Smart wearable device and method for mitigating tremor

Abstract

The invention discloses an intelligent wearable device and method for relieving tremor. The intelligent wearable device comprises a wrist fixing part and a working part which are connected with each other. The wrist fixing component comprises a bracelet used for being fixed to the wrist of a user. The working component comprises a control unit, and a multi-mode sensing unit, a percutaneous sensory nerve stimulation unit and an environment sensing unit which are connected with the control unit. According to the technical scheme, the control signal is obtained through the mapping relation between the tremor mode and the stimulation parameters of the two-channel percutaneous stimulation model, the control signal is compensated according to the current scene information of the user and the current environment data, the adaptability is improved, and the tremor suppression effect is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent wearable devices, and particularly relates to an intelligent wearable device and method for relieving tremors. Background Art

[0002] In order to suppress hand movement tremors caused by Parkinson's syndrome or cerebellar diseases, drug treatment or brain surgery is usually adopted, but there are problems of side effects and surgical risks. There are also solutions using vibrating gloves or bracelets in the prior art. The invention patent CN119367657A provides a vibrating bracelet for treating tremors, which collects the motion state signals of the target position corresponding to the vibrating bracelet, generates target vibration parameters according to the motion state signals, and adjusts the target vibration action for treating tremors applied by the execution part according to the target vibration parameters. However, the prior art usually only controls according to the motion state of the target position and electrocorticogram of the motor cortex, and the adaptability of the solution is insufficient, unable to adapt to changes in the external environment, and the recognition of the user's scenario is not accurate enough, resulting in a deteriorated effect. Summary of the Invention

[0003] The present invention provides an intelligent wearable device for relieving tremors. The device includes a wrist fixing component and a working component connected to each other. The wrist fixing component includes a bracelet for fixing on the user's wrist. The working component includes a control unit, a multimodal sensing unit connected to the control unit, a transcutaneous sensory nerve stimulation unit, and an environment sensing unit.

[0004] The control unit includes a TAPS control module with a built-in transcutaneous sensory nerve stimulation TAPS control model, a first analysis and recognition module for recognizing the user's tremor pattern, and a second analysis and recognition module for recognizing the user's motion scenario.

[0005] The multimodal sensing unit includes an integrated MEMS inertial sensor for real-time collecting the hand movement acceleration and sending it to the first analysis and recognition module and the second analysis and recognition module. It also includes a flexible surface electromyography sensor array and a skin conductivity sensing module for respectively real-time collecting the muscle activation timing sequence and the skin conductivity signal and sending them to the first analysis and recognition module.

[0006] The environment sensing unit includes a barometric pressure sensor and a temperature sensor for respectively real-time collecting the ambient air pressure and temperature and sending them to the TAPS control module.

[0007] The first analysis and recognition module recognizes the tremor pattern based on the hand movement acceleration, muscle activation timing sequence, and skin conductivity signal and sends it to the TAPS control module. The TAPS control module generates a TAPS control signal based on the TAPS control model and the tremor pattern.

[0008] The second analysis and recognition module identifies the user scenario based on the hand movement acceleration and sends it to the TAPS control module, and the TAPS control module compensates the TAPS control signal based on the user scenario, ambient air pressure, and temperature;

[0009] The transcutaneous sensory nerve stimulation unit emits a TAPS signal according to the compensated TAPS control signal.

[0010] Furthermore, the TAPS control signal is a dynamic superimposed waveform signal of a high-frequency suppression signal and a low-frequency regulation signal; the TAPS signal is a micro-current signal less than 20 mA.

[0011] Furthermore, the specific identification of the tremor pattern includes: establishing an acceleration signal feature vector according to the hand movement acceleration information, obtaining electromyogram feature parameters according to the muscle activation timing information, obtaining skin conductivity feature parameters according to the skin conductivity signal, and using the acceleration signal feature vector, electromyogram feature parameters, and skin conductivity feature parameters as input vectors, and using the sign function to obtain the tremor pattern recognition result.

