initial reference force of the hand prosthesis grabbing the articles different in softness is evident in differentiation and estimation effects are favorable.
Description
Technical field
The present invention relates to a kind of prosthetic hand and grasp object initial reference power blur estimation method.
background technology
The essential distinction of the mankind and animal is that people not only can use instrument, can also create instrument, reforming world, and this has benefited from their that a pair of and have many fingers hands of flexible operating function to a great extent.For extremity disabled persons, the forfeiture of hands not only makes its body and mind be hit, and also makes troubles to life simultaneously.Therefore, the research of prosthetic hand has very important significance for the life that improves upper limb disabled person.
In daily life, the manipulation environment of staff is dynamically non-good often, and manipulating objects complexity is various, and this just requires prosthetic hand better than standard machinery hands adaptability and motility, and function is more powerful.Prosthetic hand will be realized flexible crawl and manipulation as staff, and the design of its control system and the research of control strategy seem particularly important.Staff, in the time grasping object, is first seen and is grasped object by eyes, rule of thumb estimates the size of grip, then by nervous system control staff driver one skeletal muscle, realizes and becomes the different object of rigidity grasping soft or hard.After prosthetic hand complete design, structure, rigidity determines, and people's brain intention at present can't complete decoding, prosthetic hand can not be realized as staff become rigidity to grasp the object of different soft and hard.
For the design of prosthetic hand control system, be mainly that the electromyographic signal (EMG) by gathering on people with disability's deformed limb is carried out feature extraction to it at present, thereby to difference, pattern recognition is carried out in action, completes exercises to control prosthetic hand.For the crawl control of prosthetic hand, adopt force tracking control more, i.e. prior given expectation grip or be multiplied by a proportionality coefficient by electromyographic signal and obtain expected force, then CONTROLLER DESIGN, follows the tracks of control to it.
But in the time capturing the different object of soft or hard degree, as can not the suitable expectation grip of given prosthetic hand, just likely cause object landing or object to damage; The unstable uncertainty that also can cause expected force of electromyographic signal.By multiple different expected force is set in advance, in concrete crawl task, the grasping that is realized soft or hard degree different objects by change-over switch is a kind of solution, but this needs visual information to hold to be grabbed the feature of object, needs manual switchover pattern simultaneously.People are desirably in the dynamic process that prosthetic hand contacts with unknown object, estimate to expect grip, to realize the grasping control to different soft and hard object by the dynamic Contact force characteristic of moment.
Summary of the invention
In order to prevent that prosthetic hand from expecting that too small the causing of grip captures unstablely when object capturing, or the excessive object that causes of grip damages distortion, the invention provides a kind of different soft and hard object that adapts to, and prosthetic hand captures the blur estimation method of the initial grip of object.
The technical solution adopted in the present invention is:
A kind of different soft and hard object that adapts to, prosthetic hand captures the blur estimation method of the initial grip of object, comprises the following steps:
1) set prosthetic hand finger and capture the different object of soft or hard degree with certain speed
2) while touching object, point the pressure transducer of dactylus measure prosthetic hand finger and grabbed the actual contact force F of object by being installed on, described actual contact force has referred to while pointing firm and object contact, the contact force that dactylus upper sensor measures;
3) reality grasping contact force F is carried out to differential and obtains contact force gradient dF:
dF＝(F
k-F
k-1)/T (1)
Wherein: k is greater than 0 natural number; F
kfor the contact force value of current sensor measurement; F
k-1it is the grip value of a upper moment sensor measurement; T is the interval of adjacent twice measurement, i.e. sampling time.
4) contact force F and contact force gradient dF are normalized and obtain f and df, make its domain be [1,1]:
f＝[2F-(F
max+F
min)]/[2(F
max-F
min)] (2)
df＝[2dF-(dF
max+dF
min)]/[2(dF
max-dF
min)] (3)
Wherein: F
max, F
minrepresent respectively maximum and the minima of contact force; DF
max, dF
minrepresent respectively maximum and the minima of contact force gradient.
