A hand rehabilitation glove anti-shake control device based on a millisecond-level anti-shake algorithm

Abstract

A kind of hand rehabilitation glove anti-shake control device based on millisecond level anti-shake algorithm, including glove body, multiple motion sensors, control module with built-in millisecond level anti-shake algorithm and drive execution module.Motion sensor real-time acquisition finger joint motion parameters;Control module collects data with 110ms period, calculates variance and change rate in window after sliding window filtering;When variance and change rate are lower than threshold, it is determined as pathological tremor signal and control instruction is blocked, otherwise it is determined as active motion signal and output auxiliary torque.The device can effectively distinguish between patient active motion intention and involuntary tremor, avoid false triggering, achieve millisecond level fast response, significantly improve the safety and accuracy of rehabilitation training.

Description

Technical Field

[0001] This application relates to the field of rehabilitation medical devices and intelligent control technology, specifically a hand rehabilitation glove anti-shake control device based on a millisecond-level anti-shake algorithm. Background Technology

[0002] With an aging population and rising incidence of neurological diseases such as stroke and Parkinson's disease, the number of patients with hand motor dysfunction has increased significantly. As the most frequently used organ in daily life, the functional recovery of the hand is crucial for improving patients' quality of life. Hand rehabilitation gloves, as an effective assistive training tool, have been widely researched and applied in recent years.

[0003] In the prior art, hand rehabilitation gloves typically include a glove body, a drive unit (such as a pneumatic muscle, shape memory alloy, or motor), and a simple control unit. For example, a Chinese patent application discloses a pneumatic hand rehabilitation glove that controls the airflow through a preset program to drive the fingers to perform flexion and extension movements. Other solutions use flexion sensors mounted on the glove to detect the patient's active movement intentions of the fingers, thereby triggering auxiliary power.

[0004] However, patients with hand dysfunction, especially those with spastic or Parkinson's disease, often experience involuntary, high-frequency, low-amplitude or low-frequency, high-amplitude hand tremors. Existing rehabilitation glove control solutions generally lack the ability to effectively identify and process these pathological tremors. When patients attempt to exert force, their hand tremor signals are easily misinterpreted by sensors as valid movement intentions, causing the actuator to produce incorrect, contradictory assistive movements. This not only affects the effectiveness of rehabilitation training but, in severe cases, may even cause secondary injury or exacerbate their feelings of frustration. Summary of the Invention

[0005] This application provides a hand rehabilitation glove anti-shake control device based on a millisecond-level anti-shake algorithm, which can identify and filter out pathological tremor signals of the patient's hand in real time, accurately capture their active movement intentions, and effectively solve the problems of existing rehabilitation gloves being easily affected by tremors and having inaccurate control in the background art.

[0006] To achieve the above objectives, this application provides the following technical solution: a hand rehabilitation glove anti-shake control device based on a millisecond-level anti-shake algorithm, comprising: a glove body; multiple motion sensors, distributedly disposed on each finger portion of the glove body, for real-time acquisition of motion parameters of each finger joint; a control module electrically connected to the multiple motion sensors, incorporating a millisecond-level anti-shake algorithm for receiving and processing the motion parameters; and a drive execution module electrically connected to the control module and disposed on the glove body at positions corresponding to each finger; the control module is configured to: perform high-frequency filtering and trend analysis on the motion parameters acquired within a continuous time window based on the millisecond-level anti-shake algorithm to distinguish between the patient's active motion signals and pathological tremor signals, and only when determined to be an active motion signal, control the drive execution module to output a corresponding auxiliary torque. This solution, through rapid processing at the algorithm level, achieves precise shielding of tremor signals and rapid response to intention signals, significantly improving the control accuracy and safety of the device.

[0007] Furthermore, the motion sensor includes an inertial measurement unit and a bending sensor; the inertial measurement unit is used to collect the angular velocity and acceleration parameters of the fingers, and the bending sensor is used to collect the bending angle parameters of the finger joints. By fusing multiple sensor data, richer judgment criteria are provided for the anti-shake algorithm, thereby improving the accuracy of shake recognition.