[0012] Furthermore, the TAPS control module generating the TAPS control signal based on the dual-channel TAPS control model and the tremor pattern specifically includes constructing a dynamic mapping relationship between the tremor pattern and the stimulation parameters of the dual-channel TAPS control model based on the TS-LSTM network, where the stimulation parameters include stimulation intensity and stimulation frequency; generating the TAPS control signal before compensation based on the dual-channel TAPS control model according to the above mapping relationship.

[0013] Furthermore, compensating the TAPS control signal based on the user scenario, ambient air pressure, and temperature specifically includes: obtaining the scenario compensation value C according to the user scenario s , obtaining the environmental compensation value C according to the ambient air pressure and temperature e , using the scenario compensation value C s and the environmental compensation value C e , compensating the TAPS control signal before compensation to obtain the compensated TAPS control signal.

[0014] The present invention also relates to a control method for an intelligent wearable device for relieving tremors. The method is used for the above-mentioned intelligent wearable device for relieving tremors, and the method includes the following steps:

[0015] S1. Establish a dual-channel transcutaneous sensory nerve stimulation TAPS control model;

[0016] S2. The integrated MEMS inertial sensor in the multimodal sensing unit collects the hand movement acceleration in real time and sends it to the first analysis and recognition module and the second analysis and recognition module; the flexible surface electromyography sensor array and the skin conductivity sensing module in the multimodal sensing unit collect the muscle activation timing and the skin conductivity signal in real time respectively and send them to the first analysis and recognition module;

[0017] S3. The barometric pressure sensor and the temperature sensor in the environmental sensing unit collect the environmental barometric pressure and temperature in real time respectively and send them to the TAPS control module;

[0018] S4. The first analysis and recognition module identifies the tremor pattern based on the hand movement acceleration, muscle activation timing, and skin conductivity signal and sends it to the TAPS control module. The TAPS control module generates a TAPS control signal based on the dual-channel TAPS control model and the tremor pattern;

[0019] S5. The second analysis and recognition module identifies the user scenario based on the hand movement acceleration and sends it to the TAPS control module. The TAPS control module compensates the TAPS control signal based on the user scenario, environmental barometric pressure, and temperature;

[0020] S6. The transcutaneous sensory nerve stimulation unit emits a TAPS signal according to the compensated TAPS control signal.

[0021] Further, the specific steps of identifying the tremor pattern in step S4 include the following steps:

[0022] S411. Establish an acceleration signal feature vector according to the hand movement acceleration information

[0023] S412. Obtain the electromyography feature parameters according to the muscle activation timing information

[0024] S413. Obtain the skin conductivity feature parameters according to the skin conductivity signal,

[0025] S414. Use the acceleration signal feature vector, electromyography feature parameters, and skin conductivity feature parameters as input vectors and use the sign function to obtain the tremor pattern recognition result.

[0026] Further, the specific steps of generating the TAPS control signal based on the dual-channel TAPS control model and the tremor pattern in step S4 include the following steps:

[0027] S421. Build a dynamic mapping relationship between the tremor pattern and the stimulation parameters of the dual-channel TAPS control model. The stimulation parameters include stimulation intensity and stimulation frequency;

[0028] S422. Based on the above dual-channel TAPS control model, generate a pre-compensation TAPS control signal according to the above mapping relationship.

[0029] The present invention also relates to a computer program product, which includes a computer program. The computer program is executed by a processor and is used to execute the above-mentioned control method for the intelligent wearable device for relieving tremors.

[0030] The present invention also relates to a computer-readable storage medium, which is used to store a computer program. The computer program is executed by a processor and is used to execute the above-mentioned control method for the intelligent wearable device for relieving tremors.