5) contact force f and contact force gradient df are multiplied by respectively quantizing factor k
1, k
2, obtain the input E of fuzzy system
fwith R
f:
6) by E
fwith R
fcarry out obfuscation, fuzzy reasoning and defuzzification, obtain blur estimation value u
f.Wherein, input E
f, R
fall can adopt and hang bell, triangular function, trapezoidal, Gaussian function or other functional forms with the membership function of output uf; Linguistic variable after contact force, contact force gradient and output obfuscation is: very energetically or very large gradient (PL), energetically or large gradient (PM), more energetically or larger gradient (PS), moderate force or middle constant gradient (ZR), less power or less gradient (NS), little power or little gradient (NM), very little power or very little gradient (NL); Design 49 fuzzy rules and formed specific fuzzy rule base; Can also increase according to specific needs or reduce linguistic variable progression and fuzzy rule number; Fuzzy reasoning mechanism can adopt Madamni inference mechanism, Larsen inference mechanism, Zadeh inference mechanism etc.; Defuzzification can adopt height method, centroid method, area-method, weighted mean method or additive method;
7) by blur estimation value u
fcarry out coordinate transform and make its codomain for [0,1], and be multiplied by proportionality coefficient k
3, obtain expecting grip f
d:
f
d＝k
3(u
f+1) (4)
8) get the maximum expectation grip f calculating in (50-300ms) in the prosthetic hand contact object a bit of time afterwards
dinitial reference power while capturing object as prosthetic hand.
The present invention proposes in the time that prosthetic hand grasps different soft and hard object, can in the time of initial contact, judge according to contact force and contact force gradient the crawl reference load size that adapts to rigidity.Its core concept is: when the finger of prosthetic hand contacts object with certain speed, due to the difference of object rigidity, the initial size of contact force and the variation speed of contact force (gradient of power) are different, so by after the gradient-norm gelatinizing of initial contact force and contact force, according to the rule base of setting, by fuzzy reasoning, defuzzification obtains the initial reference power of prosthetic hand grasping object.Described reference load can adapt to the object that soft or hard is different, obtains larger grasping reference load for hard object through blur estimation, obtains less grasping reference load for soft object through blur estimation.Described fuzzy logic is estimated reference load system, adopts graphic-arts technique, can derive the mathematic(al) representation of grip blur estimation model:
f
d＝k
3(u
f+1) (5)
g(σ)＝k(1-γ)+γσ (7)
σ＝E
f+R
，＝k
1f+k
2df (8)
Wherein u
f---fuzzy logic system output valve
K
1, k
2---fuzzy logic system input scale factor
K
3---fuzzy logic system output-scale-factor
Sgn (*)---sign function, its value for-1,0 or 1 respectively corresponding independent variable be negative value, zero or on the occasion of
F---the contact force after normalization
Df---the contact force gradient after normalization
K, γ---be the nonlinear function relevant with fuzzy input.
Beneficial effect of the present invention is mainly manifested in: the grasping object reference load blur estimation strategy of proposition, adopts starting stage contact force and contact force gradient to be input to fuzzy logic system and obtain prosthetic hand and initially grasp reference load.This reference load can adapt to the object that soft or hard is different, thereby makes prosthetic hand submissiveer, steady in the time capturing object.
Accompanying drawing explanation
Fig. 1 is that prosthetic hand of the present invention grasps control strategy.
Fig. 2 is contact force and the contact force gradient that prosthetic hand of the present invention grasps different-stiffness object.
Fig. 3 is that prosthetic hand of the present invention grasps object initial reference power blur estimation model.
Fig. 4 is blur estimation system input and output triangular form membership function of the present invention.
Fig. 5 is blur estimation system input and output Gaussian membership function of the present invention.
Fig. 6 is blur estimation system convention of the present invention storehouse.
Fig. 7 is that prosthetic hand of the present invention grasps hard object initial reference power estimation experimental result.
Fig. 8 is that prosthetic hand of the present invention grasps moderate stiffness object initial reference power estimation experimental result.
Fig. 9 is that prosthetic hand of the present invention grasps soft object initial reference power estimation experimental result.
the specific embodiment
Below in conjunction with accompanying drawing, the present invention is described further.
