Active training method of hand rehabilitation robot

Abstract

The invention discloses an active training method of a hand rehabilitation robot, and relates to the technical field of hand rehabilitation therapy, and the method comprises the following steps: S1, designing a corresponding circuit of a myoelectricity sensor; s2, drawing a corresponding schematic diagram according to the designed circuit diagram, selecting a four-layer board through the schematic diagram to carry out PCB design, and then manufacturing a corresponding myoelectricity sensor; and S3, selecting an Arduino UNO R3 development board to carry out electromyographic signal processing and steering engine control. According to the active training method of the hand rehabilitation robot, through the design of the low-cost myoelectricity sensor, the operation of the steering engine is controlled only by collecting the myoelectricity signals, different active rehabilitation trainings can be achieved through the cooperation of the corresponding structure and the steering engine, the design and manufacturing of a multi-channel myoelectricity sensor are not needed, and the cost is reduced. And the collected electromyographic signals do not need to be subjected to feature extraction and action recognition, so that the manufacturing cost is reduced, the interference process of electromyographic signal processing is reduced, and the reaction speed is increased.

Description

Technical Field

[0001] The present invention relates to the technical field of hand rehabilitation therapy, and in particular to an active training method for a hand rehabilitation robot. Background Art

[0002] With the increase of population aging, traffic accidents, industrial accidents, and neurological diseases (such as stroke, spinal cord injury), the number of patients with hand dysfunction is also increasing. Although traditional hand rehabilitation treatment methods are effective, they are often limited by the number of therapists, treatment time, and individual differences of patients. Therefore, it is particularly important to develop a device that can assist or replace artificial therapists and provide personalized, efficient, and continuous hand rehabilitation training. Hand rehabilitation robot technology came into being in this context.

[0003] Early active treatment is the key to improve the condition of stroke patients. Rationalized and standardized rehabilitation training can improve patients' self-care ability and quality of life. Passive rehabilitation is an exercise method performed subjectively by patients with the help of external forces. In the early stages of rehabilitation, this rehabilitation mode can ensure the flexibility of joint movements, but the rehabilitation robot assists too much in patient training, and the patient is too dependent on the robot's assistance, which seriously affects the effect of rehabilitation training. Therefore, in the later stages of the entire rehabilitation cycle of hemiplegic patients, active rehabilitation training becomes more important. Active rehabilitation is to enable hemiplegic patients to autonomously participate in rehabilitation training through control algorithms, so that patients can exercise with rehabilitation manipulators in their hands. Only in this way can the patient's dependence on rehabilitation equipment be gradually reduced, and then gradually get rid of the assistance of rehabilitation equipment to achieve the patient's final rehabilitation.

[0004] The existing implementation scheme is to collect corresponding electromyographic signals or electroencephalographic signals through multiple channels, and then pre-process these signals, extract features and classify them, and recognize actions through programs. After recognition, the flexible hand rehabilitation robot is driven to perform corresponding rehabilitation action training. However, in the prior art, the cost of designing and manufacturing multi-channel electromyographic sensors is high, and the collection, processing and recognition of multi-channel electromyographic or electroencephalographic signals are difficult and susceptible to interference, and the recognition accuracy is not high.

[0005] Therefore, there is an urgent need to improve this shortcoming. The present invention studies and improves the existing technology and its shortcomings, and provides an active training method for a hand rehabilitation robot. Summary of the invention

[0006] The purpose of the present invention is to provide an active training method for a hand rehabilitation robot to solve the problems raised in the above background technology.

[0007] To achieve the above object, the present invention provides the following technical solution: a method for active training of a hand rehabilitation robot, comprising the following steps:

[0008] S1. Design of the corresponding circuit of the electromyography sensor:

[0009] 1) Input buffer circuit: It has noise suppression capability and can filter out the noise components in the input signal;

[0010] 2) Pre-amplifier circuit: Use INA129 instrument amplifier and set the voltage gain to 5-15 times;

[0011] 3) Double T trap circuit: It consists of two T-type networks, and each T-type network consists of a resistor and two capacitors. This circuit is used to filter out or attenuate specific frequency interference in the circuit;

[0012] 4) Filter circuit: An important component of the electromyography detection circuit, including low-pass filter circuit and high-pass filter circuit. The filter circuit can filter out the low-frequency and high-frequency noise signals mixed in the electromyography signal;

[0013] 5) Post-amplification circuit: adopts inverting proportional amplification design, the circuit structure has only differential signal input, strong anti-interference ability, and can be compatible with a larger signal input range; the amplification gain formula is G = -R25 / R23, which is about 50 times amplification;

[0014] 6) Adder: After the reverse amplification, part of the signal is still negative. The adder adds 1.25V to the circuit signal as a whole, so that the overall amplitude range of the signal falls within the dynamic detection range of the microcontroller ADC.

