KR20220081346A - Brain plasticity-based motion assist device and its control method and circuit - Google Patents

Abstract

The present invention relates to the field of human-computer interaction technology, and more particularly, to a brain plasticity-based motion assisting device, and a control method and circuit thereof. The drive mechanism adopts a wire drive design, so that the drive mechanism and the actuation mechanism can be arranged separately. In other words, the drive mechanism can be placed on the user's arm, waist, or pocket, or even within the backpack being carried, making the entire piece of equipment lightweight and easy to wear. In addition, the present invention overcomes the difficulty of inaccurately identifying the exercise intention by the EMG signal by combining the classifier, voting machine, threshold comparison, and self-adaptive control together, and reduces the control delay of the entire system to directly reflect the user's intention to operate. can do. In the control of the execution mechanism, the control method of self-adaptive adaptive control is adopted, and the electromyography signal main power estimation can be combined to provide power as needed, so that a better human-computer interaction experience and rehabilitation effect can be realized. .

Description

Brain plasticity-based motion assist device and its control method and circuit

The present invention relates to the field of human-computer interaction technology, and more particularly, to a brain plasticity-based motion assisting device, and a control method and circuit thereof.

Motion assist devices are mainly targeted for people with limb movement disorders and are widely used in neurorehabilitation and medical and military fields. Since the use of the hand is very important and frequent in life, the design of the hand motion assist device is particularly important.

Patients with hand dysfunction can be broadly classified into two categories: one is caused by damage to the brain and nerve pathways, and the other is caused by muscle damage. The first type of patient can be treated with neurotransplantation, and the second type of patient can be treated with targeted muscle neurotransplantation. However, since the nerves used in both neurotransplantation and target nerve transplantation are not normally distributed in the original muscle, the patient after surgery typically receives a certain amount of time rehabilitation training to achieve the purpose of brain remodeling.

Most of the conventional motion assisting devices are made of a rigid material, and they are heavy and bulky, which makes them very uncomfortable to wear and imposes an additional physical burden on the user.

In terms of motion control, most of the conventional equipment is controlled by a mechanical switch. That is, the movement is stopped by pressing the switch operation auxiliary device and releasing the switch. This exercise method is very rigid, and even the exercise force becomes excessively large, which can cause secondary damage.

In terms of motion intention interpretation, there are mainly intention identification based on mechanical information and intention identification based on bioelectrical information. Here, intention identification employing mechanical information mainly uses human body kinematics and dynamics information, but mainly employs electromyography and brain waves for bioelectrical signals. Mechanical information is easy to collect and the signal is stable, but can only be obtained after the user has started exercising, with significant delays. At the same time, mechanical information cannot be collected with a few small hand movements, and the mechanical information cannot directly reflect the user's exercise intention.

Although bioelectrical signal-based exercise intention identification can solve the delay problem well, since there are interference signals (for example, overlapping signals, noise signals, etc.) can

In particular, the patient can indirectly identify and use the movement intention of a new functional area of brain remodeling by effectively identifying the bioelectrical signal that changes after brain remodeling, and the changed bioelectrical signal. Therefore, it is possible to further expand the control ability of the new functional area of brain remodeling and to improve the motor function of patients with limb disorders.

For this purpose, it is urgent to design a brain plasticity-based motion assisting device, its control method and circuit, which is light, easy to wear, flexible to control, and accurate in identifying exercise intentions.

In order to compensate for the deficiencies of the prior art, the present invention designed a brain plasticity-based motion assisting device, a control method and circuit thereof, that is light, easy to wear, flexible to control, and accurately identifies an exercise intention.

In order to achieve the above object, the present invention designed a brain plasticity-based motion assisting device,

used to collect and transmit a user's bioelectrical signal, the signal collection mechanism comprising a signal sensor and a signal transmission line, wherein the signal sensor is disposed on the user's body, and one end of the signal transmission line is connected with the signal sensor, the signal The other end of the transmission line is connected to the control mechanism - ;

an operation mechanism used for controlling on/off of the operation auxiliary device, comprising a signal transmitting and receiving module and a control module, wherein the control module is connected to the signal transmitting and receiving module, and the signal transmitting and receiving module is further connected to the control mechanism;

A control mechanism used to identify and process the information transmitted by the signal acquisition mechanism and the manipulation mechanism, obtain a result of the user's motion intent, and feed it back to the driving mechanism, wherein the control mechanism includes an analog digital processing module, motion intent An identification system, a driving force calculation module and a signal transmission module are installed, the signal transmission/reception module and the signal transmission line are all connected to the analog digital processing module, and the signal output terminal of the analog digital processing module is connected to the signal input terminal of the motion intention identification system, the signal output end of the movement intention identification system is connected with the signal input end of the driving force calculation module, the signal output end of the driving force calculation module is connected with the signal input end of the signal transmission module, and the signal output end of the signal transmission module is connected with the driving mechanism;

a power supply mechanism for supplying power to the control mechanism and the drive mechanism, the power mechanism including a power supply and a charge/discharge controller, the power source is connected to the charge/discharge controller, the charge/discharge controller is further connected to the control mechanism and the drive mechanism;

A driving mechanism receiving a command from the control mechanism to control the execution mechanism to move, wherein the driving mechanism includes a driver and an encoder, a signal input terminal of the driver is connected to the control mechanism, and a power input terminal of the driver is connected to a charge/discharge controller a signal output terminal of the driver is connected to the execution mechanism, and a signal output terminal of the encoder is connected to a signal input terminal of the control mechanism; and

an execution mechanism, which is used to assist a user's movement, and includes a power unit, an electric mechanism, and an auxiliary terminal - a power input terminal of the power unit is connected to a power output terminal of a charge/discharge controller, and the power unit converts power into power Then, output to the encoder and the electric mechanism, respectively, and the electric mechanism controls the auxiliary terminal to operate under power control - included.