After prosthetic hand complete design, structure, rigidity determines, and people's brain intention at present can't complete decoding, prosthetic hand can not be realized as the staff become rigidity to grasp object.So, adopt ACTIVE CONTROL to realize the function of prosthetic hand grasping different-stiffness object.Based on the control principle of people's grasped, propose prosthetic hand and grasped control strategy, as shown in Figure 1, the grip that wherein F is prosthetic hand, F
dfor the grip of expecting.According to the metrical information of grip, replace the estimation of the grip of human brain vision with grip blur estimation model, replace people's cerebral nerve control with fuzzy controller, thereby make prosthetic hand can realize submissive, crawl stably.
When prosthetic hand finger captures the different object of rigidity with certain speed, finger and object contact moment, the dynamic property that contact force between the two shows is different.Take dixie cup, carafe and three kinds of daily life familiar object of mobile phone with metal shell as example, represent respectively soft object, the object of moderate stiffness object and hard three kinds of different-stiffness of object.Driving prosthetic hand to point with certain voltage grasps these three kinds of objects, get grip and gradient thereof as analytic target, as shown in Figure 2, when prosthetic hand finger captures different-stiffness object, the gradient of grip and grip has obvious difference, to rigidly connect tactile moment particularly outstanding.Thus, get the gradient of grip and grip as the dynamic characteristic of contact force, as the estimation foundation of initial reference power.
As Fig. 3, according to the present invention, content design prosthetic hand grasps object initial reference power blur estimation strategy, and the detailed process that control algolithm and parameter thereof are formulated is as follows:
The first step: content of the present invention is mentioned, the enforcement of Cooperation Strategy, must point dactylus setting pressure sensor at prosthetic hand, and grip measured sensor and gradient thereof are carried out to fuzzy reasoning, obtains estimating reference load.Before estimating reference load, must set up obfuscation and reverse gelatinizing method, fuzzy rule base and fuzzy reasoning mechanism by the characteristic that grasps contact force so.As Fig. 2, prosthetic hand finger is captured to different-stiffness object and carry out many experiments, measure grasping contact force by pressure transducer, and grip is got to differential obtain power gradient:
dF＝(F
k-F
k-1)/T (1)
Described grip refers to the prosthetic hand finger grasping contact force measured with the dactylus of object contact at first.
Second step: from many experiments result, get the maximum F of contact force
maxwith minima F
min, the maximum dF of contact force gradient
maxwith minima dF
min, the normalized parameter when estimating with reference to power.Normalized object is to make contact force and the unification of contact force gradient in domain [1,1], is convenient to quantification and the obfuscation of input.Normalization formula is as follows:
f＝[2F-(F
max+F
min)]/[2(F
max-F
min)] (2)
df＝[2dF-(dF
max+dF
min)]/[2(dF
max-dF
min)] (3)
Wherein: F
max, F
minrepresent respectively maximum and the minima of contact force; DF
max, dF
minrepresent respectively maximum and the minima of contact force gradient.
The 3rd step: the membership function of determining blur estimation input and output.Because contact force and contact force gradient have been carried out to normalization in second step, unified in domain domain [1,1], so, input membership function domain should be also [1,1],, as Fig. 4, the type of membership function can be selected triangular function form, as Fig. 5, also can select Gaussian, or the membership function that adopts other functional form, the present invention to test input and output has all adopted the triangular function form of standard.
The 4th step: set up fuzzy rule base.In daily life, it is various that staff touches object complexity, and their range in stiffness is very wide, and the object that cannot may touch each is tested.So the present invention has chosen three kinds of representational objects: dixie cup (soft object), carafe (moderate stiffness object), mobile phone (hard object), has carried out many groups of experiments.In order to distinguish better the object of different-stiffness, estimate suitable reference load, the linguistic variable after contact force, contact force gradient and output obfuscation is expanded to seven grades: very energetically or very large gradient (PL), energetically or large gradient (PM), more energetically or larger gradient (PS), moderate force or middle constant gradient (ZR), less power or less gradient (NS), little power or little gradient (NM), very little power or very little gradient (NL).Known with reference to Fig. 2, in the time that object rigidity is larger, the gradient of contact force and power is also larger, now should adopt larger reference load to realize stable crawl; When the rigidity of object hour, the gradient of contact force and power is also less, now should adopt less reference load to realize submissive crawl, prevents object to grab bad.Based on this, as Fig. 6, design 49 fuzzy rules and formed specific fuzzy rule base, can also increase according to specific needs linguistic variable progression and fuzzy rule number.
