WO2021174705A1 - 一种多自由度肌电假手控制系统及其使用方法 - Google Patents
一种多自由度肌电假手控制系统及其使用方法 Download PDFInfo
- Publication number
- WO2021174705A1 WO2021174705A1 PCT/CN2020/094132 CN2020094132W WO2021174705A1 WO 2021174705 A1 WO2021174705 A1 WO 2021174705A1 CN 2020094132 W CN2020094132 W CN 2020094132W WO 2021174705 A1 WO2021174705 A1 WO 2021174705A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- degree
- wrist
- freedom
- gesture
- circuit board
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 24
- 230000003183 myoelectrical effect Effects 0.000 title claims abstract description 15
- 210000000707 wrist Anatomy 0.000 claims abstract description 46
- 238000013528 artificial neural network Methods 0.000 claims abstract description 19
- 230000033001 locomotion Effects 0.000 claims abstract description 16
- 210000003205 muscle Anatomy 0.000 claims description 31
- 230000005540 biological transmission Effects 0.000 claims description 16
- 230000007246 mechanism Effects 0.000 claims description 14
- 210000002569 neuron Anatomy 0.000 claims description 13
- 230000004913 activation Effects 0.000 claims description 11
- 210000003414 extremity Anatomy 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 7
- 230000009471 action Effects 0.000 claims description 5
- 210000000245 forearm Anatomy 0.000 claims description 5
- 238000012847 principal component analysis method Methods 0.000 claims description 3
- 230000003042 antagnostic effect Effects 0.000 claims description 2
- 239000005557 antagonist Substances 0.000 claims description 2
- 238000002567 electromyography Methods 0.000 abstract 2
- 230000008569 process Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 7
- 238000012549 training Methods 0.000 description 6
- 239000011159 matrix material Substances 0.000 description 5
- 210000004247 hand Anatomy 0.000 description 3
- 241001227561 Valgus Species 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 210000002364 input neuron Anatomy 0.000 description 2
- 210000004205 output neuron Anatomy 0.000 description 2
- 238000000513 principal component analysis Methods 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 241000469816 Varus Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000007123 defense Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 231100001261 hazardous Toxicity 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000013138 pruning Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000000284 resting effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61F—FILTERS IMPLANTABLE INTO BLOOD VESSELS; PROSTHESES; DEVICES PROVIDING PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, e.g. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS; BANDAGES, DRESSINGS OR ABSORBENT PADS; FIRST-AID KITS
- A61F2/00—Filters implantable into blood vessels; Prostheses, i.e. artificial substitutes or replacements for parts of the body; Appliances for connecting them with the body; Devices providing patency to, or preventing collapsing of, tubular structures of the body, e.g. stents
- A61F2/50—Prostheses not implantable in the body
- A61F2/54—Artificial arms or hands or parts thereof
-
- 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
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61F—FILTERS IMPLANTABLE INTO BLOOD VESSELS; PROSTHESES; DEVICES PROVIDING PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, e.g. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS; BANDAGES, DRESSINGS OR ABSORBENT PADS; FIRST-AID KITS
- A61F2/00—Filters implantable into blood vessels; Prostheses, i.e. artificial substitutes or replacements for parts of the body; Appliances for connecting them with the body; Devices providing patency to, or preventing collapsing of, tubular structures of the body, e.g. stents
- A61F2/50—Prostheses not implantable in the body
- A61F2/68—Operating or control means
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61F—FILTERS IMPLANTABLE INTO BLOOD VESSELS; PROSTHESES; DEVICES PROVIDING PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, e.g. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS; BANDAGES, DRESSINGS OR ABSORBENT PADS; FIRST-AID KITS
- A61F2/00—Filters implantable into blood vessels; Prostheses, i.e. artificial substitutes or replacements for parts of the body; Appliances for connecting them with the body; Devices providing patency to, or preventing collapsing of, tubular structures of the body, e.g. stents
- A61F2/50—Prostheses not implantable in the body
- A61F2/68—Operating or control means
- A61F2/70—Operating or control means electrical
- A61F2/72—Bioelectric control, e.g. myoelectric
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60K—ARRANGEMENT OR MOUNTING OF PROPULSION UNITS OR OF TRANSMISSIONS IN VEHICLES; ARRANGEMENT OR MOUNTING OF PLURAL DIVERSE PRIME-MOVERS IN VEHICLES; AUXILIARY DRIVES FOR VEHICLES; INSTRUMENTATION OR DASHBOARDS FOR VEHICLES; ARRANGEMENTS IN CONNECTION WITH COOLING, AIR INTAKE, GAS EXHAUST OR FUEL SUPPLY OF PROPULSION UNITS IN VEHICLES