Further, the transmission mechanism is

a main shaft used to provide the power required for the transmission mechanism;

The main shaft is detachably mounted on the side wall of the base, one end of which passes through one side wall of the base and is connected to the power unit, and the other end of the main shaft is fixed to the other side wall of the base - The main shaft has a unidirectional bearing set 1 is installed to cover - ;

A driven shaft that is detachably mounted on the side wall of the base and installed so that a unidirectional bearing set 2 is covered thereon, the driven shaft is connected to the main shaft using a pair of gears, and rotates together with the main shaft;

a steering shaft which is respectively detachably mounted on the bottom plate of the base and used to change the direction of the cord; and

Includes code used to control the terminal to move.

The cord is wound on the unidirectional bearing of the main shaft, the unidirectional bearing of the driven shaft and the steering shaft, and then is connected to the external terminal through a wire hole on the side wall of the base.

Here, the gear pair includes a main gear and a driven gear, the main gear is mounted on the main shaft, and the driven gear is mounted on the driven shaft.

Furthermore, the auxiliary terminal is a flexible exoskeleton glove or a rigid flexible exoskeleton glove, and includes a glove body, a concentrator, and a wire organizer 2 installed between the concentrator and the glove, and the glove body includes a finger cover part, a back part of a hand and a palm part. The three elements are connected using a driving line, and the driving line is arranged in Wire Organizer 2 and then goes into the concentrator and is connected with the driving mechanism.

Here, the collecting device includes a housing and a rotatable wheel shaft installed inside the housing, a wire passage groove is opened at a rear portion of the housing, a wire inlet hole is opened on one side of the housing, and the wire A wire lead-out hole is installed in an upper portion of the lead-in hole, a collection hole is installed in a lower portion of the wire lead-in, and one end of the wire lead-in hole and the wire lead-out hole located inside the housing is close to the wheel shaft, and the collecting hole is inside the housing one end positioned at the bottom of the wheel shaft is passed through from below the wheel shaft to communicate with the wire passage groove, and the wire passage groove is detachably connected to the drive head so that the drive line harness communicates with the drive line passing through the collecting device. , make the drive mechanism to control the auxiliary terminal motion.

The present invention designed a control method of a motion assistance device based on brain plasticity to accurately determine an exercise intention,

Step S1, wherein the signal collection mechanism collects the user's bioelectrical signal and transmits it to the control mechanism;

Step S2, wherein the control mechanism receives the bioelectrical signal, and analog-digital processing is performed in the analog-to-digital processing module;

Step S3 of transmitting the processed electrical signal to the exercise intention identification system, and obtaining an exercise intention result by identifying the exercise intention;

transmitting the motion intention result to the driving force calculation module, calculating the driving force, and transmitting the driving force signal to the driving mechanism; and

S5 the driving mechanism receives the driving force signal, generates a corresponding driver command, and controls the driver movement, so as to drive the execution mechanism to assist the user in performing the relevant exercise; step S5 is included.

Furthermore, the present invention further includes the step S0 of training the motion assistance device.

A specific method for training the motion assistance device is as follows.

The manipulation mechanism sends a training command to the motion assistance device, the manipulation mechanism displays the motion screen, the user performs a corresponding action according to the screen prompt, and the signal acquisition mechanism collects the user's bioelectrical signal, analog digital processing is performed, electrical signal feature extraction is performed in the signal feature extraction module, and the extracted electrical signal features are stored in the LDA classifier to obtain the user-based LDA classifier, and then training is completed.

Furthermore, a sensing circuit and a filter sampling circuit are installed in the analog-to-digital processing module.

Here, the sensing circuit includes an instrumentation amplifier and a bandpass filter.

The filter sampling circuit includes a programmable tunable amplifier, a low pass filter and an ADC.

The signal input terminal of the instrumentation amplifier is connected to a signal transmission line in the signal collection mechanism, receives the collected bioelectrical signal, amplifies it, and sends it to a bandpass filter to filter out interference sound, and then transmits the electrical signal to the filter sampling circuit. It is transferred to a programmable amplifier, which adaptively amplifies the electrical signal according to the programming above, and then passes through a low-pass filter to the ADC.

Furthermore, the exercise intention identification system includes a storage module, a signal feature extraction module, an LDA classifier, a voting machine, and an exercise intention determination module.

Here, the specific method of identifying the exercise intention is,

Step S31, wherein the storage module receives the electrical signal transmitted by the ADC among the analog-to-digital processing module and stores it;

After the stored electrical signal reaches the storage threshold, the signal feature extraction module performs electrical signal feature extraction S32;

Step S32 of transferring the extracted features to the LDA classifier to obtain a classification result;

Step S33 of transferring the classification result to a voting machine to obtain a voting result;

Comparing the threshold value while transmitting the voting result to the exercise intention determination module S34; and

S35 A step S35 of comparing the result of the voting machine and the threshold comparison result, and outputting the final exercise intention result if the two results match, and outputting the idle state result if the two results do not match.

Here, the specific method of comparing the threshold value is to first set the threshold values TH1 and TH2 in the exercise intention determination module, TH1 indicates the height threshold, TH2 indicates the curvature threshold, and then the calculated surface EMG signal The absolute average value is compared with TH1 and TH2. If MAV ₁ > TH1, it is determined that the motion intention is a stretching motion, and when MAV ₂ > TH2, it is determined that the motion intent is a curved motion.

Furthermore, the specific method of calculating the driving force is,

Step S41 of calculating the driving force according to the exercise intention result - The driving force calculation formula is as follows,

Here, F ₁ is the driving force required when the user performs a stretching exercise, c ₁ is the electrical signal-force conversion coefficient when the muscle performs stretching, and MAV ₁ is the surface EMG signal of the user during stretching. is the absolute average value, F ₂ is the driving force required when the user performs a curve exercise, c ₁ is the electrical signal-force conversion coefficient when the muscle performs a curve exercise, and MAV ₁ is the surface EMG signal when the user performs a curve exercise is the absolute mean value of - ; and

The calculated driving force is self-adaptively controlled, that is, the control law is adjusted through the self-adaptation law using the self-adaptive algorithm to make the output of the actual system model equal to the reference system model, and finally the controlled driving force is a step S42 of outputting .

Furthermore, the bioelectrical signal according to the present invention is preferably a surface electromyography signal.

When the bioelectrical signal is a surface EMG signal, the surface EMG signal feature extracted by the signal feature extraction module is the absolute average value MAV of the surface EMG signal and the number of zero crossing points ZC,

Here, N represents the number of data points of the surface EMG signal collected within the set time, and x _i represents the surface EMG signal of the ith channel, and is i∈N*.