[0015] 7) Voltage follower: OPA227 operational amplifier is used to form a voltage follower;

[0016] S2. According to the designed circuit diagram, use professional software to draw the corresponding schematic diagram. After comprehensively considering the volume and cost, select a four-layer board for PCB design through the schematic diagram, and then make the corresponding electromyographic sensor;

[0017] S3. Select the Arduino UNO R3 development board for EMG signal processing and servo control. Use Arduino to design the program. First, call the servo library that comes with Arduino and the EMG signal library that you insert yourself, and define the pins corresponding to the servo and EMG sensor.

[0018] Furthermore, in step S1, the voltage gain of INA129 is adjusted by the resistor RG, and the voltage gain G=1+49.4 kΩ / RG.

[0019] Furthermore, in step S1, the low-pass filter circuit is designed using a Butterworth filter, and a Sallen-Key structure is used as a topological structure, and a fourth-order low-pass filter is designed to attenuate signals above 500 Hz.

[0020] Furthermore, in step S1, the high-pass filter circuit and the low-pass filter circuit use the Sallen-Key topology and Butterworth filter, and the fourth-order high-pass filter is designed to attenuate signals below 20 Hz.

[0021] Furthermore, in step S3, the servo is a 20kg high-torque single-axis servo TBS-K20, which can rotate 270°.

[0022] Furthermore, in step S3, the program design content includes the following two modes:

[0023] The first mode: The electromyographic signal controls the servo to rotate 90° each time. After rotating 3 times, it automatically returns to its original position. The program statement controlling the servo directly outputs the servo angle.

[0024] The second mode: After receiving the electromyographic signal, the servo starts to move from the initial position to the 270° position, and then receives the electromyographic signal again to return to the initial position. Microseconds are used as parameters to set the servo rotation angle.

[0025] The present invention provides an active training method for a hand rehabilitation robot, which has the following beneficial effects:

[0026] The technical idea of ​​the present invention is to design a corresponding electromyographic detection circuit, make a single-channel electromyographic signal sensor, obtain the electromyographic signal of the patient's arm, trigger the program to control the operation of the steering gear through the electromyographic signal, and the operation of different steering gears drives the hand rehabilitation robot with corresponding rigid structure, so as to realize the flexion and extension of the patient's hand, as well as the rehabilitation training of adduction and abduction. The present invention uses the design of a low-cost electromyographic sensor, and only needs to collect electromyographic signals to control the operation of the steering gear. Different active rehabilitation training can be realized by coordinating the steering gear with the corresponding structure, without the need to design and make a multi-channel electromyographic sensor, and there is no need to perform feature extraction and motion recognition on the collected electromyographic signals, so as to reduce the production cost and reduce the interference process of electromyographic signal processing, and improve the reaction speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 A schematic diagram of the steps of an active training method for a hand rehabilitation robot according to the present invention;

[0028] Figure 2 This is a schematic diagram of an input buffer circuit of an active training method for a hand rehabilitation robot according to the present invention;

[0029] Figure 3 This is a schematic diagram of a preamplifier circuit of an active training method for a hand rehabilitation robot of the present invention;

[0030] Figure 4 A schematic diagram of a double-T trap circuit of an active training method for a hand rehabilitation robot according to the present invention;

[0031] Figure 5 A schematic diagram of a low-pass filter circuit for an active training method for a hand rehabilitation robot according to the present invention;

[0032] Figure 6 A schematic diagram of a high-pass filter circuit of an active training method for a hand rehabilitation robot according to the present invention;

[0033] Figure 7 It is a schematic diagram of a post-amplification circuit of an active training method for a hand rehabilitation robot of the present invention;