- B60K6/00—Arrangement or mounting of plural diverse prime-movers for mutual or common propulsion, e.g. hybrid propulsion systems comprising electric motors and internal combustion engines ; Control systems therefor, i.e. systems controlling two or more prime movers, or controlling one of these prime movers and any of the transmission, drive or drive units Informative references: mechanical gearings with secondary electric drive F16H3/72; arrangements for handling mechanical energy structurally associated with the dynamo-electric machine H02K7/00; machines comprising structurally interrelated motor and generator parts H02K51/00; dynamo-electric machines not otherwise provided for in H02K see H02K99/00
- B60K6/20—Arrangement or mounting of plural diverse prime-movers for mutual or common propulsion, e.g. hybrid propulsion systems comprising electric motors and internal combustion engines ; Control systems therefor, i.e. systems controlling two or more prime movers, or controlling one of these prime movers and any of the transmission, drive or drive units Informative references: mechanical gearings with secondary electric drive F16H3/72; arrangements for handling mechanical energy structurally associated with the dynamo-electric machine H02K7/00; machines comprising structurally interrelated motor and generator parts H02K51/00; dynamo-electric machines not otherwise provided for in H02K see H02K99/00 the prime-movers consisting of electric motors and internal combustion engines, e.g. HEVs
- B60K6/22—Arrangement or mounting of plural diverse prime-movers for mutual or common propulsion, e.g. hybrid propulsion systems comprising electric motors and internal combustion engines ; Control systems therefor, i.e. systems controlling two or more prime movers, or controlling one of these prime movers and any of the transmission, drive or drive units Informative references: mechanical gearings with secondary electric drive F16H3/72; arrangements for handling mechanical energy structurally associated with the dynamo-electric machine H02K7/00; machines comprising structurally interrelated motor and generator parts H02K51/00; dynamo-electric machines not otherwise provided for in H02K see H02K99/00 the prime-movers consisting of electric motors and internal combustion engines, e.g. HEVs characterised by apparatus, components or means specially adapted for HEVs
- B60K6/36—Arrangement or mounting of plural diverse prime-movers for mutual or common propulsion, e.g. hybrid propulsion systems comprising electric motors and internal combustion engines ; Control systems therefor, i.e. systems controlling two or more prime movers, or controlling one of these prime movers and any of the transmission, drive or drive units Informative references: mechanical gearings with secondary electric drive F16H3/72; arrangements for handling mechanical energy structurally associated with the dynamo-electric machine H02K7/00; machines comprising structurally interrelated motor and generator parts H02K51/00; dynamo-electric machines not otherwise provided for in H02K see H02K99/00 the prime-movers consisting of electric motors and internal combustion engines, e.g. HEVs characterised by apparatus, components or means specially adapted for HEVs characterised by the transmission gearings
- B60K6/365—Arrangement or mounting of plural diverse prime-movers for mutual or common propulsion, e.g. hybrid propulsion systems comprising electric motors and internal combustion engines ; Control systems therefor, i.e. systems controlling two or more prime movers, or controlling one of these prime movers and any of the transmission, drive or drive units Informative references: mechanical gearings with secondary electric drive F16H3/72; arrangements for handling mechanical energy structurally associated with the dynamo-electric machine H02K7/00; machines comprising structurally interrelated motor and generator parts H02K51/00; dynamo-electric machines not otherwise provided for in H02K see H02K99/00 the prime-movers consisting of electric motors and internal combustion engines, e.g. HEVs characterised by apparatus, components or means specially adapted for HEVs characterised by the transmission gearings with the gears having orbital motion
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61F—FILTERS IMPLANTABLE INTO BLOOD VESSELS; PROSTHESES; DEVICES PROVIDING PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, e.g. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS; BANDAGES, DRESSINGS OR ABSORBENT PADS; FIRST-AID KITS
- A61F2/00—Filters implantable into blood vessels; Prostheses, i.e. artificial substitutes or replacements for parts of the body; Appliances for connecting them with the body; Devices providing patency to, or preventing collapsing of, tubular structures of the body, e.g. stents
- A61F2/50—Prostheses not implantable in the body
- A61F2/54—Artificial arms or hands or parts thereof
- A61F2/58—Elbows; Wrists ; Other joints; Hands
- A61F2/583—Hands; Wrist joints