The present invention is further designed a computer-readable storage medium, and the computer-readable storage medium stores a control method of a brain plasticity-based motion assistance device, wherein the control method of the brain plasticity-based motion assistance device is performed by a processor when run,

Step S0 for training the motion assistance device - The specific method is,

The manipulation mechanism sends a training command to the motion assistance device, the manipulation mechanism displays the motion screen, the user performs a corresponding action according to the screen prompt, and the signal acquisition mechanism collects the user's bioelectrical signal, analog digital to perform processing, to perform electrical signal feature extraction in the signal feature extraction module, and to store the extracted electrical signal features in the LDA classifier to obtain the user-based LDA classifier and then complete training;

Step S1, wherein the signal collection mechanism collects the user's bioelectrical signal and transmits it to the control mechanism;

Step S2, wherein the control mechanism receives the bioelectrical signal, and the analog-to-digital processing module performs analog-to-digital processing.

The instrumentation amplifier receives the bioelectrical signal collected by the signal acquisition mechanism, amplifies it, and passes it to a bandpass filter to filter out the interference sound, and then passes the electrical signal to the programmable amplifier of the filter sampling circuit, and the After adaptively amplifying an electrical signal according to the above programming, it is transferred to the ADC through a low-pass filter - ;

Transmitting the processed electrical signal to the exercise intention identification system, and identifying the exercise intention to obtain the exercise intention result S3 - A specific method for identifying the exercise intention is:

Step S31, wherein the storage module receives the electrical signal transmitted by the ADC among the analog-to-digital processing module and stores it;

After the stored electrical signal reaches the storage threshold, the signal feature extraction module performs electrical signal feature extraction S32;

Step S32 of transferring the extracted features to the LDA classifier to obtain a classification result;

Step S33 of transferring the classification result to a voting machine to obtain a voting result;

Comparing the threshold value while transmitting the voting result to the exercise intention determination module S34; and

Step S35 of comparing the voting machine result and the threshold comparison result, and outputting the final exercise intention result (extension exercise or curvature exercise) if the two results match, and outputting the idle state result if the two results do not match;

,

이고; 획득한 구동력을 자가적응적으로 제어하며, 즉 자가적응적 알고리즘을 이용하여 자가적응 법칙을 통해 제어 법칙을 조정하여 실제 시스템 모델의 출력이 기준 시스템 모델과 같도록 만들고, 마지막으로 제어된 구동력 신호를 출력하는 단계S4; 및The motion intention result is transmitted to the driving force calculation module, the driving force is calculated, and the driving force calculation formula is

,

ego; The obtained driving force is self-adaptively controlled, that is, the output of the actual system model is the same as the reference system model by adjusting the control law through the self-adaptive law using the self-adaptive algorithm, and finally, the controlled driving force signal is outputting step S4; and

Step S5 is implemented in which the driving mechanism receives the driving force signal to generate a corresponding driver command, and controls the driver movement to drive the execution mechanism to assist the user in performing the relevant movement.

Compared with the prior art, the drive mechanism adopts a wire drive design, so that the drive mechanism and the actuation mechanism can be arranged separately. In other words, the drive mechanism can be placed on the user's arm, waist, or pocket, or even within the backpack being carried, making the entire piece of equipment lightweight and easy to wear. In addition, the present invention overcomes the difficulty of inaccurately identifying the exercise intention by the EMG signal by combining the classifier, voting machine, threshold comparison, and self-adaptive control together, and reduces the control delay of the entire system to directly reflect the user's intention to operate. can do. In the control of the execution mechanism, the control method of self-adaptive adaptive control is adopted, and the electromyography signal main power estimation can be combined to provide power as needed, so that a better human-computer interaction experience and rehabilitation effect can be realized. .

1 is a structural diagram of an apparatus for assisting motion based on brain plasticity according to a specific embodiment.
FIG. 2 is a structural diagram of a transmission mechanism in the structure of a motion assisting device based on brain plasticity according to a specific embodiment.
3 is a flowchart of a method for controlling a motion assisting device based on brain plasticity according to a specific embodiment.
4 is a structural diagram of a surface EMG signal collection circuit in a method for controlling a motion assistance device based on brain plasticity according to a specific embodiment.
5 is a flowchart of identifying an exercise intention in a method for controlling a motion assistance device based on brain plasticity according to a specific embodiment.
6 is a processing diagram of bioelectrical signal feature extraction in a method for controlling a motion assistance device based on brain plasticity according to a specific embodiment.
7 is a flowchart illustrating a driving force self-adaptive control in a method for controlling a motion assistance device based on brain plasticity according to a specific embodiment.
8 is a diagram illustrating a use of a brain plasticity-based motion assistance device according to a specific embodiment.
9 is a side view of a concentrator in a brain plasticity-based motion assisting device according to a specific embodiment.
10 is a cross-sectional view of a concentrator in a brain plasticity-based motion assistance device according to a specific embodiment.
11 is a front view of a concentrator in a brain plasticity-based motion assisting device according to a specific embodiment.

The present invention will be further described with reference to the accompanying drawings, but the present invention is not limited thereto.

The accompanying drawings of the present invention are all in simplified form and all use inaccurate proportions. Note that this is for easily and clearly explaining the embodiments of the present invention.

1, 3, and 5, the present invention designed a brain plasticity-based motion assisting device and its control method. Specifically, it is used to collect and transmit a user's surface EMG signal and includes a signal collection mechanism 1 including a signal sensor 1-2 and a signal transmission line 1-1.

In a specific embodiment, the signal sensor 1-2 adopts an electrode sheet. A total of three electrode sheets are designed, among which the first and third electrode sheets collect the user's surface EMG signal, and the second electrode sheet is used as a reference motor so that the absolute voltage of the electrical signal is within the working range of the instrumentation amplifier. do not deviate from

Referring to FIG. 8 , the electrode sheet is disposed on the user's body and connects the electrode sheet with the control mechanism 2 using a signal transmission line 1-1.