[0031] The technical solution of the present invention obtains a control signal through the mapping relationship between the tremor mode and the stimulation parameters of the dual-channel transcutaneous stimulation model, and compensates the control signal according to the current scenario information and current environmental data of the user, improving the adaptability and the tremor suppression effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a structural block diagram of the working components in the intelligent wearable device of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0033] The present invention will be further described below with reference to the drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention and cannot be used to limit the protection scope of the present invention. It should be pointed out that the following detailed description is exemplary and is intended to provide further description of the present invention.

[0034] Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs. It should be noted that the terms used herein are only for describing the specific embodiments and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0035] Embodiment 1 of the present invention relates to an intelligent wearable device for relieving tremors. The device includes a wrist fixing component and a working component connected to each other. The wrist fixing component includes a bracelet for fixing to the user's wrist. As shown in the Figure 1 drawing, the working component includes a control unit, a multimodal sensing unit, a transcutaneous sensory nerve stimulation unit, and an environmental sensing unit connected to the control unit.

[0036] The control unit includes a TAPS control module for presetting a dual-channel transcutaneous sensory nerve stimulation (TAPS) control model, a first analysis and recognition module for identifying the user's tremor pattern, and a second analysis and recognition module for identifying the user's motion scenario.

[0037] The dual-channel transcutaneous sensory nerve stimulation (TAPS) control model is specifically as follows:

[0038]

[0039] Wherein, S (t) is the TAPS control signal, I1 and I2 are the stimulation intensities of the high-frequency suppression signal channel and the low-frequency adjustment signal channel respectively, t is the pulse width, f1 and f2 are the stimulation frequencies corresponding to the high-frequency suppression signal channel and the low-frequency adjustment signal channel respectively, α and β are weight coefficients, is the phase difference parameter, which is used to optimize the signal superposition effect. The frequency range of the high-frequency suppression signal is 80 Hz - 120 Hz, and the frequency range of the low-frequency adjustment signal is 5 Hz - 30 Hz. The TAPS control signal is a dynamic superposition waveform signal of the high-frequency suppression signal and the low-frequency adjustment signal.

[0040] The multi-modal sensing unit includes an integrated MEMS inertial sensor for real-time acquisition of the hand movement acceleration and sending it to the first analysis and recognition module and the second analysis and recognition module. It also includes a flexible surface electromyography sensor array and a skin conductivity sensing module for real-time acquisition of the muscle activation timing and the skin conductivity signal respectively and sending them to the first analysis and recognition module.

[0041] The environmental sensing unit includes a barometric pressure sensor and a temperature sensor, which respectively acquire the environmental barometric pressure and temperature in real-time and send them to the TAPS control module.

[0042] The first analysis and recognition module identifies the tremor pattern based on the hand movement acceleration, muscle activation timing, and skin conductivity signal and sends it to the TAPS control module. The TAPS control module generates a TAPS control signal based on the dual-channel TAPS control model and the tremor pattern.

[0043] The specific identification of the tremor pattern includes:

[0044] Establish an acceleration signal feature vector F according to the hand movement acceleration information,

[0045]

[0046] Wherein, are the average accelerations in the x, y, and z directions respectively, are the variances of the accelerations in the x, y, and z directions respectively, max(|a x |), max(|a y |), max(|az The peak acceleration values in the x, y, and z directions are A respectively x a(k) is the acceleration in the x direction x The frequency domain feature obtained through Fourier transform, k is the frequency component serial number, A y (k), A z (k) is the same. FFT() is the Fourier transform function, Re() represents the real part of the result value, and Im represents the imaginary part of the result value

[0047] Obtain the electromyogram feature parameter R xy (τ) according to the muscle activation timing information

[0048]

[0049] where x(t) and y(t) are the activation signals of different muscle groups is the signal mean value, τ is the time delay, t is the sampling moment, and N is the total number of signal sampling points

[0050] Obtain the skin conductivity feature parameter GSR(t) according to the skin conductivity signal

[0051]

[0052] where R(t) is the skin resistance at time t, GSR(t) is the skin conductivity feature parameter at time t, and b s is the normal skin conductivity reference constant