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61F—FILTERS IMPLANTABLE INTO BLOOD VESSELS; PROSTHESES; DEVICES PROVIDING PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, e.g. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS; BANDAGES, DRESSINGS OR ABSORBENT PADS; FIRST-AID KITS
- A61F2/00—Filters implantable into blood vessels; Prostheses, i.e. artificial substitutes or replacements for parts of the body; Appliances for connecting them with the body; Devices providing patency to, or preventing collapsing of, tubular structures of the body, e.g. stents
- A61F2/50—Prostheses not implantable in the body
- A61F2/54—Artificial arms or hands or parts thereof
- A61F2/58—Elbows; Wrists ; Other joints; Hands
- A61F2/583—Hands; Wrist joints
- A61F2/585—Wrist joints
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61F—FILTERS IMPLANTABLE INTO BLOOD VESSELS; PROSTHESES; DEVICES PROVIDING PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, e.g. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS; BANDAGES, DRESSINGS OR ABSORBENT PADS; FIRST-AID KITS
- A61F2/00—Filters implantable into blood vessels; Prostheses, i.e. artificial substitutes or replacements for parts of the body; Appliances for connecting them with the body; Devices providing patency to, or preventing collapsing of, tubular structures of the body, e.g. stents
- A61F2/50—Prostheses not implantable in the body
- A61F2/54—Artificial arms or hands or parts thereof
- A61F2/58—Elbows; Wrists ; Other joints; Hands
- A61F2/583—Hands; Wrist joints
- A61F2/586—Fingers
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61F—FILTERS IMPLANTABLE INTO BLOOD VESSELS; PROSTHESES; DEVICES PROVIDING PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, e.g. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS; BANDAGES, DRESSINGS OR ABSORBENT PADS; FIRST-AID KITS
- A61F2/00—Filters implantable into blood vessels; Prostheses, i.e. artificial substitutes or replacements for parts of the body; Appliances for connecting them with the body; Devices providing patency to, or preventing collapsing of, tubular structures of the body, e.g. stents
- A61F2/50—Prostheses not implantable in the body
- A61F2/68—Operating or control means
- A61F2002/6836—Gears specially adapted therefor, e.g. reduction gears
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61F—FILTERS IMPLANTABLE INTO BLOOD VESSELS; PROSTHESES; DEVICES PROVIDING PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, e.g. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS; BANDAGES, DRESSINGS OR ABSORBENT PADS; FIRST-AID KITS
- A61F2/00—Filters implantable into blood vessels; Prostheses, i.e. artificial substitutes or replacements for parts of the body; Appliances for connecting them with the body; Devices providing patency to, or preventing collapsing of, tubular structures of the body, e.g. stents
- A61F2/50—Prostheses not implantable in the body
- A61F2/68—Operating or control means
- A61F2/70—Operating or control means electrical
- A61F2002/701—Operating or control means electrical operated by electrically controlled means, e.g. solenoids or torque motors
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61F—FILTERS IMPLANTABLE INTO BLOOD VESSELS; PROSTHESES; DEVICES PROVIDING PATENCY TO, OR PREVENTING COLLAPSING OF, TUBULAR STRUCTURES OF THE BODY, e.g. STENTS; ORTHOPAEDIC, NURSING OR CONTRACEPTIVE DEVICES; FOMENTATION; TREATMENT OR PROTECTION OF EYES OR EARS; BANDAGES, DRESSINGS OR ABSORBENT PADS; FIRST-AID KITS
- A61F2/00—Filters implantable into blood vessels; Prostheses, i.e. artificial substitutes or replacements for parts of the body; Appliances for connecting them with the body; Devices providing patency to, or preventing collapsing of, tubular structures of the body, e.g. stents
- A61F2/50—Prostheses not implantable in the body
- A61F2/68—Operating or control means
- A61F2/70—Operating or control means electrical
- A61F2002/704—Operating or control means electrical computer-controlled, e.g. robotic control
-
- 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
- B25J9/161—Hardware, e.g. neural networks, fuzzy logic, interfaces, processor
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/45—Nc applications
- G05B2219/45172—Prosthesis
Definitions
- the present invention relates to a prosthetic hand control system and a use method thereof, and more specifically, to a multi-degree-of-freedom myoelectric prosthetic hand control system and a use method thereof.
- the multi-degree-of-freedom myoelectric prosthetic hand control system of the present invention includes a manipulator, a manipulator wrist, a residual limb receiving cavity and a data processor.
- the manipulator and residual limb receiving cavity are respectively installed at both ends of the mechanical wrist, and the residual limb receiving cavity
- a multi-channel EMG array electrode sleeve is connected, the multi-channel EMG array electrode sleeve is connected with a control unit circuit board and a battery, and the other end of the control unit circuit board is connected with a manipulator and a manipulator wrist.