The operation mechanism 3 is used to control the on-off of the operation auxiliary device and includes a signal transmission/reception module 3-1 and a control module 3-2.

The control module 3-2 is any electronic device or electronic program, and may be one or more of an Android app, an iOS app, a WeChat mini-app, and an Alipay mini-app, or may be implemented with next-generation software technology. The signal transmission/reception module 3-1 may be any electronic equipment or electronic program capable of receiving a wireless signal or a wired signal, or may be a next-generation software technology.

In a specific embodiment, it is an app that is adopted in the control module 3-2, and that is adopted in the signal transmission/reception module 3-1 is a Bluetooth transceiver.

The user opens the app and clicks on or off, the Bluetooth signal sends a start signal to the control mechanism 2 through the Bluetooth transceiver to turn the present invention on or off.

The control mechanism 2 identifies and processes the information transmitted by the signal collection mechanism 1 and the operation mechanism 3 , and is used to obtain the user's action intention result and feed it back to the drive mechanism 4 . .

In the control mechanism 2, an analog digital processing module 2-1, an exercise intention identification system 2-2, a driving force calculation module 2-3 and a signal transmission module 2-4 are installed.

Preferably, as shown in Fig. 4, the analog-to-digital processing module 2-1 includes a sensing circuit 2-1-1 and a filter sampling circuit 2-1-2.

Here, the sensing circuit 2-1-1 includes an instrumentation amplifier 2-1-1-1 and a band-pass filter 2-1-1-2, and a filter sampling circuit 2-1-2. contains a programmable tunable amplifier (2-1-2-1), a low-pass filter (2-1-2-2) and an ADC (2-1-2-3).

In a specific embodiment, the analog-to-digital processing module 2-1 receives the signal transmitted from the Bluetooth transceiver, turns on or off the entire device, receives the surface EMG signal transmitted from the electrode sheet, and then responds to the analog-to-digital signal perform processing. The instrumentation amplifier (2-1-1-1-1) receives the differential voltage signal collected from the electrode sheet, amplifies it 500 times, and transfers it to the 10- to 500 Hz band-pass filter (2-1-1-2), the electrode sheet After filtering out low-frequency interference caused by displacement and high-frequency noise caused by environmental coupling, it enters the filter sampling circuit 2-1-2 through a cable with a length of about 0.5 m. This partial circuit is an analog-to-digital mixed circuit, where interference sources such as a crystal oscillator, MCU, and electrode driver can be integrated, and the versatility of the EMG sensor may be reduced due to differences in the user's own EMG signal strength. Therefore, in the filter sampling circuit (2-1-2), a programmable tunable amplifier (2-1-2-1) was added, and for different users, the program automatically adjusts the circuit to an appropriate amplification factor, and then the low-pass filter ( 2-1-2-2) to the ADC (2-1-2-3), and the ADC continuously receives the collected EMG signal and transfers it to the exercise intention identification system (2-2).

Preferably, the exercise intention identification system 2-2 includes a storage module 2-2-1, a signal feature extraction module 2-2-2, an LDA classifier 2-2-3, and a voting machine 2- 2-4) and an exercise intention determination module (2-2-5).

,

이다.In a specific embodiment, the storage threshold of the storage module 2-2-1 is set to 128, and when the storage module stores 128 EMG signals, feature extraction is performed. That is, the absolute mean value MAV and zero-crossing point ZC of the surface EMG signal are extracted, and the formula is

,

to be.

Here, N represents the number of data points of the surface EMG signal collected within the set time, and x _i represents the surface EMG signal of the ith channel, and is i∈N*.

To ensure real-time performance and identification rate, the length of the data processing window in the system is 256 data points and the length of the sliding window is 128 data points. Under a sampling rate of 1860 Hz, the processing latency is 68.8 ms, which is less than the acceptable latency of 300 ms. In this embodiment, since two EMG channels of electrode sheet No. 1 and No. 3 electrode sheet are designed, when 128 data points are obtained for each channel, these data are combined with the previous 128 data to extract features. That is, the window length is 256 data points, and the sliding window length is 128 data points. For a detailed processing procedure, refer to FIG. 6 .

Then, the extracted MAV and ZC are transferred to the LDA classifier (2-2-3) to perform linear discriminant analysis, and the analysis result is transferred to the voting machine to perform voting. The result obtained with the most votes is input to the exercise intention determination module 2-2-5, and the simultaneously extracted MAV is transferred to the exercise intention determination module 2-2-5 to perform threshold comparison.

The specific method of comparing the threshold value is to first set the threshold values TH1 and TH2 in the exercise intention determination module 2-2-5, TH1 indicates the height threshold, TH2 indicates the curvature threshold, and then the calculated The absolute average value of the surface EMG signal is compared with TH1 and TH2. If MAV>TH1, it is determined that the motion intention is a stretching exercise, and when MAV>TH2, it is determined that the motion intention is a curved motion.

Finally, two results may be obtained from the exercise intention determination module 2-2-5. One is the voting result of the voting machine, and the other is the threshold value comparison result. It compares the two results, and if they are the same, the identification result is output, and if they are different, the idle state result is output.

The output motion intention result enters the driving force calculation module 2-3. The main role of the module is to calculate the assisting force so that the execution mechanism is appropriate, so that the user can properly assist the user to complete the stretching and flexion exercise so that the mechanism does not block the movement or the mechanism moves too quickly.

,

을 이용하여 대응하는 구동력을 계산한 후, 계산된 구동력에 대해 자가적응적 제어를 수행한다. 도 7을 참고하면, 자가적응적 알고리즘을 이용하여 자가적응 법칙을 통해 제어 법칙을 조정하여 실제 시스템 모델의 출력이 기준 시스템 모델과 같도록 만들고, 마지막으로 제어된 구동력을 구동 메커니즘에 출력한다.Preferably, the specific method of calculating the driving force is as follows. That is, the formula for calculating the driving force

,

After calculating the corresponding driving force using , self-adaptive control is performed on the calculated driving force. Referring to FIG. 7 , the control law is adjusted through the self-adaptation law using the self-adaptive algorithm so that the output of the actual system model is the same as the reference system model, and finally, the controlled driving force is output to the driving mechanism.