[0008] Furthermore, the millisecond-level anti-shake algorithm includes: acquiring the motion parameters in real time with a sampling period of 1-10 milliseconds; performing sliding window filtering on the motion parameters to filter out high-frequency noise; calculating the variance and rate of change of the motion parameters within the sliding window; when the variance is less than a first threshold and the rate of change is less than a second threshold, determining the current signal as a pathological jitter signal and blocking the output of control commands; when the variance is greater than or equal to the first threshold and the rate of change is greater than or equal to the second threshold, determining it as an active motion signal. This algorithm has low computational complexity and low latency, and can complete signal identification within milliseconds, ensuring the real-time performance and smoothness of rehabilitation training.

[0009] Furthermore, the control module also includes an adaptive adjustment unit, used to dynamically adjust the first threshold and the second threshold based on the patient's historical motion data. This unit can personalize parameters for different patients and different stages of rehabilitation, further optimizing the tremor stabilization effect.

[0010] Furthermore, the drive actuation module is a shape memory alloy wire arranged longitudinally along each finger and contracts when energized to drive the finger to bend. Shape memory alloys have the advantages of good flexibility, light weight, and quiet operation, improving the wearing comfort and safety of rehabilitation gloves.

[0011] Furthermore, the drive actuation module is a miniature pneumatic artificial muscle, fixed to the back of the hand side of the glove body. The pneumatic artificial muscle can provide significant driving force, making it suitable for patients in the early stages of rehabilitation who require high-intensity assistance.

[0012] Furthermore, it also includes an electromyography (EMG) sensor, installed on the wrist or back of the hand of the glove body, to collect EMG signals from relevant muscle groups in the patient's forearm. Using EMG signals as an auxiliary basis for judging movement intention can help identify movement intention in advance when the patient's hand muscle strength is extremely weak and the motion sensor signal is not obvious, thus improving the sensitivity of the device.

[0013] Furthermore, it also includes a feedback module connected to the control module, used to provide vibration or sound cues to the patient when a shaking signal is detected. This feedback helps the patient consciously control hand stability, forming a biofeedback training loop and accelerating the rehabilitation process.

[0014] Furthermore, the control module is encapsulated within a detachable wristband-style control box, and the glove body is electrically connected to the control box via a flexible circuit board or wires. This design facilitates the cleaning, maintenance, and upgrade / replacement of the control module.

[0015] Furthermore, it also includes a wireless communication module for uploading the patient's motion data and image stabilization results to an external terminal. This provides data support for doctors to remotely assess the patient's rehabilitation progress and adjust the rehabilitation plan.

[0016] Compared with the prior art, the beneficial effects of this application are: 1. This application, by incorporating a millisecond-level anti-shake algorithm into the control module, can perform real-time filtering and trend analysis on high-frequency collected motion parameters, accurately distinguish between the patient's active movement intention and pathological unconscious shaking, effectively avoid erroneous auxiliary movements caused by misjudgment of shaking, and significantly improve the safety and effectiveness of rehabilitation training.

[0017] 2. This application adopts a dual threshold judgment mechanism of sliding window variance and rate of change. The algorithm is simple and efficient, and can complete signal identification within millisecond delay, ensuring that the rehabilitation gloves respond quickly to the patient's intentions and making rehabilitation movements smoother and more natural.

[0018] 3. By setting up an adaptive adjustment unit and integrating an electromyography signal sensor, this application can dynamically adjust the anti-shake parameters or provide auxiliary judgment according to individual patient differences and real-time physiological state, which enhances the intelligence and clinical applicability of the device and helps to meet the diverse needs of different rehabilitation stages. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the overall structure of the hand rehabilitation glove anti-shake control device based on millisecond-level anti-shake algorithm in Embodiment 1 of this application.

[0020] Figure 2 This is a block diagram of the control system principle of this application.

[0021] Figure 3 This is a flowchart illustrating the millisecond-level anti-shake algorithm of this application.

[0022] In the diagram: 1. Glove body; 2. Bending sensor; 3. Inertial measurement unit; 4. Electromyography (EMG) sensor; 5. Drive execution module (shape memory alloy wire); 6. Control module (wristband control box); 7. Feedback module; 8. Wireless communication module; 9. Finger area; 10. Back of hand. Detailed Implementation

[0023] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0024] In the description of this application, any descriptions of orientation, such as up, down, front, back, left, right, etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are merely for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. When a feature is referred to as being set, fixed, or connected to another feature, it can be directly set, fixed, or connected to the other feature, or it can be indirectly set, fixed, or connected to the other feature.