- the data processor sends an instruction to collect the surface EMG signal to the control unit circuit board, so that the multi-channel EMG array electrode sleeve collects the surface EMG signal, and performs neural network processing on the received data to generate a gesture prediction model.
- the mechanical wrist includes a bevel gear set mechanism, a pulley transmission mechanism, a servo motor and a wrist support frame.
- the bevel gear set mechanism uses four bevel gears to mesh with each other to form a cross-shaped arrangement.
- the left and right bevel gears are installed on the wrist support.
- the frame is respectively connected with transmission wheels.
- the belt pulley transmission mechanism is connected to the transmission wheel and is connected with a servo motor.
- the two horizontal gears in the bevel gear set mechanism are the sun gear, and the sun gear is connected to the transmission wheel through the driving shaft.
- the sun gear is fixed on the driving shaft through the sun gear top wire, and the transmission wheel is fixed on the driving shaft through the driving shaft top wire.
- the upper part of the vertical direction is the first planetary gear connected to the manipulator, and the lower part is the second planetary gear, the first planetary gear and A hollow passive shaft passes between the second planetary gears, and deep groove bearings are installed between the hollow passive shaft and the first planetary gear and the second planetary gear.
- the wrist support frame is composed of a left panel, a right panel, a beam and a bottom plate. The upper end of the panel and the right panel are connected by beams, and the lower end is fixedly connected with the bottom plate.
- the servo motor is installed on the left and right panels. The servo motor pulley and the transmission wheel are sleeved through a belt, and a pressure roller is fixed on the outside of the belt.
- the method of using the multi-degree-of-freedom myoelectric prosthetic hand control system of the present invention includes the following steps:
- the data processor receives the surface EMG signal and inputs it into the neural network algorithm to generate a gesture prediction model
- the neural network algorithm in step S3 includes the following steps for data processing:
- (S31) Preprocess the original surface EMG signal to extract the muscle activation signal, and then divide it with a fixed-length time window and use it as the input layer of the unsupervised neural network.
- the first hidden layer of the network uses the principal component analysis method to compress the time -Spatial characteristics;
- the second hidden layer uses an autoencoder to learn the muscle signal characteristics of 2N forearm muscles that cooperate with each other when completing different gestures, and generate continuous gesture labels based on the muscle coordination characteristics and the experimental action sequence, where 2N represents the 2N gestures to be recognized Degree of freedom, N is the number of antagonistic muscles in the forearm muscles participating in gesture movement;
- the third hidden layer fits the muscle synergy feature with the continuous gesture label to generate a regression network.
- the output layer of the regression network contains N neurons, which respectively output the continuous kinematics and dynamics of the antagonist muscles. Data, where different neurons represent different gestures, and the continuous data output by the neuron represents the strength of the gesture.
- the present invention has significant advantages: 1. It can recognize continuous gestures and the strength of gestures; 2. It can make gestures with multiple degrees of freedom, and the artificial hand is more dexterous.
- Figure 1 is a schematic diagram of the installation of the prosthetic hand of the present invention
- Figure 2 is a schematic diagram of the wrist structure
- Figure 3 is a schematic diagram of the structure of the bevel gear set mechanism
- Figure 4 is a schematic diagram of the structure of the wrist support frame
- Figure 5 is a side view of the wrist structure
- Figure 6 is a block diagram of a hierarchical neural network
- Figure 7 is a block diagram of the second hidden layer autoencoder of the gesture recognition algorithm
- Figure 8 is a schematic diagram of a label self-generation method
- Figure 9 is a flowchart of a gesture recognition algorithm
- Figure 10 is a flow chart of the neural network.
- the multi-degree-of-freedom electromyographic prosthetic hand control system is characterized by including a manipulator, a manipulator wrist 2, a residual limb receiving cavity 1 and a data processor 3.
- the manipulator and residual limb receiving cavity are respectively installed in the mechanical
- a multi-channel EMG array electrode sleeve is connected in the receiving cavity of the residual limb.
- the multi-channel EMG array electrode sleeve is connected with a control unit circuit board and a battery.
- the other end of the control unit circuit board is connected with the manipulator and the manipulator wrist.
- the data processor 3 sends an instruction to collect the surface EMG signal to the control unit circuit board, so that the multi-channel EMG array electrode sleeve collects the surface EMG signal, and performs neural network processing on the received data to generate a gesture prediction model.
- the mechanical wrist 2 includes a bevel gear set mechanism 4, a belt pulley transmission mechanism 5, a servo motor 7 and a wrist support frame 6.
- the bevel gear set mechanism 4 uses four bevel gears to mesh with each other to form a cross-shaped arrangement.
- the two horizontal gears are
- the sun gear 15 is installed on the wrist support frame 6.