The reference model of the present invention selects a second-order system. It is not necessary to know the exact actual system model here, and the self-adaptive controller is designed based on the Narendra method.

Here, F ₁ is the driving force required when the user performs a stretching exercise, c ₁ is the electrical signal-force conversion coefficient when the muscle performs stretching, and MAV ₁ is the surface EMG signal of the user during stretching. is the absolute average value, F ₂ is the driving force required when the user performs a curve exercise, c ₁ is the electrical signal-force conversion coefficient when the muscle performs a curve exercise, and MAV ₁ is the surface EMG signal when the user performs a curve exercise is the absolute average of

Preferably, the drive mechanism 4 includes a driver 4-1 and an encoder 4-2. The driver 4-1 outputs a corresponding current according to the control signal of the controller to drive the movement of the execution mechanism, and the encoder 4-2 is used for pulse counting and transmits it to the control mechanism 2 .

Accordingly, the present invention should include a power supply mechanism (5) for supplying power to the control mechanism (2) and the drive mechanism (4). The power mechanism 5 includes a power source 5-1 and a charge/discharge controller 5-2, the power source 5-1 is connected to the charge/discharge controller 5-2, and the charge/discharge controller ( 5-2) is connected with the control mechanism 2 and the drive mechanism 4 .

Preferably, the actuation mechanism includes a power unit 6-1, a transmission mechanism 6-2 and an auxiliary terminal 6-3. In a specific embodiment, the power unit is a motor and the transmission mechanism is a linear mechanism.

The motor may provide power to the transmission mechanism 6-2 to make the transmission mechanism 6-2 control the operation of the auxiliary terminal 6-3.

To ensure accuracy during use of motion aids, training should be performed prior to use. That is, the LDA classifier must be trained, and the specific method is:

The operation mechanism 3 sends a training command to the operation assisting device, and displays an operation screen on the operation mechanism 3, the user performs a corresponding operation according to the screen prompt, and the signal collection mechanism 1 receives the user's Bioelectrical signals are collected, analog digital processing is performed, electrical signal feature extraction is performed in the signal feature extraction module (2-2-2), and the extracted electrical signal features are transferred to the LDA classifier (2-2-3) By storing, training is completed after obtaining the user-based LDA classifier (2-2-3).

Preferably, the transmission mechanism includes a main shaft (6-2-1) for providing the power required for the transmission mechanism (6-2);

The main shaft (6-2-1) is detachably mounted on the side wall of the base (6-2-2), and one end of the power unit (6-1) passes through one side wall of the base (6-2-2). ) and the other end of the main shaft (6-2-1) fixed to the other side wall of the base (6-2-2) - The main shaft (6-2-1) has a unidirectional bearing set 1 (6) -2-3) is installed to cover - ;

Driven shaft (6-2-5) that is detachably mounted on the side wall of the base (6-2-2) and installed so that the one-way bearing set 2 (6-2-4) is covered thereon - the driven shaft (6) -2-5) is connected to the main shaft (6-2-1) using a gear pair, and rotates together with the main shaft (6-2-1);

a steering shaft (6-2-6) which is detachably mounted on the bottom plate of the base (6-2-2), respectively, and used to change the direction of the cord (6-2-7); and

and a code (6-2-7) used to control the auxiliary terminal (6-3) to move.

The above code (6-2-7) is unidirectional bearing set 1 (6-2-3) of the main shaft (6-2-1), unidirectional bearing set 2 (6-2) of the driven shaft (6-2-5) -4) and after being wound around the steering shaft (6-2-6), it is connected to the external auxiliary terminal (6-3) through a wire hole on the side wall of the base (6-2-2).

In a specific embodiment with reference to FIG. 2 , four unidirectional bearings are adopted, and two unidirectional bearings are used as a set, respectively, mounted on the main shaft and the driven shaft, and the unidirectional bearings on the main shaft are the first unidirectional bearings and the second unidirectional bearings. divided In the driven shaft, the unidirectional bearing is divided into a third unidirectional bearing and a fourth unidirectional bearing.

Preferably, two steering shafts are adopted and are denoted as a first steering shaft and a second steering shaft, respectively.

Preferably, the code also adopts two sets and is denoted by a first code and a second code, respectively.

Preferably, two sets of pressing shafts are also adopted, respectively, located on both upper and lower sides of the driven shaft, so that the cord is in close contact with the unidirectional bearing on the driven shaft, so that the wire withdrawal is reliable.

The driven gear of the gear pair is mounted on the driven shaft, and the main gear of the gear pair is mounted on the driven shaft. Since the driven gear meshes with the main gear, the driven shaft is driven to rotate along the main shaft.

Preferably, the first cord is installed to be wound around the first unidirectional bearing of the main shaft, the third unidirectional bearing of the driven shaft, and the first steering shaft, and then passes through the first cord on the side wall of the base to be connected to the external terminal.

Preferably, the second cord is installed to be wound around the second unidirectional bearing of the main shaft, the fourth unidirectional bearing of the driven shaft, and the second steering shaft, and then passes through the second cord on the side wall of the base to be connected to the external terminal.

Here, the main shaft of the two-stranded cord that controls the curvature and elongation of the terminal is controlled by a single motor, and time-division independent control of the winding of the two-stranded cord is implemented using the clutch effect of the one-way bearing, and the single cord does not affect the unwinding and winding of the other strand of cord when it is working. By using gear pair matching, a single motor can be implemented to simultaneously drive the rotation of the main shaft and the driven shaft.

In a specific implementation, according to the requirements for the direction of the cord of the auxiliary terminal, the direction of the cord is changed by using the steering axis. According to the requirements for the unwinding speed ratio of the two cords of the auxiliary terminal (each design method and the requirements of each wearer are different), the main shaft is actively rotated by the motor, the driven shaft is moved along it, and the core is It drives to perform the take-up and unwind motion, and the speed is controlled by the auxiliary terminal.

The winding and winding of the cord set corresponding to the two different operations (bending and stretching) of the auxiliary terminal are completely opposite and the speed remains the same.