[0034] Figure 8 A schematic diagram of an adder for an active training method for a hand rehabilitation robot according to the present invention;

[0035] Fig. 9 This is a schematic diagram of a voltage follower of an active training method for a hand rehabilitation robot according to the present invention. DETAILED DESCRIPTION

[0036] The following embodiments of the present invention are described in further detail in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0037] like Figure 1-Figure 9 As shown, a hand rehabilitation robot active training method comprises the following steps:

[0038] S1. Design of the corresponding circuit of the electromyography sensor:

[0039] 1) Input buffer circuit: It has a certain noise suppression capability and can filter out the noise components in the input signal to improve the signal quality. For circuit design, see Figure 2 ;

[0040] 2) Pre-amplifier circuit: INA129 instrument amplifier is selected, which has the characteristics of high common mode rejection ratio and low noise; INA129 voltage gain is adjusted by resistor RG, voltage gain G = 1 + 49.4kΩ / RG, the voltage gain of the pre-amplifier circuit is set to 5-15 times, the circuit design is shown in Figure 3 , making the amplified signal easy to carry out subsequent signal conditioning and avoiding nonlinear distortion of the amplified signal;

[0041] 3) Double T trap circuit: It is composed of two T-type networks, and each T-type network is composed of a resistor and two capacitors. This circuit is used to filter out or attenuate specific frequency interference in the circuit. In this embodiment, the designed double T trap circuit is used to filter out 50Hz power frequency interference. By accurately adjusting the values ​​of the resistor and the capacitor, a narrow stop band can be formed at 50Hz to effectively filter out the signal of this frequency. The circuit design is shown in Figure 4 ;

[0042] 4) Filter circuit: An important component of the electromyography detection circuit, including low-pass filter circuit and high-pass filter circuit. The filter circuit can filter out the low-frequency and high-frequency noise signals mixed in the electromyography signal;

[0043] 4A, low-pass filter circuit: the main function is to allow low-frequency signals to pass through, while suppressing or attenuating high-frequency signals. For circuit design, see Figure 5 ;

[0044] Active filters include but are not limited to Butterworth filters, Chebyshev filters and Bessel filters, among which Butterworth filters: have flat amplitude responses in both passband and stopband, and the attenuation slope from passband to stopband is relatively gentle, that is, the stopband drop rate is continuous and gradually tends to negative infinity; Chebyshev filters: compared with other types of filters of the same order, Chebyshev filters have better attenuation effects in the stopband, but the cost is that there will be equal ripple fluctuations in the passband; Bessel filters: have good linear phase, but the stopband drops gently and the frequency selection effect is not ideal; therefore, this embodiment uses Butterworth filters for filter circuit design;

[0045] The filter topology structure includes a multi-stage negative feedback structure and a Sallen-Key topology structure. The Sallen-Key structure has the characteristics of high input impedance and low output impedance. Compared with the multi-stage negative feedback structure, it has lower noise gain and requires fewer components. In this embodiment, the filter is designed using a Butterworth filter, and the Sallen-Key structure is a topology structure. A fourth-order low-pass filter is designed to attenuate signals above 500HZ.

[0046] 4B. High-pass filter circuit: a circuit that allows signals with frequencies higher than a certain frequency threshold to pass through and has an attenuation effect on frequency components below a given frequency. In this embodiment, the high-pass filter circuit uses the Sallen-Key topology and Butterworth filter in the same manner as the low-pass filter circuit. A fourth-order high-pass filter is designed to attenuate signals below 20 Hz. See the circuit design for details. Figure 6 ; A fourth-order high-pass filter and a fourth-order low-pass filter can be used to form a band-pass filter to filter out noise signals and obtain better filtering effects;

[0047] 5) Post-amplifier circuit: adopts inverting proportional amplification design. This circuit structure has only differential signal input, strong anti-interference ability, and can be compatible with a larger signal input range. See circuit design for details. Figure 7 ; The formula for amplification gain is G = -R25 / R23, which is about 50 times amplification;

[0048] 6) Adder: After the reverse amplification, part of the signal is still negative. The adder adds 1.25V to the circuit signal as a whole, so that the overall amplitude range of the signal falls within the dynamic detection range of the microcontroller ADC. See the circuit design for details. Figure 8 ;