[0053] Generate the tremor pattern recognition result y z ,

[0054] y z = sign(wX + b)

[0055] where X is the input feature vector, including the acceleration signal feature vector F, the electromyogram feature parameter R xy (τ), and the skin conductivity feature parameter CSR(t), w is the weight parameter, b is the bias parameter, and sign is the sign function

[0056] The TAPS control module generates TAPS control signals based on the dual-channel TAPS control model and the tremor pattern, specifically including constructing a dynamic mapping relationship between the tremor pattern and the stimulation parameters of the dual-channel TAPS control model. The stimulation parameters include the stimulation intensity and the stimulation frequency. Specifically, for the high-frequency inhibition signal and the low-frequency regulation signal, the following method is used to establish the mapping relationship with the tremor pattern

[0057] I0(t) = I 0,base + k * TS-LSTM(yz )

[0058] f0(t) = f base + k f *TS-LSTM(y z )

[0059] where I 0,base is the basic stimulus intensity of the signal, k is the adjustment coefficient, f base is the basic frequency of the signal, k f is the frequency adjustment coefficient, I0(t) is the value of the adjusted stimulus intensity, f0(t) is the value of the adjusted stimulus frequency, TS-LSTM() is the LSTM function, and y z is the tremor pattern recognition result. Furthermore, based on the above dual-channel TAPS control model, the TAPS control signal S before compensation is generated according to the above mapping relationship (t) .

[0060] The second analysis and recognition module identifies the user scenario based on the hand movement acceleration and sends it to the TAPS control module, and the TAPS control module compensates the TAPS control signal based on the user scenario, environmental air pressure, and temperature

[0061] The user scenario specifically includes the user being in a dining, writing, or resting scenario. Compensating the TAPS control signal based on the user scenario, environmental air pressure, and temperature specifically includes obtaining the scenario compensation value C s and the environmental compensation value C e ,

[0062] C s = Classifier(a x (t), a y (t), a z (t))

[0063] C e = I base *(1 + k p (P - P0) + k T (T - T0))

[0064] where Classifier() is the classifier function, a x (t), a y (t), a z (t) are the measured values of the hand movement acceleration in the x, y, and z directions respectively, kx, k T are the calibration coefficients of air pressure and temperature respectively, P and P0 are the current air pressure and the reference air pressure respectively, and T and T0 are the current temperature and the reference temperature respectively

[0065] Obtain the compensated TAPS control signal Scomp ,

[0066] S comp = C e S (t) + C s

[0067] Among them, S (t) is the TAPS control signal before compensation.

[0068] The transcutaneous sensory nerve stimulation unit emits a TAPS signal according to the compensated TAPS control signal. The TAPS signal is a micro-current signal less than 20 mA, which can stimulate the median nerve and radial nerve in the wrist, trigger thalamic neuron oscillations, and thus reduce hand tremors.

[0069] Embodiment 2 of the present invention relates to a control method for an intelligent wearable device for relieving tremors, and the method is used to control the intelligent wearable device described in Embodiment 1. The method includes the following steps:

[0070] S1. Establish a dual-channel transcutaneous sensory nerve stimulation TAPS control model. Specifically:

[0071]

[0072] Among them, S (t) is the TAPS control signal, I1 and I2 are the stimulation intensities of the high-frequency suppression signal channel and the low-frequency adjustment signal channel respectively, t is the pulse width, f1 and f2 are the stimulation frequencies corresponding to the high-frequency suppression signal channel and the low-frequency adjustment signal channel respectively, α and β are weight coefficients, is the phase difference parameter, which is used to optimize the signal superposition effect. The frequency range of the high-frequency suppression signal is 80 Hz - 120 Hz, and the frequency range of the low-frequency adjustment signal is 5 Hz - 30 Hz. The TAPS control signal is a dynamic superposition waveform signal of the high-frequency suppression signal and the low-frequency adjustment signal.