[0025] Please see Figure 1 , Figure 2 and Figure 3 This application provides the following technical solutions: A hand rehabilitation glove anti-shake control device based on a millisecond-level anti-shake algorithm includes: a glove body 1; multiple motion sensors, distributedly disposed on each finger portion 9 of the glove body 1, for real-time acquisition of motion parameters of each finger joint; a control module 6, electrically connected to the multiple motion sensors, with a built-in millisecond-level anti-shake algorithm, for receiving and processing the motion parameters; and a drive execution module 5, electrically connected to the control module 6, and disposed on the glove body 1 at the position corresponding to each finger.

[0026] Specifically, in this embodiment, the motion sensor includes an inertial measurement unit (IMU) 3 (e.g., an MPU6050, used to acquire angular velocity and acceleration) and a bending sensor 2 (e.g., a flexible resistive bending sensor, used to acquire the bending angle of the finger joints). The core of the control module 6 is an STM32F407 series microcontroller, which has a high operating frequency and a DSP instruction set, suitable for running complex filtering and anti-shake algorithms. The drive execution module 5 is a 0.2mm diameter titanium-nickel alloy shape memory alloy wire, which is tightly sewn along the longitudinal direction of each finger (from fingertip to finger base) on the inner side of the glove body.

[0027] The control module 6 is configured to: perform high-frequency filtering and trend analysis on the motion parameters collected within a continuous time window based on the millisecond-level anti-shake algorithm, so as to distinguish between the patient's active motion signal and pathological tremor signal, and only when it is determined to be an active motion signal, control the drive execution module 5 to be energized and contracted, and output the corresponding auxiliary torque to help the fingers complete the expected action.

[0028] The specific steps of the millisecond-level anti-shake algorithm are as follows (please refer to the following carefully). Figure 3 ): 1. Data Acquisition: The control module 6 reads the data from the bending sensor 2 and the inertial measurement unit 3 in real time with a sampling period of 5 milliseconds.

[0029] 2. Preprocessing: Apply sliding window (window width set to 3 sampling points) mean filtering to the raw data to eliminate circuit noise.

[0030] 3. Feature Calculation: Calculate the variance of the bending angle signal (characterizing jitter amplitude) and the rate of change of the angular velocity signal (characterizing jitter frequency) within the current sliding window (e.g., a data window containing the most recent 20 sampling points, i.e., a 100-millisecond data window).

[0031] 4. Intent judgment: When the calculated variance is less than the first threshold (preset to 0.5) and the rate of change is less than the second threshold (preset to 50 degrees / second), the current signal is determined to originate from involuntary pathological tremors of the hand. At this time, the control module 6 blocks the output of control commands, and the drive execution module 5 remains in its current state without taking any action.

[0032] Conversely, when the variance reaches the first threshold and the rate of change reaches the second threshold, it is determined that the patient has a valid intention to move. The control module 6 then calculates the corresponding PWM duty cycle based on the bending angle and drives the shape memory alloy wire to contract to the corresponding length.

[0033] 5. Loop: Enter the next sampling period.

[0034] In use: The patient wears the hand rehabilitation glove on the affected hand and secures the wristband control box 6 to the wrist. After activating the device, the control module 6 begins high-speed acquisition of data from various sensors at millisecond intervals. When the patient experiences involuntary finger tremors due to spasticity or Parkinson's disease, the anti-shake algorithm quickly calculates extremely low variance and rate of change, classifying it as an invalid tremor. The drive module 5 remains inactive, and the glove does not provide incorrect assistance, thus avoiding the risk of the hand being pulled incorrectly. When the patient consciously attempts to bend a finger, the bending sensor 2 and IMU 3 detect continuous, unidirectional changes in angle and speed. The variance and rate of change calculated by the algorithm then exceed the threshold, classifying it as a valid intention. The control module 6 immediately outputs current, energizing and contracting the shape memory alloy wire 5 on the corresponding finger to assist the patient in completing the finger bending action. Simultaneously, the feedback module 7 emits a slight beep each time a tremor is filtered out, reminding the patient to relax. All raw data and anti-shake processing results during the rehabilitation process are uploaded in real-time to the doctor's tablet via the wireless communication module 8 (e.g., Bluetooth) for assessing rehabilitation progress.