[0049] 7) Voltage follower: OPA227 operational amplifier is used to form a voltage follower. See the circuit design for details. Fig. 9 , the circuit plays the role of buffer isolation, improving the load capacity and improving the stability and reliability of the entire system;

[0050] S2. According to the designed circuit diagram, use Jiali Chuang EDA to draw the corresponding schematic diagram. After comprehensively considering the volume and cost, select a four-layer board for PCB design through the schematic diagram, and then make the corresponding electromyographic sensor;

[0051] S3. Select Arduino UNO R3 development board for EMG signal processing and servo control. The selected servo is 20kg high torque single-axis servo TBS-K20, which can rotate 270°. Program design through Arduino. First call the servo library provided by Arduino and the EMG signal library inserted by yourself, and define the corresponding pins of the servo and EMG sensor.

[0052] In this embodiment, the first mode is designed first, that is, the electromyographic signal controls the servo to rotate 90° each time, and after rotating 3 times, it automatically returns to the original position; the program statement for controlling the servo directly outputs the servo angle;

[0053] #include "EMGFilters.h" / / Call the electromyographic signal library

[0054] #include<Servo.h> / / Call the servo library

[0055] #define servo_pin 9 / / macro definition of servo control pin

[0056] #define SensorInputPin A0 / / define electrode pin

[0057] unsigned long ServoThreshold = 10000; / / Servo threshold

[0058] unsigned long threshold=100; / / EMG filter threshold

[0059] unsigned long EMG_num = 0;

[0060] Servo TBS; / / Create a servo class

[0061] EMGFilters myFilter; / / Create an EMG signal class

[0062] SAMPLE_FREQUENCY samplerate = SAMPLE_FREQ_500HZ; / / sampling rate

[0063] NOTCH_FREQUENCY notch=NOTCH_FREQ_50HZ; / / notch

[0064] void setup(){

[0065] myFilter.init(samplerate,notch,true,true,true); / / Initialize the filter

[0066] TBS.attach(servo_pin);

[0067] Serial.begin(115200);

[0068] }

[0069] void loop(){

[0070] int data = analogRead(SensorInputPin); / / Read the current analog value of the sensor

[0071] int dataAfterFilter=myFilter.update(data);

[0072] int envelope=sq(dataAfterFilter);

[0073] envelope = (envelope>threshold)? envelope: 0; / / The value is retained only when it is greater than a certain threshold

[0074] if(threshold>0){

[0075] if(envelope>ServoThreshold){

[0076] if(EMG_num>2){

[0077] EMG_num = 0;

[0078] TBS.write(0);

[0079]

[0080] The second mode is that when the electromyographic signal is received, the servo starts to move from the initial position to the position of 270°, and then receives the electromyographic signal again to return to the initial position. Microseconds are used as parameters to set the rotation angle of the servo.

[0081] #include "EMGFilters.h" / / Call the electromyographic signal library

[0082] #include<Servo.h> / / Call the servo library

[0083] #define servo_pin 9 / / macro definition of servo control pin

[0084] #define SensorInputPin A0 / / define electrode pin

[0085] unsigned int PWM = 0;

[0086] unsigned long ServoThreshold = 10000; / / Servo threshold

[0087] unsigned long threshold=70; / / EMG filter threshold

[0088] unsigned long EMG_num = 0;

[0089] Servo TBS; / / Create a servo class

[0090] EMGFilters myFilter; / / Create an EMG signal class

[0091] SAMPLE_FREQUENCY sampleRate = SAMPLE_FREQ_500HZ; / / sampling rate

[0092] NOTCH_FREQUENCY notch=NOTCH_FREQ_50HZ; / / notch

[0093] void setup(){

[0094] myFilter.init(sampleRate,notch,true,true,true); / / Initialize the filter

[0095] TBS.attach(servo_pin);

[0096] Serial.begin(115200);

[0097] }

[0098] void loop(){

[0099] int data = analogRead(SensorInputPin); / / Read the current analog value of the sensor

[0100] int dataAfterFilter=myFilter.update(data);

[0101] int envelope=sq(dataAfterFilter);

[0102] envelope = (envelope>threshold)? envelope: 0; / / The value is retained only when it is greater than a certain threshold

[0103] if(threshold>0){

[0104] if(envelope>ServoThreshold){

[0105] if(EMG_num>0){

[0106] EMG_num = 0;