[0073] S2. The integrated MEMS inertial sensor in the multi-modal sensing unit collects the hand movement acceleration in real time and sends it to the first analysis and recognition module and the second analysis and recognition module; the flexible surface electromyography sensor array and the skin conductivity sensing module in the multi-modal sensing unit collect the muscle activation timing and the skin conductivity signal in real time and send them to the first analysis and recognition module.

[0074] S3. The barometric pressure sensor and the temperature sensor in the environmental sensing unit collect the environmental barometric pressure and temperature in real time and send them to the TAPS control module.

[0075] S4. The first analysis and recognition module identifies the tremor pattern based on the hand movement acceleration, muscle activation timing, and skin conductivity signal and sends it to the TAPS control module. The TAPS control module generates a TAPS control signal based on the dual-channel TAPS control model and the tremor pattern.

[0076] The specific steps for identifying the tremor pattern are as follows:

[0077] S411. Establish an acceleration signal feature vector F based on the hand movement acceleration information.

[0078]

[0079] Among them, are the average accelerations in the x, y, and z directions respectively. are the variances of the accelerations in the x, y, and z directions respectively. max(|a x |), max(|a y |), max(|a z |) are the peak accelerations in the x, y, and z directions respectively. A x (k) is the frequency domain feature obtained by performing a Fourier transform on the x-direction acceleration a x , where k is the frequency component number. A y (k) and A z (k) are the same. FFT() is the Fourier transform function, Re() represents the real part of the result value, and Im represents the imaginary part of the result value.

[0080] S412. Obtain the electromyogram feature parameter R[[ID=3 forty]] xy (τ) according to the muscle activation timing information.

[0081]

[0082] Among them, x(t), y(t) are the activation signals of different muscle groups. is the signal mean, τ is the time delay, t is the sampling moment, and N is the total number of signal sampling points.

[0083] S413. Obtain the skin conductivity feature parameter GSR(t) according to the skin conductivity signal.

[0084]

[0085] Among them, R(t) is the skin resistance at time t, GSR(t) is the skin conductivity feature parameter at time t, and b s is the normal skin conductivity reference constant.

[0086] S414. Generate the tremor pattern recognition result y z ,

[0087] y z = sign(wX + b)

[0088] Wherein, X is the input feature vector, including the acceleration signal feature vector F, the electromyogram feature parameter R xy (τ), and the galvanic skin response feature parameter GSR(t), w is the weight parameter, b is the bias parameter, and sign is the sign function.

[0089] The TAPS control module generates a TAPS control signal based on the dual-channel TAPS control model and the tremor pattern, which specifically includes the following steps:

[0090] S421. Based on the TS-LSTM network, construct a dynamic mapping relationship between the tremor pattern and the stimulation parameters of the dual-channel TAPS control model. The stimulation parameters include the stimulation intensity and the stimulation frequency.

[0091] Specifically, for the high-frequency suppression signal and the low-frequency regulation signal, the following method is used to establish the mapping relationship with the tremor pattern.

[0092] I0(t) = I 0,base + k * TS-LSTM(y z )

[0093] f0(t) = f base + k f * TS-LSTM(y z )

[0094] Wherein, I 0,base is the basic stimulation intensity of the signal, k is the adjustment coefficient, f base is the basic frequency of the signal, k f is the frequency adjustment coefficient, I0(t) is the value of the adjusted stimulation intensity, f0(t) is the value of the adjusted stimulation frequency, TS-LSTM() is the LSTM function, and y z is the tremor pattern recognition result.

[0095] S422. Based on the above dual-channel TAPS control model, generate the TAPS control signal S before compensation according to the above mapping relationship. (t) .

[0096] S5. The second analysis and recognition module identifies the user scenario according to the hand movement acceleration and sends it to the TAPS control module. The TAPS control module compensates the TAPS control signal based on the user scenario, the ambient air pressure, and the temperature.