- the sun gear 15 is connected to the driving wheel 10 through the driving shaft 9.
- the sun gear 15 is fixed on the driving shaft 9 through the sun gear top wire 12, and the driving wheel 10 is passed through the driving shaft top wire 11
- Fixed on the driving shaft 9, the upper part of the vertical direction is the first planetary gear 14 connected to the manipulator, the lower part is the second planetary gear 13, and a hollow driven shaft 8 passes between the first planetary gear 14 and the second planetary gear 13.
- a deep groove bearing is installed between the hollow driven shaft 8 and the first planetary gear 14 and the second planetary gear 13.
- the wrist support frame 6 is composed of a left panel, a right panel, a beam 16 and a bottom plate 17. The upper ends of the left and right panels are connected by the beam 18, and the lower end is fixedly connected with the bottom plate.
- the servo motors 7 are installed on the left and right panels.
- the belt wheel 19 and the transmission wheel 10 are sleeved through a belt 18.
- a pressure roller 20 is fixed on the outside of the belt 21.
- the flow of the gesture recognition algorithm is shown in Figure 9.
- Each gesture needs to be completed three times in a row before making the next gesture.
- the multi-channel EMG array electrode cuff collects EMG signals and stores them to the control unit circuit board and uploads them to the data processor 3, and inputs the neural network algorithm for processing.
- the processing flow is shown in Figure 10.
- the training set is processed by the three-layer network of the hierarchical neural network, as shown in Figure 6.
- the original EMG signal of 8 channels is preprocessed, and the root mean square RMS average value is used to obtain the activation signal.
- these 8 activation signals are divided by a fixed-length time window and used as the input layer of the neural network.
- Each input sample will contain the spatial and temporal information of the array EMG signal.
- the first hidden layer of the network uses principal component analysis Method to reduce the dimensionality of the input signal.
- the second hidden layer uses an autoencoder to learn six muscle coordination features to further reduce the feature dimensionality.
- the third hidden layer fits the muscle coordination features with automatically generated motion intent labels, and finally
- the output layer of the network contains three neurons, which output continuous motion data with three degrees of freedom.
- the weight matrix of each hidden layer of the neural network is independently trained and stacked together, layer by layer in the actual process of fitting the deep neural network Fine adjustment, in which the predicted wrist movement information is used to control the manipulator wrist 2, and the hand opening and closing movement information is used to control the manipulator mounted on the manipulator wrist 2.
- the present invention performs principal component analysis on the time scale of the EMG activation signal of each channel, and the T sample points of the EMG activation signal in the time window are Substitute the features of principal component analysis to generate samples in the form of a sliding window. Set the sliding step length to 1, and the number of sampling points of the surface EMG signal of a single channel to m, then the number of generated samples is m-T+1.
- the figure describes each Each EMG signal channel selects two principal components to characterize the data in the time window.
- Figure 7 shows the process of extracting 2N muscle synergy features from the principal component features of the 8 channels in the second hidden layer in Figure 6,
- 2N represents the 2N gesture degrees of freedom to be recognized, and N is the forearm muscles participating in the gesture movement.
- N is the forearm muscles participating in the gesture movement.
- the number of neurons that continue to be input into the neural network is reduced from c ⁇ T to c ⁇ 2.
- the second layer of neurons is used to extract muscle synergy related Features.
- 6 types of muscle synergy features are extracted from the electromyographic activation features, corresponding to wrist valgus, wrist varus, clockwise wrist rotation, counterclockwise wrist rotation, hand open, and hand fist, and the muscle synergy feature value They are all non-negative values, and the weight matrix of the second layer of neurons is obtained by training an autoencoder.
- the main feature of the autoencoder is that the input neurons and output neurons are exactly the same, and the number of hidden layer neurons is less than the number of input and output neurons.
- This neural network structure can obtain some potential features in the input data.
- the text uses the encoding part of the self-encoder to complete the training as the weight matrix of the second layer of the network.
- the activation function of the encoding process uses the Relu function. Since the input layer contains features with negative values, in order to enable the output layer to recover negative features, the activation function of the decoding process uses Tanh function.
- the loss function of the autoencoder uses the cross-entropy function; the weight matrix of the encoder is initialized with the Xavier method, which can make the initial weights have a normal distribution with a mean value of 0; the iterative training process uses the pruning algorithm to reduce In case of small over-fitting, the network learning rate decays exponentially with the number of iterations, and the ADAM gradient descent method and Mini-Batch method are used to speed up the training speed. Compared with the non-negative matrix factorization method, the model fitted by this method is The calculation of the nonlinear activation function is improved, so it has a better approximation effect.