The 5th step: fuzzy reasoning mechanism and defuzzification, fuzzy reasoning mechanism can adopt Madamni inference mechanism, Larsen inference mechanism, Zadeh inference mechanism etc., and defuzzification can adopt height method, centroid method, area-method, weighted mean method or additive method
Can see in this strategy, the foundation of blur estimation model need to be take the dynamic characteristic of contact force as foundation, and contact force dynamic characteristic experiment directly affects the accuracy of estimating reference load.
In order to verify that the prosthetic hand of above-mentioned proposition grasps the effectiveness of object initial reference power blur estimation strategy, to build prosthetic hand initial reference power and estimate experimental program, this prosthetic hand drives thumb and forefinger to move by direct current generator.Select three kinds of different-stiffness objects: hard cup, carafe, can paper cup, grasp object as prosthetic hand, to check blur estimation strategy estimate suitable initial reference power to different-stiffness object.In experiment, drive prosthetic hand finger motion with constant voltage, finger carries out reference load estimation with the front 200ms of object contact.Experimental result, as Fig. 7 to Fig. 9, is respectively hard cup, the estimation reference load result of carafe and paper cup, and prosthetic hand grasps hard cup, and the blur estimation reference load of carafe and paper cup is about respectively: 2.3N, 1.4N, 0.9N; Prosthetic hand carries out force tracking grasping with this reference load, can guarantee that soft object (paper cup) does not damage, and can realize again the stable crawl to different-stiffness object.Result shows: the prosthetic hand that the present invention proposes grasps object initial reference power blur estimation strategy, can realize fast the initial reference power of different soft and hard object is estimated, the reference load estimating can guarantee that soft object does not damage, and realizes the stable grasping to different soft and hard object.
Claims (3)
1. prosthetic hand grasps an object initial reference power blur estimation method, it is characterized in that: described grasping object initial reference power blur estimation method comprises the following steps:
Set prosthetic hand finger and capture the different object of soft or hard degree with certain speed;
Measure prosthetic hand finger and grabbed the actual contact force F of object by being installed on the pressure transducer of finger dactylus;
Reality grasping contact force F is carried out to differential and obtain contact force gradient dF;
Contact force F and contact force gradient dF are normalized and obtain f and df, make its domain be [1,1];
Contact force f and contact force gradient df are multiplied by respectively quantizing factor k
1, k
2, obtain the input E of fuzzy system
fwith R
f;
By E
fwith R
fcarry out obfuscation, fuzzy reasoning and defuzzification, obtain blur estimation value u
f;
By blur estimation value u
fcarry out coordinate transform and make its codomain for [0,1], and be multiplied by proportionality coefficient k
3, obtain expecting grip f
d;
Get the maximum expectation grip f calculating in (50-300ms) in the prosthetic hand contact object a bit of time afterwards
dinitial reference power while capturing object as prosthetic hand.
2. prosthetic hand as claimed in claim 1 captures object initial reference power blur estimation method, it is characterized in that: the detailed process that described control algolithm and parameter thereof are formulated is as follows:
The first step: need capture different-stiffness object for prosthetic hand finger and carry out many experiments, measure grasping contact force by the pressure transducer being installed on finger, and grip is got to differential obtain power gradient:
dF＝(F
k-F
k-1)/T (1)
Described grip refers to the prosthetic hand finger grasping contact force measured with the dactylus of object contact at first;
Second step: from many experiments result, get the maximum F of contact force
maxwith minima F
min, the maximum dF of contact force gradient
maxwith minima dF
min, the normalized parameter when estimating with reference to power.Normalized object is to make contact force and the unification of contact force gradient in domain [1,1], is convenient to quantification and the obfuscation of input.Normalization formula is as follows:
f＝[2F-(F
max+F
min)]/[2(F
max-F
min)] (2)
df＝[2dF-(dF
max+dF
min)]/[2(dF
max-dF
min)] (3)
Wherein: F
max, F
minrepresent respectively maximum and the minima of contact force; DF
max, dF
minrepresent respectively maximum and the minima of contact force gradient;
The 3rd step: the membership function of determining blur estimation input and output;
The 4th step: set up fuzzy rule base;
The 5th step: fuzzy reasoning mechanism and defuzzification.