[0035] It is worth noting that the sampling period (5 milliseconds), sliding window width (20 points), variance threshold (0.5 degrees), and rate of change threshold (50 degrees / second) disclosed in this embodiment are initial values ​​set for typical mild to moderate Parkinson's disease patients. In practical applications, the control module 6 contains an adaptive adjustment unit that can dynamically optimize these threshold parameters under the authorization of a rehabilitation physician or by self-learning based on the patient's continuous week of exercise data. The core microcontroller is an STM32F407 series, and the control of the shape memory alloy wire (such as PWM power adjustment via MOSFET switching transistors) and the wireless communication protocol (such as Bluetooth 4.0) both adopt methods commonly used in existing technologies.

[0036] Although embodiments of this application have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of this application, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A hand rehabilitation glove anti-shake control device based on a millisecond-level anti-shake algorithm, characterized in that, include: The glove body (1) includes multiple motion sensors distributed across the finger sections (9) of the glove body (1) for real-time acquisition of motion parameters of each finger joint; and a control module (6) electrically connected to the multiple motion sensors, with a built-in millisecond-level anti-shake algorithm for receiving and processing the motion parameters. The drive execution module (5) is electrically connected to the control module (6) and is set on the glove body (1) at the position corresponding to each finger. The control module (6) is configured to: perform high-frequency filtering and trend analysis on the motion parameters collected within a continuous time window based on the millisecond-level anti-shake algorithm, so as to distinguish between the patient's active motion signal and pathological tremor signal, and only when it is determined to be an active motion signal, control the drive execution module (5) to output the corresponding auxiliary torque.

2. The hand rehabilitation glove anti-shake control device based on millisecond-level anti-shake algorithm according to claim 1, characterized in that: The motion sensor includes an inertial measurement unit (3) and a bending sensor (2); the inertial measurement unit (3) is used to collect the angular velocity and acceleration parameters of the finger, and the bending sensor (2) is used to collect the bending angle parameters of the finger joint.

3. The hand rehabilitation glove anti-shake control device based on millisecond-level anti-shake algorithm according to claim 1, characterized in that: The millisecond-level anti-shake algorithm includes: acquiring the motion parameters in real time with a sampling period of 1-10 milliseconds; performing sliding window filtering on the motion parameters to filter out high-frequency noise; calculating the variance and rate of change of the motion parameters within the sliding window; when the variance is less than a first threshold and the rate of change is less than a second threshold, determining that the current signal is a pathological jitter signal and blocking the output of control commands; when the variance is greater than or equal to the first threshold and the rate of change is greater than or equal to the second threshold, determining that it is an active motion signal.

4. The hand rehabilitation glove anti-shake control device based on millisecond-level anti-shake algorithm according to claim 3, characterized in that: The control module (6) further includes an adaptive adjustment unit for dynamically adjusting the first threshold and the second threshold based on the patient's historical motion data.

5. The hand rehabilitation glove anti-shake control device based on millisecond-level anti-shake algorithm according to claim 1, characterized in that: The drive execution module (5) is a shape memory alloy wire, which is arranged along the longitudinal direction of each finger and contracts when energized to drive the finger to bend.

6. The hand rehabilitation glove anti-shake control device based on millisecond-level anti-shake algorithm according to claim 1, characterized in that: The drive execution module (5) is a miniature pneumatic artificial muscle, which is fixed to the back of the hand (10) of the glove body (1).

7. The hand rehabilitation glove anti-shake control device based on millisecond-level anti-shake algorithm according to claim 1, characterized in that: It also includes an electromyography (EMG) sensor (4), which is set on the wrist or back of the hand of the glove body (1) to collect EMG signals of the relevant muscle groups of the patient's forearm as an auxiliary basis for judging the intention of movement.

8. The hand rehabilitation glove anti-shake control device based on millisecond-level anti-shake algorithm according to claim 1, characterized in that: It also includes a feedback module (7), which is connected to the control module (6) and is used to provide vibration or sound prompts to the patient when a shaking signal is detected.

9. The hand rehabilitation glove anti-shake control device based on millisecond-level anti-shake algorithm according to claim 1, characterized in that: The control module (6) is encapsulated in a detachable wristband-type control box, and the glove body (1) is electrically connected to the control box via a flexible circuit board or wires.

10. The hand rehabilitation glove anti-shake control device based on millisecond-level anti-shake algorithm according to claim 1, characterized in that: It also includes a wireless communication module (8) for uploading the patient’s motion data and stabilization processing results to an external terminal.

Images