8, preferably, the auxiliary terminal 6-3 according to the present invention is a flexible exoskeleton glove 41 or a rigid flexible exoskeleton glove 42, a glove body, a concentrator 6-3-1, and and a wire organizer 2 installed between the concentrator 6-3-1 and the glove. The glove body is composed of a finger cover part, a back part of a hand, and a palm part, and the three parts are connected using a driving wire, and the driving wire is arranged in Wire Organizer 2 and then transferred to a collecting device (6-3-1). It enters and connects with the drive mechanism (14).

9 to 11 , here the collecting device 6-3-1 is installed in the housing 6-3-1-1 and the housing 6-3-1-1 and the rotatable wheel shaft 6 -3-1-2), and a wire passage groove 6-3-1-3 is opened at the rear portion of the housing 6-3-1-1, and the housing 6-3- A wire lead-in hole 6-3-1-4 is opened on one side of 1-1), and a wire lead-out hole 6-3-1 is located above the wire lead-in hole 6-3-1-4. -5) is installed, and a collection hole (6-3-1-6) is installed in the lower portion of the wire inlet, and a wire inlet hole (6-3-1-4) and a wire lead out hole (6-3-1) are installed. -5) has one end positioned inside the housing 6-3-1-1 is close to the wheel shaft 6-3-1-2, and the collection hole 6-3-1-6 is located in the housing 6 -3-1-1) One end located inside is penetrated from below the wheel shaft 6-3-1-2 and communicates with the wire passage groove 6-3-1-3. The wire passage groove 6-3-1-3 is detachably connected to the drive head so that the drive line harness communicates with the drive line penetrating through the collecting device 6-3-1, so that the drive mechanism (4) make the auxiliary terminal 6-3 control the movement.

In the selection of the auxiliary terminal, the flexible exoskeleton glove is made of a flexible material and adopts a wire driving method to control the glove movement, comfortable to wear and light to use. The flexible exoskeleton glove also adopts a flexible linear motorized design, and the back plate, palm plate and finger cover made of metal materials provide the rigidity required for training while making it easy to mold to accommodate a variety of patients.

In the concentrator, the present invention designs a wire passage groove to detachably connect the present invention and the external driving mechanism, so that the driving mechanism of different models and the exoskeleton device of different models can be freely assembled and matched.

In a specific embodiment, the present invention further designs a computer-readable storage medium, in which a control method of a brain plasticity-based motion assistance device is stored, wherein the brain plasticity-based motion assistance device control method is configured by a processor When executed by

Step S0 for training the motion assistance device - The specific method is,

The manipulation mechanism sends a training command to the motion assistance device, the manipulation mechanism displays the motion screen, the user performs a corresponding action according to the screen prompt, and the signal acquisition mechanism collects the user's bioelectrical signal, analog digital to perform processing, to perform electrical signal feature extraction in the signal feature extraction module, and to store the extracted electrical signal features in the LDA classifier to obtain the user-based LDA classifier and then complete training;

Step S1, wherein the signal collection mechanism collects the user's bioelectrical signal and transmits it to the control mechanism;

Step S2, wherein the control mechanism receives the bioelectrical signal, and the analog-to-digital processing module performs analog-to-digital processing.

The instrumentation amplifier receives the bioelectrical signal collected by the signal acquisition mechanism, amplifies it, and passes it to a bandpass filter to filter out the interference sound, and then passes the electrical signal to the programmable amplifier of the filter sampling circuit, and the After adaptively amplifying an electrical signal according to the above programming, it is transferred to the ADC through a low-pass filter - ;

Transmitting the processed electrical signal to the exercise intention identification system, and identifying the exercise intention to obtain the exercise intention result S3 - A specific method for identifying the exercise intention is:

Step S31, wherein the storage module receives the electrical signal transmitted by the ADC among the analog-to-digital processing module and stores it;

After the stored electrical signal reaches the storage threshold, the signal feature extraction module performs electrical signal feature extraction S32;

Step S32 of transferring the extracted features to the LDA classifier to obtain a classification result;

transferring the classification result to a voting machine to obtain a voting result S3S;

Step S34 of comparing the threshold while transmitting the voting result to the exercise intention determination module; and

Comparing the voting machine result and the threshold comparison result, if the two results match, outputting the final exercise intention result (stretching exercise or curvature exercise), and outputting the idle state result if the two results do not match;

,

이고, 획득한 구동력을 자가적응적으로 제어하며, 즉 자가적응적 알고리즘을 이용하여 자가적응 법칙을 통해 제어 법칙을 조정하여 실제 시스템 모델의 출력이 기준 시스템 모델과 같도록 만들고, 마지막으로 제어된 구동력 신호를 출력하는 단계; 및S4 Sends the motion intention result to the driving force calculation module, calculates the driving force, and the driving force calculation formula is

,

and the obtained driving force is self-adaptively controlled, that is, the output of the actual system model is the same as the reference system model by adjusting the control law through the self-adaptive law using the self-adaptive algorithm, and finally the controlled driving force outputting a signal; and

Step S5 is implemented in which the driving mechanism receives the driving force signal and generates a corresponding driver command, and controls the driver movement to drive the execution mechanism to assist the user in performing the relevant movement.

A person of ordinary skill in the art can complete all or part of the process in the method of the above embodiment by instructing the related hardware by a computer program, and the program can be stored in one computer-readable storage medium. can understand As in the embodiment of the present invention, the program may be stored in a storage medium of a computer system and executed by at least one processor of the computer system to implement a process including the above-described method embodiment. Here, the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

The technical features of the above-described embodiments may be arbitrarily combined. All possible combinations of technical features according to the above-described embodiments have not been described for concise description, but all possible combinations of technical features should be considered within the scope of the present invention, unless there is a contradiction in the combinations of technical features.

The above-described examples merely show some embodiments of the present invention, and the description is specific and detailed, but should not be construed as limiting the scope of the present invention. A person skilled in the art to which the present invention pertains can make various modifications and improvements without departing from the spirit of the present invention, all of which fall within the protection scope of the present invention.