[0097] The user scenario specifically includes that the user is in a diet, writing, or resting scenario. Compensating the TAPS control signal based on the user scenario, the ambient air pressure, and the temperature specifically includes obtaining the scenario compensation value Cs and the environmental compensation value C e ,

[0098] C s = Classifier(a x (t), a y (t), a z (t))

[0099] C e = I base *(1 + k p (P - P0)+k T (T - T0))

[0100] wherein, Classifier() is a classifier function, a x (t), a y (t), a z (t) are respectively the measured values of the hand movement acceleration in the x, y, and z directions, k p and k T are respectively the calibration coefficients of air pressure and temperature, P and P0 are the current air pressure and the reference air pressure respectively, and T and T0 are the current temperature and the reference temperature respectively.

[0101] Obtain the compensated TAPS control signal S comp ,

[0102] S comp = C e S (t) + C s

[0103] wherein, S (t) is the TAPS control signal before compensation.

[0104] S6. The transcutaneous sensory nerve stimulation unit emits a TAPS signal according to the compensated TAPS control signal. The TAPS signal is a microcurrent signal less than 20 mA, which can stimulate the median nerve and the radial nerve in the wrist, trigger thalamic neuron oscillation, and thus reduce hand tremors.

[0105] Embodiment 3 of the present invention relates to a computer program product, the computer program product includes a computer program, and the computer program is executed by a processor for executing the control method of the intelligent wearable device for relieving tremors in Embodiment 2.

[0106] Embodiment 4 of the present invention relates to a computer-readable storage medium, the computer-readable storage medium is used to store a computer program, and the computer program is executed by a processor for executing the control method of the intelligent wearable device for relieving tremors in Embodiment 2.

[0107] As described above, this is only the preferred embodiment of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the technical principle of the present invention, several improvements and modifications can be made, and these improvements and modifications should also be regarded as the protection scope of the present invention.

Claims

1. An intelligent wearable device for relieving tremors, characterized in that, The device includes a wrist fixation component and a working component connected to each other; the wrist fixation component includes a bracelet for fixing to the user's wrist; the working component includes a control unit, a multimodal sensing unit, a transcutaneous sensory nerve stimulation unit, and an environmental sensing unit connected to the control unit; The control unit includes a TAPS control module with a built-in transcutaneous sensory nerve stimulation TAPS control model, a first analysis and recognition module for identifying the user's tremor pattern, and a second analysis and recognition module for identifying the user's motion scenario; The multimodal sensing unit includes an integrated MEMS inertial sensor that real-time collects the hand movement acceleration and sends it to the first analysis and recognition module and the second analysis and recognition module. It also includes a flexible surface electromyography sensor array and a skin conductivity sensing module that respectively real-time collect the muscle activation timing sequence and skin conductivity signal and send them to the first analysis and recognition module; The environmental sensing unit includes a barometric pressure sensor and a temperature sensor, which respectively real-time collect the environmental barometric pressure and temperature and send them to the TAPS control module; The first analysis and recognition module identifies the tremor pattern based on the hand movement acceleration, muscle activation timing sequence, and skin conductivity signal and sends it to the TAPS control module. The TAPS control module generates a TAPS control signal based on the TAPS control model and the tremor pattern; The second analysis and recognition module identifies the user scenario based on the hand movement acceleration and sends it to the TAPS control module. The TAPS control module compensates the TAPS control signal based on the user scenario, environmental barometric pressure, and temperature; The transcutaneous sensory nerve stimulation unit emits a TAPS signal according to the compensated TAPS control signal.

2. The intelligent wearable device for relieving tremors according to claim 1, wherein The TAPS control signal is a dynamic superimposed waveform signal of a high-frequency suppression signal and a low-frequency regulation signal; the TAPS signal is a microcurrent signal less than 20 mA.