- Figure 8 shows the process of extracting kinematics and dynamics labels from the muscle coordination features obtained in Figure 7.
- the muscle coordination features learned by the autoencoder cannot directly obtain the desired motion intention, when the 6 coordination features are vector superimposed
- the oscillating waveform shown in Figure 8 will be obtained, in which each peak represents the maximum level of muscle coordination when a certain action is completed, and the troughs on both sides indicate that the muscles are in a resting state, so a complete trough -The crest-trough segment indicates the process from the completion of a certain gesture to the strongest muscle activation and then rest recovery.
- the kinematic parameter labels of the hands and wrists can be reconstructed with three degrees of freedom.
- the muscle synergy features and label data calculated by the upper layer network are finally substituted into a feedforward neural network for regression fitting.
- the obtained network layer is stacked with the network layer calculated in the previous two sections to form the final deep regression model. All network layers are trained and stacked to form the final gesture recognition network and fed back to the control unit circuit board.
- the actual use is to make the user wear the residual limb receiving cavity 1 and connect the manipulator and the manipulator 2.
- the user controls the wrist and hand movements of the prosthetic hand by imagining gestures, and the control unit circuit board controls the prosthesis at the same time.
- Manipulator wrist 2 multiple degrees of freedom movement of the manipulator, the speed of the manipulator movement is regulated by the muscle force predicted by the hierarchical network.
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Transplantation (AREA)
- Mechanical Engineering (AREA)
- General Health & Medical Sciences (AREA)
- Cardiology (AREA)
- Heart & Thoracic Surgery (AREA)
- Vascular Medicine (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Biomedical Technology (AREA)
- Automation & Control Theory (AREA)
- Robotics (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Transportation (AREA)
- Orthopedic Medicine & Surgery (AREA)
- Prostheses (AREA)
Abstract
Description
Claims (8)
- 一种多自由度肌电假手控制系统,其特征在于,包括机械手、机械手腕(2)、残肢接受腔(1)和数据处理器(3),所述机械手和残肢接受腔分别安装在机械手腕两端,所述残肢接受腔内连接有多通道肌电阵列电极袖套,所述多通道肌电阵列电极袖套连接有控制单元电路板和电池,所述控制单元电路板另一端连接机械手和机械手腕,所述数据处理器向控制单元电路板发出采集表面肌电信号的指令使多通道肌电阵列电极袖套采集表面肌电信号并接受数据进行处理生成手势预测模型。
- 根据权利要求1所述的多自由度肌电假手控制系统,其特征在于,所述机械手腕(2)包括锥齿轮组机构(4)、皮带轮传动机构(5)、伺服电机(7)和手腕支撑框架(6),所述锥齿轮组机构(4)采用四个锥齿轮相互啮合,构成十字型排布,左、右两个锥齿轮安装在手腕支撑框架(6)上,并分别连接有传动轮(10),所述皮带轮传动机构(5)连接在传动轮(10)上,并连接有伺服电机(7)。
- 根据权利要求2所述的多自由度肌电假手控制系统,其特征在于,所述锥齿轮组机构(4)中水平方向的两个齿轮为太阳轮(15),太阳轮(15)通过主动轴(9)与传动轮(10)相连,太阳轮(15)通过太阳轮顶丝(12)固定在主动轴(9)上,传动轮(10)通过传动轴顶丝(11)固定在主动轴(9)上,垂直方向上部为连接机械手的第一行星齿轮(14),下部为第二行星齿轮(13),所述第一行星齿轮(14)和第二行星齿轮(13)之间穿过一空心被动轴(8)。
- 根据权利要求3所述的多自由度肌电假手控制系统,其特征在于,所述空心被动轴(8)与第一行星齿轮(14)和第二行星齿轮(13)之间安装有深沟轴承。
- 根据权利要求2所述的多自由度肌电假手控制系统,其特征在于,所述手腕支撑框架(6)由左面板、右面板、梁(16)和底板(17)构成,所述左面板、右面板上端通过梁(18)连接,下端与底板固定连接,所述伺服电机(7)安装在左、右面板上,伺服电机皮带轮(19)与传动轮(10)通过皮带(18)套接。
- 根据权利要求5所述的多自由度肌电假手控制系统,其特征在于,所述皮带(21)外侧固定一压轮(20)。
- 一种权利要求1所述的多自由度肌电假手控制系统的使用方法,其特征在于,包括以下步骤:(S1)令使用者戴上多通道肌电阵列电极袖套,然后连接好控制单元电路板、 电池;(S2)令使用者根据实验动作序列完成手势,数据处理器(3)向控制单元电路板发出采集表面肌电信号的指令,控制单元电路板控制多通道肌电阵列电极袖套采集表面肌电信号后储存至控制单元电路板并上传至数据处理器(3);(S3)数据处理器(3)接收表面肌电信号并输入神经网络算法生成手势预测模型;(S4)使用者穿戴上残肢接受腔,并连接好机械手和机械手腕,利用生成的手势预测模型进行实时手势识别,控制单元电路板控制手腕、机械手的多个自由度运动。