Claims (17)

In the brain plasticity-based motion assisting device,
A signal collection mechanism (1), which is used to collect and transmit a user's bioelectrical signal, and includes a signal sensor (1-2) and a signal transmission line (1-1) - the signal sensor (1-2) is a user's body is disposed in, one end of the signal transmission line (1-1) is connected to the signal sensor (1-2), and the other end of the signal transmission line (1-1) is connected to the control mechanism (2);
Operation mechanism 3, which is used to control the on/off of the operation auxiliary device, comprising a signal transmission/reception module 3-1 and a control module 3-2 - the control module 3-2 is a signal transmission/reception module (3-2), and the signal transmitting and receiving module (3-1) is further connected with the control mechanism (2);
A control mechanism (2) used to identify and process the information transmitted by the signal acquisition mechanism (1) and the manipulation mechanism (3), obtain the result of the user's operation intention, and feed it back to the driving mechanism (4) - An analog digital processing module 2-1, an exercise intention identification system 2-2, a driving force calculation module 2-3 and a signal transmission module 2-4 are installed in the control mechanism 2, and the The signal transmission/reception module 3-1 and the signal transmission line 1-1 are both connected to the analog-to-digital processing module 2-1, and the signal output terminal of the analog-to-digital processing module 2-1 is connected to the exercise intention identification system 2 -2), the signal output terminal of the exercise intention identification system 2-2 is connected to the signal input terminal of the driving force calculation module 2-3, and the signal output terminal of the driving force calculation module 2-3 is connected with the signal input terminal of the signal transmission module 2-4, and the signal output terminal of the signal transmission module 2-4 is connected with the driving mechanism 4;
a power supply mechanism 5 for supplying power to the control mechanism 2 and the driving mechanism 4 - the power supply mechanism 5 includes a power supply 5-1 and a charge/discharge controller 5-2, and a power supply ( 5-1) is connected to a charge/discharge controller 5-2, and the charge/discharge controller 5-2 is further connected to a control mechanism 2 and a drive mechanism 4;
a drive mechanism 4 receiving a command from the control mechanism 2 to control the execution mechanism 6 to move, the drive mechanism 4 comprising a driver 4-1 and an encoder 4-2, The signal input terminal of the driver 4-1 is connected to the control mechanism 2, the power input terminal of the driver 4-1 is connected to the charge/discharge controller 5-2, and the signal of the driver 4-1 the output end is connected with the actuation mechanism 6 , and the signal output end of the encoder 4-2 is connected with the signal input end of the control mechanism 2 ; and
An actuation mechanism (6) used to assist the movement of a user, comprising a power unit (6-1), a power mechanism (6-2) and an auxiliary terminal (6-3) - the power unit (6-1) The power input terminal of the power supply is connected to the power output terminal of the charge/discharge controller 5-2, and the power unit 6-1 converts the power into power, and then the encoder 4-2 and the electric mechanism 6-2, respectively. ), and the electric mechanism (6-2) controls the auxiliary terminal (6-3) to operate under power control. According to claim 1,
The transmission mechanism 6-2 is,
a main shaft (6-2-1) used to provide the power required for the transmission mechanism (6-2);
The main shaft (6-2-1) is detachably mounted on the side wall of the base (6-2-2), and one end of the power unit (6-1) passes through one side wall of the base (6-2-2). ) and the other end of the main shaft (6-2-1) fixed to the other side wall of the base (6-2-2) - The main shaft (6-2-1) has a unidirectional bearing set 1 (6) -2-3) is installed to cover - ;
Driven shaft (6-2-5) that is detachably mounted on the side wall of the base (6-2-2) and installed so that the one-way bearing set 2 (6-2-4) is covered thereon - the driven shaft (6) -2-5) is connected to the main shaft (6-2-1) using a gear pair, and rotates together with the main shaft (6-2-1);
a steering shaft (6-2-6) which is detachably mounted on the bottom plate of the base (6-2-2), respectively, and used to change the direction of the cord (6-2-7); and
and a code (6-2-7) used to control the auxiliary terminal (6-3) to move;
The above code (6-2-7) is unidirectional bearing set 1 (6-2-3) of the main shaft (6-2-1), unidirectional bearing set 2 (6-2) of the driven shaft (6-2-5) -4) and after being wound around the steering shaft (6-2-6), the brain plasticity characterized in that it is connected to the external auxiliary terminal (6-3) through the wire hole on the side wall of the base (6-2-2) based motion aids. 3. The method of claim 2,
The gear pair includes a main gear (6-2-8) and a driven gear (6-2-9), the main gear (6-2-8) is installed on the main shaft (6-2-1) and the driven gear (6-2-9) is a brain plasticity-based motion assistance device, characterized in that it is mounted on the driven shaft (6-2-5). According to claim 1,
The auxiliary terminal 6-3 is a flexible exoskeleton glove 41 or a rigid flexible exoskeleton glove 42, and the glove body, the concentrator 6-3-1, and the concentrator 6-3-1 and the glove and a wire organizer 2 installed on the Brain plasticity-based motion assisting device, characterized in that it enters the concentrator (6-3-1) and is connected to the driving mechanism (4). 5. The method of claim 4,
The collector 6-3-1 includes a housing 6-3-1-1 and a rotatable wheel shaft 6-3-1-2 installed inside the housing 6-3-1-1. and a wire passage groove (6-3-1-3) is opened at the rear portion of the housing (6-3-1-1), and a wire is formed on one side of the housing (6-3-1-1) A lead-in hole (6-3-1-4) is opened, and a wire lead-out hole (6-3-1-5) is installed in an upper portion of the wire lead-in hole (6-3-1-4), and the wire A collection hole (6-3-1-6) is installed in the lower part of the inlet, and a wire inlet hole (6-3-1-4) and a wire outlet hole (6-3-1-5) are provided in the housing (6-3). -1-1) One end located inside is all close to the wheel shaft (6-3-1-2), and the collecting hole (6-3-1-6) is located inside the housing (6-3-1-1). One end positioned is passed through from below the wheel shaft 6-3-1-2 and communicates with the wire passage groove 6-3-1-3, and the wire passage groove 6-3-1-3 is detachably connected to the drive head, such that the drive line harness communicates with the drive line penetrating through the collecting device 6-3-1, so that the drive mechanism 4 controls the movement of the auxiliary terminal 6-3 Brain plasticity-based motion assistance device, characterized in that to make it. According to claim 1,
The bioelectrical signal is a brain plasticity-based motion assisting device, characterized in that it is a surface EMG signal. In the control method of a motion assistance device based on brain plasticity,
Step S1, wherein the signal collection mechanism (1) collects the user's bioelectrical signal and transmits it to the control mechanism (2);
step S2 of the control mechanism 2 receiving the bioelectric signal, and performing analog-digital processing in the analog-digital processing module 2-1;
Step S3 of transmitting the processed electrical signal to the exercise intention identification system 2-2, and identifying the exercise intention to obtain an exercise intention result;
step S4 of sending the motion intention result to the driving force calculation module 2-3, calculating the driving force, and sending the driving force signal to the driving mechanism 4; and
The driving mechanism 4 receives the driving force signal to generate a corresponding driver 4-1 command, and controls the driver 4-1 movement, so that the execution mechanism 6 assists the user in performing the relevant movement A control method of a motion assistance device based on brain plasticity, comprising: a step S5 of driving to do so. 8. The method of claim 7,
The method of controlling a motion assistance device based on brain plasticity, characterized in that it further comprises the step S0 of training the motion assistance device. 8. The method of claim 7,
The analog-to-digital processing module (2-1) is provided with a sensing circuit (2-1-1) and a filter sampling circuit (2-1-2);
the sensing circuit (2-1-1) includes an instrumentation amplifier (2-1-1-1-1) and a band-pass filter (2-1-1-2);
The filter sampling circuit (2-1-2) includes a programmable amplifier (2-1-2-1), a low-pass filter (2-1-2-2) and an ADC (2-1-2-3). including;
The signal input terminal of the instrumentation amplifier 2-1-1-1 is connected to the signal transmission line 1-1 of the signal collection mechanism 1, and after receiving the collected bioelectrical signal and amplifying it, a band-pass filter ( 2-1-2) to filter out the interference, and then transfer the electrical signal to the programmable amplifier (2-1-2-1) of the filter sampling circuit (2-1-2-1), Brain plasticity-based, characterized in that after adaptively amplifying an electrical signal according to the programming above, it is transferred to the ADC (2-1-2-3) through a low-pass filter (2-1-2-2) control method of the operation auxiliary device. 8. The method of claim 7,
The exercise intention identification system (2-2) includes a storage module (2-2-1), a signal feature extraction module (2-2-2), an LDA classifier (2-2-3), and a voting machine (2-2-4). ), a control method of a brain plasticity-based motion assistance device comprising a motion intention determination module (2-2-5). 8. The method of claim 7,
A specific way to identify exercise intent is:
a step S31 in which the storage module (2-2-1) receives the electrical signal transmitted by the ADC among the analog-to-digital processing module (2-1) and stores it;
After the stored electrical signal reaches the storage threshold, the signal feature extraction module 2-2-2 performs electrical signal feature extraction S32;
Step S32 of transferring the extracted features to the LDA classifier (2-2-3) to obtain a classification result;
Step S33 to obtain a voting result by transferring the classification result to the voting machine (2-2-4);
Step S34 of comparing the threshold value while transmitting the voting result to the exercise intention determination module (2-2-5); and
Step S35 of comparing the result of the voting machine (2-2-4) and the threshold comparison result, outputting the final exercise intention result if the two results match, and outputting the idle state result if the two results do not match; characterized in that A control method of a motion assist device based on brain plasticity. 8. The method of claim 7,
The specific method of calculating the driving force is:
Step S41 of calculating the driving force according to the exercise intention result - The driving force calculation formula is as follows,