3. The intelligent wearable device for relieving tremors according to claim 1, wherein The specific process of identifying the tremor pattern includes: establishing an acceleration signal feature vector based on the hand movement acceleration information, obtaining electromyography feature parameters according to the muscle activation timing sequence information, obtaining skin conductivity feature parameters according to the skin conductivity signal, and using the acceleration signal feature vector, electromyography feature parameters, and skin conductivity feature parameters as input vectors, and using the sign function to obtain the tremor pattern recognition result.

4. The intelligent wearable device for relieving tremors according to claim 1, wherein The TAPS control module generates a TAPS control signal based on the dual-channel TAPS control model and the tremor pattern, which specifically includes constructing a dynamic mapping relationship between the tremor pattern and the stimulation parameters of the dual-channel TAPS control model based on the TS-LSTM network. The stimulation parameters include stimulation intensity and stimulation frequency; based on the dual-channel TAPS control model, generate a pre-compensation TAPS control signal according to the above mapping relationship.

5. The intelligent wearable device for relieving tremors according to claim 4, characterized in that , Compensate the TAPS control signal based on the user scenario, ambient air pressure, and temperature, specifically including: obtaining the scenario compensation value C according to the user scenario s , obtaining the environmental compensation value C according to the ambient air pressure and temperature e , using the scenario compensation value C s and the environmental compensation value C e , compensate the TAPS control signal before compensation to obtain the compensated TAPS control signal.

6. A control method for an intelligent wearable device for relieving tremors, the method being used for an intelligent wearable device for relieving tremors according to any one of claims 1-5, characterized in that, The method includes the following steps: S1. Establish a dual-channel transcutaneous sensory nerve stimulation TAPS control model; S2. The integrated MEMS inertial sensor in the multimodal sensing unit real-time collects the hand movement acceleration and sends it to the first analysis and recognition module and the second analysis and recognition module; the flexible surface electromyography sensor array and the skin conductivity sensing module in the multimodal sensing unit respectively real-time collect the muscle activation timing sequence and skin conductivity signal and send them to the first analysis and recognition module; S3. The barometric pressure sensor and temperature sensor in the environmental sensing unit respectively collect the ambient barometric pressure and temperature in real time and send them to the TAPS control module; S4. The first analysis and recognition module identifies the tremor pattern based on the hand movement acceleration, muscle activation timing, and skin conductivity signal and sends it to the TAPS control module. The TAPS control module generates a TAPS control signal based on the dual-channel TAPS control model and the tremor pattern; S5. The second analysis and recognition module identifies the user scenario based on the hand movement acceleration and sends it to the TAPS control module. The TAPS control module compensates the TAPS control signal based on the user scenario, ambient barometric pressure, and temperature; S6. The transcutaneous sensory nerve stimulation unit emits a TAPS signal according to the compensated TAPS control signal.

7. The control method according to claim 6, wherein The specific steps for identifying the tremor pattern in step S4 are as follows: S411. Establish an acceleration signal feature vector based on the hand movement acceleration information S412. Obtain the electromyography characteristic parameters according to the muscle activation timing information S413. Obtain the skin conductivity characteristic parameters according to the skin conductivity signal, S414. Use the acceleration signal feature vector, electromyography characteristic parameters, and skin conductivity characteristic parameters as input vectors, and use the sign function to obtain the tremor pattern recognition result.

8. The control method according to claim 7, characterized in that The specific steps for generating the TAPS control signal based on the dual-channel TAPS control model and the tremor pattern in step S4 are as follows: S421. Build a dynamic mapping relationship between the tremor pattern and the stimulation parameters of the dual-channel TAPS control model based on the TS-LSTM network. The stimulation parameters include stimulation intensity and stimulation frequency; S422. Based on the above dual-channel TAPS control model, generate the TAPS control signal before compensation according to the above mapping relationship.

9. A computer program product, characterized in that, The computer program product includes a computer program, and the computer program is executed by a processor for executing the control method of the intelligent wearable device for relieving tremors as claimed in claim 7 or 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program, and the computer program is executed by a processor for executing the control method of the intelligent wearable device for relieving tremors as claimed in claim 7 or 8.

Images