3. prosthetic hand as claimed in claim 1 captures object initial reference power blur estimation method, it is characterized in that: described membership function all can adopt and hang bell, triangular function, trapezoidal, Gaussian function or other functional forms; Described fuzzy reasoning mechanism can adopt Madamni inference mechanism, Larsen inference mechanism, Zadeh inference mechanism etc.; Described defuzzification can adopt height method, centroid method, area-method, weighted mean method or additive method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310329353.1A CN103876867B (en) | 2013-08-01 | 2013-08-01 | A kind of prosthetic hand grasps object initial reference power blur estimation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310329353.1A CN103876867B (en) | 2013-08-01 | 2013-08-01 | A kind of prosthetic hand grasps object initial reference power blur estimation method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103876867A true CN103876867A (en) | 2014-06-25 |
| CN103876867B CN103876867B (en) | 2018-01-19 |
Family
ID=50946216
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310329353.1A Active CN103876867B (en) | 2013-08-01 | 2013-08-01 | A kind of prosthetic hand grasps object initial reference power blur estimation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103876867B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105892331A (en) * | 2014-11-25 | 2016-08-24 | 中南大学 | Grasping force estimation and control method for motor-driven manipulator |
| CN110859688A (en) * | 2019-12-06 | 2020-03-06 | 中国科学院长春光学精密机械与物理研究所 | Intelligent shoe for artificial limb control and control method of artificial limb |
| CN111805545A (en) * | 2020-07-13 | 2020-10-23 | 河北省科学院应用数学研究所 | Dexterous hand control method and device and terminal equipment |
| CN112873263A (en) * | 2020-11-24 | 2021-06-01 | 北京邮电大学 | Air cylinder driving type dexterous hand reflection anti-skid control system and method |
| WO2023163669A1 (en) * | 2022-02-28 | 2023-08-31 | Ozyegin Universitesi | A controllable artificial hand system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0516932A1 (en) * | 1991-06-04 | 1992-12-09 | Schweizerische Unfallversicherungsanstalt Suva | Control device for the grasping force of a motordriven gripper, especially for a hand prothesis |
| CN101057795A (en) * | 2007-05-18 | 2007-10-24 | 天津大学 | Artificial hand using muscle electrical and electroencephalogram cooperative control and controlling method thereof |
| US20080215162A1 (en) * | 2007-02-06 | 2008-09-04 | Troy Farnsworth | System and method for using a digit to position a prosthetic or orthotic device |
| US20110071664A1 (en) * | 2009-09-22 | 2011-03-24 | Gm Global Technology Operations, Inc. | Human grasp assist device and method of use |
| CN202060919U (en) * | 2011-05-04 | 2011-12-07 | 上海理工大学 | Dynamic artificial-hand speed adjusting system based on FMG (ForceMyography) proportion |
| CN102309365A (en) * | 2011-08-30 | 2012-01-11 | 西安交通大学苏州研究院 | Wearable brain-control intelligent prosthesis |
| WO2012071343A1 (en) * | 2010-11-22 | 2012-05-31 | Vanderbilt University | Control system for a grasping device |
-
2013
- 2013-08-01 CN CN201310329353.1A patent/CN103876867B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0516932A1 (en) * | 1991-06-04 | 1992-12-09 | Schweizerische Unfallversicherungsanstalt Suva | Control device for the grasping force of a motordriven gripper, especially for a hand prothesis |
| US20080215162A1 (en) * | 2007-02-06 | 2008-09-04 | Troy Farnsworth | System and method for using a digit to position a prosthetic or orthotic device |
| CN101057795A (en) * | 2007-05-18 | 2007-10-24 | 天津大学 | Artificial hand using muscle electrical and electroencephalogram cooperative control and controlling method thereof |
| US20110071664A1 (en) * | 2009-09-22 | 2011-03-24 | Gm Global Technology Operations, Inc. | Human grasp assist device and method of use |
| WO2012071343A1 (en) * | 2010-11-22 | 2012-05-31 | Vanderbilt University | Control system for a grasping device |
| CN202060919U (en) * | 2011-05-04 | 2011-12-07 | 上海理工大学 | Dynamic artificial-hand speed adjusting system based on FMG (ForceMyography) proportion |
| CN102309365A (en) * | 2011-08-30 | 2012-01-11 | 西安交通大学苏州研究院 | Wearable brain-control intelligent prosthesis |