Here, F ₁ is the driving force required when the user performs a stretching exercise, c ₁ is the electrical signal-force conversion coefficient when the muscle performs a stretching exercise, and MAV ₁ is the absolute value of the surface EMG signal when the user performs a stretching exercise is the average value, F ₂ is the driving force required when the user performs a curved motion, c ₁ is the electrical signal-force conversion coefficient when the muscle performs a curved motion, and MAV ₁ is the surface EMG signal when the user performs a curved motion. Absolute average value - ; and
The calculated driving force is self-adaptively controlled, that is, the output of the actual system model is the same as the reference system model by adjusting the control law through the self-adaptive law using the self-adaptive algorithm, and finally the controlled driving force A control method of a motion assistance device based on brain plasticity, characterized in that the step S42 of outputting a. 9. The method of claim 8,
Specific methods for training motion aids include:
The operation mechanism 3 sends a training command to the operation assisting device, and displays an operation screen on the operation mechanism 3, the user performs a corresponding operation according to the screen prompt, and the signal collection mechanism 1 receives the user's Bioelectrical signals are collected, analog digital processing is performed, electrical signal feature extraction is performed in the signal feature extraction module (2-2-2), and the extracted electrical signal features are transferred to the LDA classifier (2-2-3) The method of controlling a motion assistance device based on brain plasticity, characterized in that the training is completed after obtaining the user-based LDA classifier (2-2-3) by storing it. 14. The method of claim 13,
The bioelectrical signal is a control method of a motion assisting device based on brain plasticity, characterized in that it is a surface EMG signal. 15. The method of claim 14,
when the bioelectric signal is a surface EMG signal, the surface EMG signal feature extracted by the signal feature extraction module 2-2-2 is the absolute average value MAV of the surface EMG signal and the zero-crossing point number ZC;

Here, N represents the number of data points of the surface EMG signal collected within the set time, x _i represents the surface EMG signal of the ith channel, and i∈N* control method. 12. The method of claim 11,
The specific method of comparing the threshold value is to first set the threshold values TH1 and TH2 in the exercise intention determination module 2-2-5, TH1 indicates the height threshold, TH2 indicates the curvature threshold, and then calculate Brain plasticity, characterized in that the absolute average value of the surface EMG signal is compared with TH1 and TH2, and when MAV ₁ > TH1, it is determined that the motion intention is a stretching exercise, and when MAV ₂ > TH2, it is determined that the motion intention is a curved motion. A method of controlling a motion aid based on a motion aid. A computer-readable storage medium comprising:
A computer-readable storage medium stores a method for controlling a motion assistance device, wherein when the control method for a motion assistance device is executed by a processor, the method for controlling a motion assistance device according to any one of claims 7 to 16 A computer readable storage medium characterized in that this is implemented.

Images