- 根据权利要求7所述的多自由度肌电假手控制系统的使用方法,其特征在于,步骤S3中神经网络算法对数据处理包括以下步骤:(S31)对原始表面肌电信号进行预处理以提取肌肉激活信号,然后用固定长度的时间窗口分割并作为无监督神经网络的输入层,网络的第一个隐藏层利用主成分分析方法压缩时间-空间特征;(S32)第二个隐藏层采用自编码器学习2N个前臂肌肉完成不同手势时相互协同的肌肉信号特征,根据肌肉协同特征和实验动作序列生成连续手势标签,其中2N表示要识别的2N个手势自由度,N为参与手势运动的前臂肌肉中互为拮抗肌肉的个数;(S33)第三个隐藏层将肌肉协同特征与连续手势标签进行拟合,生成回归网络,回归网络的输出层包含N个神经元,分别输出N对拮抗肌表现出的连续运动学与动力学数据,其中不同神经元表示不同的手势,神经元输出的连续数据表示该手势的力度。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US17/628,753 US20220355469A1 (en) | 2020-03-03 | 2020-06-03 | Multi-degree-of-freedom myoelectric artificial hand control system and method for using same |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010139635.5A CN111300413B (zh) | 2020-03-03 | 2020-03-03 | 一种多自由度肌电假手控制系统及其使用方法 |
CN202010139635.5 | 2020-03-03 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2021174705A1 true WO2021174705A1 (zh) | 2021-09-10 |
Family
ID=71153579
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2020/094132 WO2021174705A1 (zh) | 2020-03-03 | 2020-06-03 | 一种多自由度肌电假手控制系统及其使用方法 |
Country Status (3)
Country | Link |
---|---|
US (1) | US20220355469A1 (zh) |
CN (1) | CN111300413B (zh) |
WO (1) | WO2021174705A1 (zh) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112732090B (zh) * | 2021-01-20 | 2022-08-09 | 福州大学 | 基于肌肉协同的用户无关实时手势识别方法 |
CN114053007B (zh) * | 2021-10-29 | 2022-06-21 | 哈尔滨工业大学 | 一种面向健康受试者的多自由度假肢实验装置 |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5413611A (en) * | 1992-07-21 | 1995-05-09 | Mcp Services, Inc. | Computerized electronic prosthesis apparatus and method |
US20060155386A1 (en) * | 2005-01-10 | 2006-07-13 | Wells David L | Electromyographic sensor |
CN103892945A (zh) * | 2012-12-27 | 2014-07-02 | 中国科学院深圳先进技术研究院 | 肌电假肢控制系统 |
CN106980367A (zh) * | 2017-02-27 | 2017-07-25 | 浙江工业大学 | 一种基于肌电地形图的手势识别方法 |
CN108491077A (zh) * | 2018-03-19 | 2018-09-04 | 浙江大学 | 一种基于多流分治卷积神经网络的表面肌电信号手势识别方法 |
CN109464227A (zh) * | 2018-12-20 | 2019-03-15 | 中国科学院深圳先进技术研究院 | 一种多自由度的假肢手臂关节 |
Family Cites Families (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2003305682A (ja) * | 2002-04-11 | 2003-10-28 | Fanuc Ltd | ロボット手首駆動機構 |
CN102379760B (zh) * | 2011-08-25 | 2014-11-05 | 东南大学 | 肌电假手力触觉反馈方法及触觉反馈的肌电假手系统 |
CN202458789U (zh) * | 2011-12-28 | 2012-10-03 | 上海科生假肢有限公司 | 适用于前臂腕离断和手掌根部截肢者的接受腔 |
CN204909745U (zh) * | 2015-07-30 | 2015-12-30 | 沈阳工业大学 | 带有高精度小型化拉线控制系统的智能肌电假肢手 |
CN104997579B (zh) * | 2015-07-30 | 2018-02-16 | 沈阳工业大学 | 带有高精度小型化拉线控制系统的智能肌电假肢手 |
CN209004339U (zh) * | 2018-06-05 | 2019-06-21 | 北京通和营润智能科技发展有限公司 | 一种多通道肌电前臂假肢接受腔 |
CN108983973B (zh) * | 2018-07-03 | 2021-01-26 | 东南大学 | 一种基于手势识别的仿人灵巧肌电假手控制方法 |
CN109866250B (zh) * | 2019-03-01 | 2021-08-13 | 达闼科技(北京)有限公司 | 机器人手腕结构及机器人 |
-
2020
- 2020-03-03 CN CN202010139635.5A patent/CN111300413B/zh active Active
- 2020-06-03 US US17/628,753 patent/US20220355469A1/en active Pending
- 2020-06-03 WO PCT/CN2020/094132 patent/WO2021174705A1/zh active Application Filing
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5413611A (en) * | 1992-07-21 | 1995-05-09 | Mcp Services, Inc. | Computerized electronic prosthesis apparatus and method |
US20060155386A1 (en) * | 2005-01-10 | 2006-07-13 | Wells David L | Electromyographic sensor |
CN103892945A (zh) * | 2012-12-27 | 2014-07-02 | 中国科学院深圳先进技术研究院 | 肌电假肢控制系统 |
CN106980367A (zh) * | 2017-02-27 | 2017-07-25 | 浙江工业大学 | 一种基于肌电地形图的手势识别方法 |
CN108491077A (zh) * | 2018-03-19 | 2018-09-04 | 浙江大学 | 一种基于多流分治卷积神经网络的表面肌电信号手势识别方法 |
CN109464227A (zh) * | 2018-12-20 | 2019-03-15 | 中国科学院深圳先进技术研究院 | 一种多自由度的假肢手臂关节 |
Also Published As
Publication number | Publication date |
---|---|
CN111300413B (zh) | 2022-10-14 |
CN111300413A (zh) | 2020-06-19 |
US20220355469A1 (en) | 2022-11-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10959863B2 (en) | Multi-dimensional surface electromyogram signal prosthetic hand control method based on principal component analysis | |
CN108785997B (zh) | 一种基于变导纳的下肢康复机器人柔顺控制方法 | |
Kwon et al. | Real-time upper limb motion estimation from surface electromyography and joint angular velocities using an artificial neural network for human–machine cooperation | |
Arslan et al. | Prediction of externally applied forces to human hands using frequency content of surface EMG signals | |
CN109199783B (zh) | 一种利用sEMG来控制踝关节康复设备刚度的控制方法 | |
CN110675933B (zh) | 一种手指镜像康复训练系统 | |
CN109262618A (zh) | 基于肌肉协同的上肢多关节同步比例肌电控制方法与系统 | |
WO2021174705A1 (zh) | 一种多自由度肌电假手控制系统及其使用方法 | |
JP6998569B2 (ja) | パワーアシスト装置およびパワーアシスト装置の制御方法 | |
Kulkarni et al. | Overview: mechanism and control of a prosthetic arm | |
Cho et al. | The MSC prosthetic hand: Rapid, powerful, and intuitive | |
CN109199654B (zh) | 一种实现假肢手柔顺运动的控制系统与方法 | |
Pham et al. | Extraction and implementation of muscle synergies in neuro-mechanical control of upper limb movement | |
Yang et al. | Vit-based terrain recognition system for wearable soft exosuit | |
Cho et al. | Estimating simultaneous and proportional finger force intention based on sEMG using a constrained autoencoder | |
Wei et al. | Continuous Motion Intention Prediction Using sEMG for Upper-Limb Rehabilitation: A Systematic Review of Model-Based and Model-Free Approaches | |
Long et al. | Online sparse Gaussian process based human motion intent learning for an electrically actuated lower extremity exoskeleton | |
Yang et al. | Design of elbow rehabilitation exoskeleton robot with semg-based torque estimation control strategy | |
Bian et al. | Reshaping Wearable Robots Using Fuzzy Intelligence: Integrating Type-2 Fuzzy Decision, Intelligent Control, and Origami Structure | |
Bo et al. | Hand gesture recognition using semg signals based on cnn | |
Ali et al. | EMG based control of a quadcopter | |
Gmerek | High-level controller for an arm rehabilitation robot-positioning algorithms with respect to EMG data | |
Souzanchi-K et al. | Control of elastic joint robot based on electromyogram signal by pre-trained Multi-Layer Perceptron | |
Soma et al. | Preliminary results of online classification of upper limb motions from around-shoulder muscle activities | |
Yan et al. | Controlling a powered exoskeleton system via electromyographic signals |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20923040 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 20923040 Country of ref document: EP Kind code of ref document: A1 |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 20923040 Country of ref document: EP Kind code of ref document: A1 |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205 DATED 06.07.2023.) |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 20923040 Country of ref document: EP Kind code of ref document: A1 |