[0107] for(PWM=2450;PWM>50;PWM-=5){

[0108] TBS.writeMicroseconds(PWM);

[0109] }

[0110] delay(500);

[0111] }else{

[0112] EMG_num++;

[0113] for(PWM=50;PWM<2450;PWM+=5){ / / The servo switches from state 50 to 2450, increasing by 5 each time

[0114] TBS.writeMicroseconds(PWM); / / Write PWM to the servo

[0115] }

[0116] delay(500);

[0117] }

[0118] }

[0119] }

[0120] Serial.println(envelope);

[0121] delayMicroseconds(500);

[0122] }

[0123] Through the above design, the electromyographic signal can be used to drive the operation of the servo, and then different structures can be designed to match the servo to realize the active training mode of the rigid hand rehabilitation robot.

[0124] The embodiments of the present invention are given for the purpose of illustration and description, and are not intended to be exhaustive or to limit the invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

Claims

1. A method for active training of a hand rehabilitation robot, characterized in that: The following steps are involved: S1. Design of the corresponding circuit of the electromyography sensor: 1) Input buffer circuit: It has noise suppression capability and can filter out the noise components in the input signal; 2) Pre-amplifier circuit: Use INA129 instrument amplifier and set the voltage gain to 5-15 times; 3) Double T trap circuit: It consists of two T-type networks, and each T-type network consists of a resistor and two capacitors. This circuit is used to filter out or attenuate specific frequency interference in the circuit; 4) Filter circuit: An important component of the electromyography detection circuit, including low-pass filter circuit and high-pass filter circuit. The filter circuit can filter out the low-frequency and high-frequency noise signals mixed in the electromyography signal; 5) Post-stage amplifier circuit: adopts inverting proportional amplifier design, the circuit structure has only differential signal input, and the amplifier gain formula is G = -R25 / R23; 6) Adder: Add 1.25V to the circuit signal through the adder, so that the overall amplitude range of the signal falls within the dynamic detection range of the microcontroller ADC; 7) Voltage follower: OPA227 operational amplifier is used to form a voltage follower; S2. According to the designed circuit diagram, use professional software to draw the corresponding schematic diagram. After comprehensively considering the volume and cost, select a four-layer board for PCB design through the schematic diagram, and then make the corresponding electromyographic sensor; S3. Select the Arduino UNO R3 development board for EMG signal processing and servo control. Use Arduino to design the program. First, call the servo library that comes with Arduino and the EMG signal library that you insert yourself, and define the pins corresponding to the servo and EMG sensor.

2. The active training method for a hand rehabilitation robot according to claim 1, characterized in that: In the step S1, the voltage gain of INA129 is adjusted by the resistor RG, and the voltage gain G=1+49.4 kΩ / RG.

3. The active training method for a hand rehabilitation robot according to claim 1, characterized in that: In the step S1, the low-pass filter circuit is designed using a Butterworth filter.

4. The active training method for a hand rehabilitation robot according to claim 3, characterized in that: In the low-pass filter circuit design, the Sallen-Key structure is a topological structure, and a fourth-order low-pass filter is designed to attenuate signals above 500 Hz.

5. The active training method for a hand rehabilitation robot according to claim 4, characterized in that: In step S1, the high-pass filter circuit and the low-pass filter circuit use the Sallen-Key topology and Butterworth filter, and the fourth-order high-pass filter is designed to attenuate signals below 20 Hz.

6. The active training method for a hand rehabilitation robot according to claim 1, characterized in that: In step S3, the servo is a 20kg high torque single-axis servo TBS-K20, which can rotate 270°.

7. The active training method for a hand rehabilitation robot according to claim 6, characterized in that: In step S3, the program design content includes the first mode: the electromyographic signal controls the servo to rotate 90 degrees each time, and after rotating three times, it automatically returns to the original position; the program statement for controlling the servo directly outputs the servo angle.

8. The active training method for a hand rehabilitation robot according to claim 7, characterized in that: In step S3, the program design content includes the second mode: when the electromyographic signal is received, the servo starts to move from the initial position to the position of 270°, and then receives the electromyographic signal again to return to the initial position, and microseconds are used as parameters to set the servo rotation angle.

Images