Non-Patent Citations (2)
| Title |
|---|
| 吴常铖等: "一种肌电假手的自适应模糊控制方法", 《仪器仪表学报》 * |
| 郝智秀等: "握取器握力自适应控制", 《清华大学学报》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105892331A (en) * | 2014-11-25 | 2016-08-24 | 中南大学 | Grasping force estimation and control method for motor-driven manipulator |
| CN110859688A (en) * | 2019-12-06 | 2020-03-06 | 中国科学院长春光学精密机械与物理研究所 | Intelligent shoe for artificial limb control and control method of artificial limb |
| CN110859688B (en) * | 2019-12-06 | 2021-07-16 | 中国科学院长春光学精密机械与物理研究所 | Intelligent shoe for artificial limb control and control method of artificial limb |
| CN111805545A (en) * | 2020-07-13 | 2020-10-23 | 河北省科学院应用数学研究所 | Dexterous hand control method and device and terminal equipment |
| CN111805545B (en) * | 2020-07-13 | 2021-06-08 | 河北省科学院应用数学研究所 | Dexterous hand control method and device and terminal equipment |
| CN112873263A (en) * | 2020-11-24 | 2021-06-01 | 北京邮电大学 | Air cylinder driving type dexterous hand reflection anti-skid control system and method |
| WO2023163669A1 (en) * | 2022-02-28 | 2023-08-31 | Ozyegin Universitesi | A controllable artificial hand system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103876867B (en) | 2018-01-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Xu et al. | Advances and disturbances in sEMG-based intentions and movements recognition: A review | |
| Tavakoli et al. | Single channel surface EMG control of advanced prosthetic hands: A simple, low cost and efficient approach | |
| Allard et al. | A convolutional neural network for robotic arm guidance using sEMG based frequency-features | |
| Geethanjali | Myoelectric control of prosthetic hands: state-of-the-art review | |
| CN103876867A (en) | Fuzzy estimation method for initial article grabbing reference force of hand prosthesis | |
| Zhao et al. | Levenberg-Marquardt based neural network control for a five-fingered prosthetic hand | |
| CN102499797B (en) | Artificial limb control method and system | |
| Fajardo et al. | Galileo bionic hand: sEMG activated approaches for a multifunction upper-limb prosthetic | |
| Oweis et al. | ANN-based EMG classification for myoelectric control | |
| Ahamed et al. | Fuzzy inference system-based recognition of slow, medium and fast running conditions using a triaxial accelerometer | |
| Abd et al. | Surface feature recognition and grasped object slip prevention with a liquid metal tactile sensor for a prosthetic hand | |
| Zhang et al. | Simultaneous estimation of joint angle and interaction force towards sEMG-driven human-robot interaction during constrained tasks | |
| Fajardo et al. | User-prosthesis interface for upper limb prosthesis based on object classification | |
| Khushaba et al. | Myoelectric feature extraction using temporal-spatial descriptors for multifunction prosthetic hand control | |
| CN103417315B (en) | A kind of anthropomorphic reflection control method of prosthetic hand | |
| Xiao et al. | Human motion intention recognition method with visual, audio, and surface electromyography modalities for a mechanical hand in different environments | |
| Potluri et al. | sEMG based fuzzy control strategy with ANFIS path planning for prosthetic hand | |
| Torres et al. | Finger movements classification from grasping spherical objects with surface electromyography using time domain based features | |
| Lu et al. | Time series modeling of surface EMG based hand manipulation identification via expectation maximization algorithm | |
| Maufroy et al. | CNN-based detection and classification of grasps relevant for worker support scenarios using sEMG signals of forearm muscles | |
| Tsujimura et al. | Hand sign classification employing myoelectric signals of forearm | |
| Tuga et al. | Concept of a co-adaptive training environment for human-machine interfaces based on EMG-control | |
| Villarejo et al. | Pattern recognition of hand movements with low density sEMG for prosthesis control purposes | |
| Zhang et al. | A method to modify initial desired force of prosthetic hand based on stiffness estimation | |
| Wang et al. | Brain-Controlled Robotic Arm Grasping System Based on Adaptive TRCA |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |