Rehabilitation instrument motion mode control method and system

Abstract

The invention relates to the technical field of rehabilitation medical treatment, in particular to a rehabilitation instrument motion mode control method and system, and the method comprises the steps: obtaining motion angle data and confrontation moment data of a user; determining whether the deviation between the actual motion track and a preset path exceeds a preset threshold value, determining whether the moment actively emitted by the user presents a continuous attenuation trend, and judging whether the motion performance of the user declines; if it is judged that the athletic performance of the user is reduced, diagnostic mechanical stimulation is applied; collecting mechanical response data; according to the mechanical response data, judging whether a counterforce source is active force reduction or non-autonomous muscle stretching reflex; if yes, the mode of the rehabilitation instrument is switched into a vibration assisting mode and a low-speed movement mode; and if the active force is reduced, switching the mode of the rehabilitation instrument to a force compensation mode. The intelligent rehabilitation instrument has the advantages that the movement mode of the rehabilitation instrument is intelligently adjusted, and the pertinence, safety and effectiveness of rehabilitation training are improved.

Description

Technical Field

[0001] The present invention relates to the field of rehabilitation medical technology, and in particular to a method and system for controlling a motion mode of a rehabilitation apparatus. Background Art

[0002] In the field of telerehabilitation, existing rehabilitation devices struggle to accurately sense and intelligently respond to changes in the user's physiological state when executing preset exercise patterns. Specifically, when a user's performance declines, the device typically intervenes by simply increasing the assist torque, but is unable to effectively distinguish whether the "counter torque" is due to muscle fatigue (lack of active force) or an involuntary muscle stretch reflex caused by increased muscle tone. This lack of information recognition can lead to the device mistakenly increasing the assist force continuously, exacerbating the involuntary stimulation of the user's muscles and creating a vicious cycle of "increased assist - enhanced reflex - enhanced counterforce." This not only reduces the effectiveness of rehabilitation training but can also cause discomfort or even muscle damage.

[0003] In view of the above problems, the existing technology is in urgent need of improvement. Summary of the Invention

[0004] In order to address the shortcomings of the existing technology, the present application provides a rehabilitation instrument movement mode control method and system, which has the advantages of intelligently adjusting the movement mode of the rehabilitation instrument, reducing the negative impact that may be caused by the blind increase of auxiliary force by traditional rehabilitation instruments, thereby improving the targetedness, safety and effectiveness of rehabilitation training.

[0005] The present application provides a method for controlling a rehabilitation apparatus movement mode, the method comprising: Obtain the user's motion angle data and resistance torque data; Determine whether the deviation between the actual motion trajectory and the preset path exceeds a preset threshold based on the motion angle data, and determine whether the torque actively exerted by the user shows a continuous attenuation trend based on the counter torque data to determine whether the user's athletic performance has declined; If it is determined that the user's athletic performance has declined, the auxiliary force output form of the rehabilitation device is changed and diagnostic mechanical stimulation is applied. The diagnostic mechanical stimulation includes superimposing a vibration component of a preset frequency and preset amplitude on the original basic driving torque of the rehabilitation device; collecting mechanical response data of the user during application of diagnostic mechanical stimulation; Based on the mechanical response data, determine whether the source of the user's resistance force is a reduction in active force or an involuntary muscle stretch reflex; If the source of the antagonistic force is determined to be an involuntary muscle stretch reflex, the rehabilitation device is switched to a vibration-assisted and low-speed motion mode. The vibration-assisted and low-speed motion modes include continuously superimposing a vibration component in the auxiliary force output, reducing the movement speed of the task target, and simplifying the complexity of the preset path. If the source of the antagonistic force is determined to be a reduction in active force, the mode of the rehabilitation device is switched to a force compensation mode, which includes increasing the constant value of the basic propulsion torque.

[0006] Through the above solution, the deep-seated reasons for the decline in the user's athletic performance can be accurately determined, and the exercise mode of the rehabilitation device can be intelligently adjusted accordingly, avoiding the negative effects that may be caused by the blind increase of auxiliary force by traditional rehabilitation devices, thereby improving the targetedness, safety and effectiveness of rehabilitation training.

[0007] Furthermore, the present application also proposes collecting the user's mechanical response data during the application of diagnostic mechanical stimulation, including: Obtaining the natural frequency and natural phase of the user's intrinsic limb tremor; Applying an anti-phase vibration torque with the same frequency as the natural frequency and opposite phase to the natural phase to suppress the intrinsic limb tremor; The user's torque data is collected as mechanical response data.

[0008] Furthermore, the present application also proposes: Continuously update the natural frequency and natural phase in a rolling time window manner; After each update of the natural frequency and natural phase, the frequency and phase of the anti-phase vibration torque are synchronously adjusted so that the frequency is consistent with the natural frequency and the phase is opposite to the natural phase; According to the changing trend of the intrinsic frequency and intrinsic phase in the rolling time window, it is judged whether the intrinsic limb tremor is in a stable state; If it is determined to be a stable state, the parameters of the current anti-phase vibration torque are kept unchanged; If it is determined to be an unstable state, the natural frequency and natural phase are updated by rolling the time window, and the frequency and phase of the anti-phase vibration torque are synchronously adjusted based on the updated natural frequency and natural phase.

[0009] Furthermore, the present application also proposes: Determine whether there is a residual tremor component in the torque response after the user applies the anti-phase vibration torque; If there is a residual tremor component, constructing a compensation signal based on the frequency and phase information of the residual tremor component; Based on the compensation signal, the output amplitude and phase of the anti-phase vibration torque are adaptively adjusted.

[0010] Furthermore, the present application also proposes that after switching the mode of the rehabilitation device to the vibration-assisted and low-speed exercise modes, the following steps are also included: Monitor the user's actual motion deviation value along the preset path; Determining whether the actual motion deviation value is continuously in a deviation stable state below a first deviation threshold; If it is determined to be in a stable state, the complexity of the preset path and the movement speed of the task target are gradually increased according to the preset adjustment rules; If it is determined that the deviation has not reached a stable state, the current preset path and movement speed settings are maintained.

[0011] Furthermore, the present application also proposes that after switching the mode of the rehabilitation device to the vibration-assisted and low-speed exercise modes, the following steps are also included: Within the preset response determination time window, based on the actual motion trajectory, determine whether the user has a continuous deviation trajectory or periodic disturbance response that is inconsistent with the task target direction in the vibration-assisted and low-speed motion modes; If it is determined that there is a continuous deviation trajectory or periodic disturbance response, it is determined that there is non-goal-oriented compensatory movement or abnormal behavior affected by external interference, and a prompt signal is sent to the rehabilitation personnel, and an amplitude below the interference threshold is applied to the user to verify the user's response stability and movement intention. Figure 1 The disturbance threshold is the preset user safety response upper limit torque amplitude; If the user produces an unpredictable response to the perturbation torque, the current mechanical response data is marked as a key evaluation interval, and the corresponding abnormal trajectory characteristic parameters are recorded.

[0012] Furthermore, the present application also proposes that the user is subjected to a force with an amplitude lower than the interference threshold to verify the user's response stability and movement intention. Figure 1 After the consistent perturbation torque, it also includes: Continuously monitor the user's mechanical response data and extract the user's response start time, maximum torque response time, and maximum perturbation torque time point; Calculate the time difference between the maximum torque response moment and the maximum value time point, and determine whether the time difference exceeds a preset response hysteresis threshold; If the judgment time difference exceeds the response hysteresis threshold, it is recorded as a hysteresis response event; If the cumulative number of hysteresis response events exceeds a preset threshold value during a preset number of perturbation torque application operations, the current user is determined to be a hysteresis response mode user; If the user is determined to be a user in the delayed response mode, the starting time of the response determination time window is delayed by a preset delay compensation time length for correction.

[0013] Furthermore, the present application also proposes that after a user is determined to be a user in a delayed response mode, the following steps are also included: Obtain the user's mechanical response peak amplitude, response duration, and vibration input delay during the application of the perturbation torque, and calculate the user's vibration tolerance parameter, which is used to characterize the user's sensitivity to the perturbation torque. Dynamically adjust the output characteristics of subsequent perturbation torque according to the vibration tolerance parameters, including the amplitude upper limit, rise time, continuous application time and / or start delay time; After applying the adjusted perturbation torque, the user's mechanical response data is collected, and the vibration tolerance parameters are updated based on the mechanical response data. The updated vibration tolerance parameters are used for adaptive adjustment of the output characteristics of the subsequent perturbation torque.

[0014] Furthermore, the present application also proposes that after switching the mode of the rehabilitation device to the force compensation mode, the following steps are also included: Within the preset monitoring time window, continuously collect the user's active force response data under the action of the basic pushing torque; The time-weighted muscle fatigue integral is calculated based on the difference between the active force response data and the actual output torque of the rehabilitation device; Determine whether the muscle fatigue score is lower than the recovery judgment threshold; If the muscle fatigue score is lower than the recovery judgment threshold, it is determined that the user has the potential to recover through independent force, and the mode of the rehabilitation device is switched from the force compensation mode to the preset path recovery assessment mode. The preset path recovery assessment mode uses a step-by-step reduction of the pushing torque to guide the user to complete the target action again independently.

[0015] Furthermore, the present application also proposes a rehabilitation apparatus motion mode control system, comprising: Motion and torque acquisition module, used to obtain the user's motion angle data and resistance torque data; The athletic performance evaluation module is used to determine whether the deviation between the actual motion trajectory and the preset path exceeds a preset threshold based on the motion angle data, and to determine whether the torque actively exerted by the user shows a continuous attenuation trend based on the counteracting torque data, thereby determining whether the user's athletic performance has declined; A diagnostic stimulation application module is used to change the auxiliary force output form of the rehabilitation device and apply diagnostic mechanical stimulation if it is determined that the user's athletic performance has declined. The diagnostic mechanical stimulation includes superimposing a vibration component of a preset frequency and preset amplitude on the original basic driving torque of the rehabilitation device; a response data acquisition module, for acquiring mechanical response data of the user during application of diagnostic mechanical stimulation; The resistance source judgment module is used to judge whether the source of the user's resistance is a reduction in active force or an involuntary muscle stretch reflex based on the mechanical response data; A vibration assistance mode switching module is used to switch the rehabilitation device to a vibration assistance and low-speed motion mode if the source of the antagonistic force is determined to be an involuntary muscle stretch reflex. The vibration assistance and low-speed motion modes include continuously superimposing a vibration component in the auxiliary force output, reducing the movement speed of the task target, and simplifying the complexity of the preset path. The force compensation mode switching module is used to switch the mode of the rehabilitation device to the force compensation mode if the source of the antagonistic force is determined to be a reduction in active force. The force compensation mode includes increasing the constant value of the basic pushing torque.

[0016] Through the above solution, a system is provided for realizing the above method for controlling the movement pattern of the rehabilitation apparatus. The modular design enables the method to be effectively implemented, thereby improving the integration and functionality of the system.

[0017] In summary, the present application provides a method and system for controlling the motion mode of a rehabilitation instrument. By introducing diagnostic mechanical stimulation and mechanical response data analysis, it can accurately determine the underlying causes of the user's decreased athletic performance (reduced active force or involuntary muscle stretch reflex), and intelligently switch the motion mode of the rehabilitation instrument accordingly (vibration assistance and low-speed motion mode or force compensation mode). This has the advantages of being able to accurately determine the underlying causes of the user's decreased athletic performance, and intelligently adjust the motion mode of the rehabilitation instrument accordingly, avoiding the negative effects that may be caused by the blind increase of auxiliary force by traditional rehabilitation instruments, thereby improving the targetedness, safety and effectiveness of rehabilitation training. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A flow chart of a method for controlling the movement mode of a rehabilitation device provided in this application.

[0019] Figure 2 This is a program flowchart of a rehabilitation device motion mode control system provided in this application.

[0020] In the figure: 1. Motion and torque acquisition module; 2. Motion performance evaluation module; 3. Diagnostic stimulation application module; 4. Response data acquisition module; 5. Resistance source judgment module; 6. Vibration assistance mode switching module; 7. Force compensation mode switching module. DETAILED DESCRIPTION

[0021] The technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all of the embodiments. The components of the present application generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application for which protection is claimed, but merely represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of this application.

[0022] It should be noted that similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and should not be understood as indicating or implying relative importance.

[0023] Reference Figure 1 , this application proposes a rehabilitation apparatus movement mode control method, comprising: S1000: Acquire the user's motion angle data and resistance torque data; S2000: Determining whether the deviation between the actual motion trajectory and the preset path exceeds a preset threshold based on the motion angle data, and determining whether the torque actively exerted by the user shows a continuous attenuation trend based on the counteracting torque data, thereby determining whether the user's athletic performance has declined; S3000: If it is determined that the user's athletic performance has declined, the auxiliary force output form of the rehabilitation device is changed to apply diagnostic mechanical stimulation, which includes superimposing a vibration component of a preset frequency and preset amplitude on the original basic driving torque of the rehabilitation device; S4000: collecting mechanical response data of the user during the application of diagnostic mechanical stimulation; S5000: Based on the mechanical response data, determine whether the source of the user's resistance force is a reduction in active force or an involuntary muscle stretch reflex; S6000: If the source of the antagonistic force is determined to be an involuntary muscle stretch reflex, the rehabilitation device is switched to vibration-assisted and low-speed motion mode. This includes continuously adding a vibration component to the auxiliary force output, reducing the target's motion speed, and simplifying the complexity of the pre-set path. S7000: If the source of the antagonistic force is determined to be a reduction in active force, the mode of the rehabilitation device is switched to a force compensation mode. The force compensation mode includes increasing the constant value of the basic propulsion torque.

[0024] Motion angle data refers to the angular position information of the user's joints or limbs relative to a reference coordinate system during rehabilitation training. This data can be obtained using technologies such as angle sensors, encoders, inertial measurement units (IMUs), or optical motion capture systems. For example, rotary encoders mounted on the joints of the rehabilitation device provide real-time measurement, or wearable IMU sensors provide limb posture data. Counteracting torque data refers to the torque, actively or passively generated by the user's limbs under the assistance or resistance of the rehabilitation device, that opposes the device's output torque. This data can be obtained using force sensors, torque sensors, or torque estimation based on motor current feedback. For example, direct measurement can be performed using a torque sensor on the device's output shaft, or reaction torque can be calculated using the drive motor's current and speed data. Diagnostic mechanical stimulation refers to mechanical input intended to elicit a specific physiological response from the user, thereby helping the system determine the user's underlying physiological state. This can be achieved by superimposing a vibration component of a specific frequency and amplitude on the device's original base torque. For example, while the device continuously provides assistive force, a small, periodic torque fluctuation is applied. Mechanical response data refers to the mechanical feedback generated by a user's limbs or muscles during the application of diagnostic mechanical stimulation. This data can be collected using devices such as torque sensors, electromyographic sensors, or accelerometers. For example, a rehabilitation device's torque sensor can record instantaneous torque changes during vibration stimulation, or surface electromyography (sEMG) can be used to monitor muscle electrical activity. The involuntary stretch reflex is a physiological reflex in which stretch receptors within a muscle are activated when it is stretched, causing involuntary contraction. This typically manifests as muscle stiffness, spasm, or resistance to external stretching. Vibration-assisted and low-speed exercise modes are rehabilitation training methods that target the involuntary stretch reflex. These can be achieved by continuously superimposing a vibration component on the auxiliary force output, reducing the target movement speed, and simplifying the complexity of the pre-set path. For example, while the rehabilitation device provides auxiliary force, it continuously applies low-amplitude vibration, reduces the speed of the user's required movement, and adjusts the movement path from a specific curved shape to a straight line. The force compensation mode refers to a rehabilitation training mode that aims to reduce the user's active force. It can be achieved by increasing the constant value of the basic driving torque. For example, the rehabilitation device adds a fixed, larger driving torque on the basis of the original auxiliary force.

[0025] The core innovation of this application lies in that by introducing diagnostic mechanical stimulation and collecting the user's mechanical response data, it is possible to determine whether the source of the user's antagonistic torque is a reduction in active force or an involuntary muscle stretch reflex, and based on this, switch the movement mode of the rehabilitation device, thereby avoiding misjudgment of the user's physiological state and improving the safety and effectiveness of rehabilitation training.

[0026] In some preferred embodiments, the present application is specifically implemented as follows: The rehabilitation device is equipped with a rotary encoder and a torque sensor to acquire real-time data on the user's elbow joint motion angle and the opposing torque on the device's output shaft. For example, when a user performs elbow flexion and extension exercises, the rotary encoder records the instantaneous elbow angle at a 100Hz sampling rate, while the torque sensor measures the interaction torque between the user and the device at the same sampling rate. The system continuously analyzes this data using a built-in processor. If the average deviation between the user's actual motion trajectory and the pre-set circular path exceeds 5 degrees for five consecutive seconds, and the user's actively generated torque (calculated by subtracting the assist torque from the device's output torque) shows a continuous decline of more than 20% over the past 30 seconds, the system determines that the user's athletic performance has deteriorated. At this point, the device's control unit immediately changes the assist force output pattern and applies diagnostic mechanical stimulation. Specifically, a sinusoidal vibration component with a frequency of 5Hz and an amplitude of 0.5N·m is superimposed on the device's original base torque (e.g., 2N·m). During this diagnostic mechanical stimulation, the rehabilitation device's torque sensor continuously collects mechanical response data from the user at a higher sampling rate (e.g., 200 Hz). This data includes the instantaneous torque feedback from the user's limbs in response to the vibration stimulation. The system then performs a Fourier transform analysis on the collected mechanical response data to identify any harmonic components corresponding to the diagnostic vibration frequency (5 Hz). If a torque peak appears near 5 Hz in the mechanical response data, and this peak has a specific lag relationship with the phase of the vibration stimulation, it is determined to be an involuntary muscle stretch reflex. Conversely, if no harmonic components are present in the mechanical response data, or if the torque response only demonstrates passive compliance to the underlying thrust, it is determined to be a reduction in active force. If this is determined to be an involuntary muscle stretch reflex, the rehabilitation device switches to vibration-assisted and low-speed exercise mode. In this mode, the rehabilitation device continuously superimposes a vibration component with a frequency of 3 Hz and an amplitude of 0.3 N·m on the auxiliary force output to continuously suppress muscle spasms. At the same time, the movement speed of the task target will be reduced from the original 30 degrees / second to 15 degrees / second, and the complexity of the preset path will be simplified from a curve of a specific shape to a straight reciprocating motion. If it is judged that the active force is reduced, the mode of the rehabilitation device will switch to the force compensation mode. In this mode, the rehabilitation device will increase the constant value of the basic pushing torque from 2 Nm to 4 Nm to directly compensate for the user's lack of strength and help him complete the training. Through the above technical solution, the present application can effectively solve the technical problem that the rehabilitation device cannot accurately distinguish the source of the user's antagonistic torque in remote rehabilitation.

[0027] In another embodiment of the present application, it is further proposed that S4000 includes: S4100: Obtaining the natural frequency and natural phase of the user's intrinsic limb tremor; S4200: applying an anti-phase vibration torque having the same frequency as the natural frequency and opposite phase to the natural phase to suppress the natural limb tremor; S4300: Collect the user's torque data as mechanical response data.

[0028] Among them, intrinsic limb tremor refers to the involuntary periodic vibration with a specific frequency and phase caused by physiological or pathological reasons when the user's limbs are at rest or in motion. It can be collected by various sensors such as accelerometers, electromyography sensors or force sensors, and identified and extracted through signal processing methods such as Fourier transform, wavelet analysis or adaptive filtering. Antiphase vibration torque refers to the torque with the opposite vibration direction, the same frequency and the opposite phase to the intrinsic limb tremor. It can be achieved by the actuator of the rehabilitation device, such as a motor or linear actuator, by precisely controlling the output torque. Its purpose is to reduce or eliminate the impact of intrinsic limb tremor on mechanical response data by actively canceling it out.

[0029] The solution of the present application first obtains the natural frequency and natural phase of the user's inherent limb tremor. In this way, the solution enables the rehabilitation device to perform analysis based on purer and more accurate mechanical response data when judging the source of the user's resistance force.

[0030] In some preferred embodiments, collecting the user's mechanical response data during the application of diagnostic mechanical stimulation may be specifically implemented as follows: First, to determine the natural frequency and phase of the user's intrinsic limb tremor, the rehabilitation device can be equipped with highly sensitive force sensors or accelerometers. These sensors continuously collect motion or torque data from the user's limbs during rehabilitation training. A built-in signal processing unit, such as a digital signal processor (DSP) or microcontroller, performs real-time spectral analysis on this raw data, using, for example, a fast Fourier transform (FFT) algorithm, to identify the primary periodic components in the data. By analyzing the spectral peaks, the tremor's natural frequency can be determined, and its natural phase can be determined through phase demodulation techniques. Subsequently, to apply an anti-phase vibration torque with the same frequency and opposite phase to suppress the intrinsic limb tremor, the rehabilitation device's actuator, such as a high-response servo motor, generates a vibration torque command with the same frequency but precisely opposite phase based on the natural frequency and phase calculated by the signal processing unit. The servo motor, through its driver, precisely outputs this anti-phase vibration torque to the user's limbs, thereby actively counteracting the user's intrinsic tremor. For example, if it is detected that the user's limbs are trembling at a frequency of 5 Hz and a phase of 0 degrees, the servo motor will output a vibration torque with a frequency of 5 Hz and a phase of 180 degrees to suppress the tremor. Finally, after the intrinsic limb tremor is effectively suppressed, the rehabilitation device continues to collect the user's torque data through its force sensor and uses this data as mechanical response data. Since the impact of the tremor has been significantly reduced at this time, the collected torque data can more accurately reflect the user's true physiological response under diagnostic mechanical stimulation, such as the active contraction force or stretch reflex force of the muscle, thereby providing reliable input for subsequent judgment of the source of the antagonistic force. Through the above technical solution, when collecting the user's mechanical response data, the natural frequency and natural phase of the user's intrinsic limb tremor can be first obtained, and an anti-phase vibration torque can be applied for active suppression.

[0031] In another embodiment of the present application, it is further proposed that, based on applying the anti-phase vibration torque to suppress intrinsic limb tremor, the steps of further optimizing the suppression effect include: S8000: Continuously updates the natural frequency and natural phase in a rolling time window manner; S9000: After each update of the natural frequency and natural phase, the frequency and phase of the anti-phase vibration torque are synchronously adjusted to keep it consistent with the frequency of the natural frequency and opposite to the phase of the natural phase; S10000: Determine whether the intrinsic limb tremor is in a stable state based on the changing trends of the intrinsic frequency and intrinsic phase in the rolling time window; S11000: If it is determined to be a stable state, the current anti-phase vibration torque parameter is kept unchanged; S12000: If it is determined to be an unstable state, the natural frequency and natural phase are updated continuously by means of a rolling time window, and the frequency and phase of the anti-phase vibration torque are synchronously adjusted based on the updated natural frequency and natural phase.

[0032] A rolling time window is a data processing technique that defines a fixed-length time period and continuously incorporates the latest data into the window while removing the oldest data from the window over time, thereby enabling real-time or quasi-real-time analysis of the data stream. Specifically, this can be achieved by setting a sliding window, such as a 5- or 10-second window, and recalculating the frequency and phase information of the user's limb tremor within the window every 1-second or shorter time step. A trend refers to the overall direction or pattern of changes in the natural frequency and phase values ​​over time. Specifically, this can be achieved by calculating the mean, standard deviation, slope, or linear regression analysis of the natural frequency and phase within the rolling time window to assess their volatility or drift direction. A stable state refers to a state in which the frequency and phase of the natural limb tremor remain relatively constant or fluctuate within an acceptable range over a specified timeframe. Specifically, this can be achieved by setting a threshold. For example, if the standard deviation or rate of change of the natural frequency and phase within the rolling time window is below a preset threshold, the state is considered stable. This approach aims to avoid unnecessary frequent adjustments when tremor parameters are relatively stable, thereby reducing system computational burden and energy consumption.

[0033] This application's solution effectively addresses the issue of decreased suppression effectiveness caused by time-varying intrinsic limb tremor parameters by introducing a dynamic, adaptive tremor suppression mechanism. This adaptive feedback closed-loop control enables the rehabilitation device to continuously track and respond to dynamic changes in the user's tremor, ensuring that the anti-phase vibration torque remains optimally suppressed, thereby maintaining long-term, effective tremor suppression.

[0034] In another embodiment of the present application, it is further proposed that the method further includes the following steps: S13000: Determine whether there is a residual tremor component in the torque response after the user applies the anti-phase vibration torque; S14000: If a residual tremor component exists, construct a compensation signal based on frequency and phase information of the residual tremor component; S15000: Adaptively adjust the output amplitude and phase of the anti-phase vibration torque based on the compensation signal.

[0035] The residual tremor component refers to detectable periodic fluctuations with a specific frequency and phase that remain in the user's torque response after the application of an anti-phase vibration torque, with an amplitude exceeding a preset noise threshold or a clinically acceptable level of residual tremor. Specifically, this can be achieved by performing spectral analysis on the torque response data to identify energy peaks close to the frequency of intrinsic limb tremor and assessing whether their amplitudes reach a level requiring further processing. The compensation signal refers to an auxiliary control signal constructed based on the frequency and phase information of the residual tremor component to further offset the residual tremor. Specifically, this can be achieved by a digital signal processor generating a sinusoidal or quasi-sinusoidal waveform signal with the same frequency, opposite phase, and adjustable amplitude as the residual tremor based on the real-time frequency and phase of the residual tremor. Adaptive adjustment refers to the dynamic and automatic correction of the output parameters of the anti-phase vibration torque based on the real-time feedback compensation signal to better adapt it to the user's actual tremor condition. Specifically, this can be achieved through a closed-loop control system that uses the compensation signal as an error input to drive the torque output module to adjust the amplitude and phase of the anti-phase vibration torque to minimize residual tremor.

[0036] The solution of the present application solves the problem of interference of residual tremor components on rehabilitation judgment in the prior art by conducting an in-depth analysis of the torque response of the user after the anti-phase vibration torque is applied.

[0037] In some preferred embodiments, the present application is implemented as follows: The control unit of the rehabilitation device can continuously monitor the user's torque response data after applying an anti-phase vibration torque. This data can be collected by a force sensor. To determine whether a residual tremor component is present, the control unit can perform real-time Fourier transform or wavelet analysis on the collected torque data. By analyzing the spectrum, if one or more energy peaks are found near the intrinsic limb tremor frequency, and the amplitude of these peaks exceeds a preset threshold (for example, a torque fluctuation threshold determined through clinical trials or expert experience), a significant residual tremor component can be determined. If a residual tremor component is determined to be present, the control unit further extracts the precise frequency and phase information of these residual tremor components. For example, a peak detection algorithm can be used to determine the dominant residual tremor frequency, and a cross-correlation analysis or Hilbert transform can be used to determine its phase. Based on this frequency and phase information, the control unit can construct a compensation signal. This compensation signal can be a sinusoidal waveform with the same frequency and precisely opposite phase as the residual tremor, and its initial amplitude can be set to match the residual tremor amplitude.

[0038] Subsequently, the control unit will adaptively adjust the output amplitude and phase of the anti-phase vibration torque based on this constructed compensation signal. This can be achieved through a proportional-integral-differential controller, in which the compensation signal can be used as the input or error signal of the controller. For example, if the amplitude of the residual tremor is still large, the proportional-integral-differential controller can increase the output amplitude of the anti-phase vibration torque; if the phase of the residual tremor deviates from the expected phase, the controller can fine-tune the phase of the anti-phase vibration torque to make it more accurately anti-phase with the residual tremor. This adjustment process is continuous and real-time, ensuring that the anti-phase vibration torque can dynamically adapt to any residual tremor in the user's torque response, thereby achieving continuous optimization and suppression of tremor. Through the above technical solution, the present application can identify and accurately compensate for the residual tremor components that may exist in the torque response of the user after applying the anti-phase vibration torque.

[0039] In another embodiment of the present application, it is further proposed that after switching the mode of the rehabilitation device to the vibration assistance and low-speed exercise mode, the sub-step S6000 further includes: S6200: monitoring the actual movement deviation value of the user along the preset path; S6300: Determine whether the actual motion deviation value is continuously in a deviation stable state lower than a first deviation threshold; S6400: If it is determined that the deviation is in a stable state, gradually increase the complexity of the preset path and the movement speed of the task target according to the preset adjustment rules; S6500: If it is determined that the deviation has not reached a stable state, the current preset path and movement speed settings are maintained.

[0040] The actual motion deviation value refers to the degree of deviation between the user's actual limb motion trajectory and the target preset path set by the rehabilitation device when performing a rehabilitation task. This can be obtained by measuring the user's joint angles or limb positions in real time, comparing them with the corresponding target points on the preset path, and calculating the Euclidean distance or angular difference. The first deviation threshold refers to a preset value used to determine whether the user's motion deviation is within an acceptable range. This can be personalized based on the user's rehabilitation stage, degree of limb dysfunction, or rehabilitation goals. Deviation stability refers to whether the user's actual motion deviation value remains below the first deviation threshold over a period of time with minimal fluctuation. This can be determined by taking a rolling average, calculating the standard deviation, or performing trend analysis on the actual motion deviation value. The preset adjustment rule refers to a specific strategy used to guide the difficulty of the rehabilitation program when the user's performance reaches a stable state of deviation. This may include gradually increasing the percentage of movement speed, increasing the curvature of the path curve, or introducing new movement directions. The complexity of the preset path refers to the geometric characteristics and difficulty level of the motion trajectory that the user needs to follow during the rehabilitation task. This can be expressed as the curvature of the path, the number of path segments, the path length, or the density of points on the path that require precise control. The movement speed of the task target refers to the speed at which the target point moves along the path when the rehabilitation device guides the user to complete the preset path. It can be expressed as angular velocity or linear velocity.

[0041] The solution of the present application introduces a dynamic adjustment mechanism, which enables fine-tuning intervention based on the user's actual exercise performance after the rehabilitation device switches to vibration assistance and low-speed exercise mode.

[0042] In some preferred embodiments, the present application is specifically implemented as follows: When the rehabilitation device is switched to vibration-assisted and low-speed exercise mode, the device's control unit continuously monitors the user's actual motion deviation along the preset path. Angle sensors or encoders installed at the device's joints collect real-time angular data of the user's limbs and compare it with the target angle along the preset path for the rehabilitation task. This data is then used to calculate the real-time angular deviation as the actual motion deviation. The control unit then determines whether the actual motion deviation remains in a stable state, below a first deviation threshold. Specifically, the control unit can set a time window, such as 5 seconds, and continuously collect actual motion deviation values ​​within this time window. The mean and standard deviation of the deviation values ​​within this time window are then calculated. If the mean is below a preset first deviation threshold and the standard deviation is below a preset fluctuation threshold, the deviation is determined to be stable. The first deviation threshold can be set based on the rehabilitation goal; for example, for elbow joint rehabilitation, it can be set to 5 degrees. If the control unit determines that the deviation is stable, it gradually increases the complexity of the preset path and the target speed according to preset adjustment rules. For example, a preset adjustment rule could be set as follows: each time the user reaches a stable deviation state, the target movement speed increases by 5%, and the complexity of the preset path increases by one level, for example, changing from a straight path to one with a curve, or increasing the number of points on the path requiring precise control. The rehabilitation device's drive motor and control algorithm adjust their output accordingly to guide the user through the new speed and path. If the control unit determines that the stable deviation state has not been reached, meaning that the user's movement deviation remains large or fluctuates erratically, the rehabilitation device maintains the current preset path and speed settings. This means that the rehabilitation device will continue training at the current low speed and simplified path until the user's performance reaches a stable deviation state. This mechanism ensures that the difficulty of rehabilitation training matches the user's actual ability, avoiding the potential for poor training results or user discomfort caused by blindly increasing the difficulty. Through this technical solution, after the rehabilitation device switches to vibration-assisted and low-speed exercise mode, it can dynamically and fine-tune the rehabilitation program based on the user's real-time performance.

[0043] In another embodiment of the present application, it is further proposed that after switching the mode of the rehabilitation device to the vibration assistance and low-speed exercise mode, the sub-step S6000 further includes: S6600: Within a preset response determination time window, determine based on the actual motion trajectory whether the user exhibits a continuous deviation trajectory or periodic disturbance response that is inconsistent with the task target direction while in vibration-assisted and low-speed motion modes; S6700: If it is determined that there is a continuous deviation trajectory or periodic disturbance response, it is determined that there is non-goal-oriented compensatory movement or abnormal behavior affected by external interference. A prompt signal is sent to the rehabilitation personnel, and an amplitude below the interference threshold is applied to the user to verify the user's response stability and movement intention. Figure 1 The disturbance threshold is the preset user safety response upper limit torque amplitude; S6800: If the user generates an unpredictable response to the perturbation torque, the current mechanical response data is marked as a key evaluation interval and the corresponding abnormal trajectory characteristic parameters are recorded.

[0044] The preset response determination time window refers to the time period used by the system to analyze the user's motion trajectory and mechanical response data. It can be dynamically set based on the stage of rehabilitation training, the user's recovery status, or the characteristics of a specific task to ensure timely detection of abnormal behavior. A continuously deviating trajectory from the task target direction refers to a situation where the user's actual motion path continuously deviates from the predetermined rehabilitation path over a period of time. This deviation is not random fluctuation but rather exhibits a trend or directionality. This can manifest as the user attempting to complete the movement using non-standard postures or muscle groups. For example, during elbow flexion and extension training, the user may use shoulder swing to assist in completing the movement. A periodic perturbation response refers to regular, repetitive fluctuations or tremors in the user's motion trajectory or mechanical response data. This can be caused by external environmental factors such as equipment vibration and ambient noise, or the user's own physiological factors, such as physiological tremor or myoclonus. Non-goal-directed compensatory movements refer to the situation where, when completing a rehabilitation task, the user unconsciously or consciously adopts unintended, non-rehabilitation-targeted limb movement patterns to achieve the task requirements. This is intended to bypass the impaired function rather than directly improve it. Abnormal behavior due to external interference refers to abnormal performance during rehabilitation training caused by unexpected influences from the external environment or the device itself. This can include loose connections, sensor drift, or sudden noise in the training environment. The interference threshold refers to the upper limit of torque applied to the user to ensure safety and comfort. It can be pre-set based on the user's physiological tolerance, rehabilitation stage, and the safety specifications of the rehabilitation device. A perturbation torque is a small, short-term torque input used to detect the user's movement intention and response stability. It can be output in the form of a pulse, step, or low-amplitude sinusoidal wave. An unpredictable response refers to an irregular, violent, or delayed mechanical response after a perturbation torque stimulus that is inconsistent with normal physiological reactions or expected movement intentions. This can manifest as sudden muscle stiffness, severe tremors, or complete loss of limb control. Key assessment intervals are periods of time identified by the system where the user's movement performance is abnormal or their response is unstable. These periods are automatically marked and highlighted by the system for in-depth analysis and diagnosis by rehabilitation personnel. Abnormal trajectory characteristic parameters refer to specific values ​​or patterns used to quantify and describe the user's abnormal motion trajectory, which may include offset amplitude, offset direction, disturbance frequency, disturbance amplitude, motion smoothness index, etc. These parameters can be used for subsequent rehabilitation program adjustment and effect evaluation.

[0045] The solution of the present application further improves the rehabilitation training control in vibration-assisted and low-speed movement modes by introducing a mechanism for identifying and responding to abnormal movement behaviors of users.

[0046] In some preferred embodiments, the present application is specifically implemented as follows: When the rehabilitation device switches to vibration-assisted and low-speed motion mode, the device's control unit can continuously collect the user's motion angle data and, based on this data, construct the user's actual motion trajectory in real time. For example, the control unit can use a Kalman filter to smooth the raw sensor data to ensure trajectory accuracy. Within a preset response determination time window, for example, every 5 seconds, the control unit can analyze the most recent motion trajectory data. Analysis methods can include calculating the Fréchet distance or Hausdorff distance between the actual motion trajectory and the preset path to quantify the degree of trajectory deviation. Simultaneously, Fourier transform or wavelet analysis can be used to detect the presence of specific frequency components in the trajectory data to identify periodic perturbation responses. If the user's elbow joint flexion and extension trajectory continuously deviates from the preset straight path and exhibits a tendency to swing outward, this can be identified as a continuously deviated trajectory. Alternatively, if the user's limbs are detected to exhibit regular tremors with a frequency range of 4 to 12 Hz during movement, this can be identified as a periodic perturbation response. If the control unit determines the presence of a continuous deviation trajectory or periodic perturbation response, the system can immediately identify non-goal-directed compensatory movement or abnormal behavior due to external interference. At this point, the rehabilitation device can send a visual or auditory prompt to the connected rehabilitation personnel's terminal (such as a tablet or workstation) via the wireless communication module. For example, a warning box pops up on the rehabilitation personnel's monitoring interface or a sound prompt plays. Simultaneously, the rehabilitation device's actuator can apply a perturbation torque to the user. This perturbation torque can be a short pulse torque with a duration of 50 milliseconds and an amplitude below the user's safe response upper limit (for example, below 0.5 N·m) to verify the user's movement intention and limb stability. The system then continuously monitors the user's mechanical response data after the perturbation torque is applied. If the user exhibits an unpredictable response to the perturbation torque, such as sudden, violent, and irregular shaking of the limb or a momentary stiffness and cessation of movement after the torque is applied, this can be considered an unpredictable response. In this case, the control unit automatically marks the mechanical response data within the current time period as a critical assessment interval and stores it in the rehabilitation device's local memory or uploads it to a cloud database. The system also records the corresponding abnormal trajectory characteristic parameters, such as the movement angle, torque amplitude, duration, maximum trajectory deviation distance, and disturbance frequency at the time of the abnormality. This data can serve as an important basis for rehabilitation personnel to conduct detailed diagnosis and adjust the rehabilitation plan.

[0047] In another embodiment of the present application, it is further proposed that the sub-step of S6700 is to apply a force with an amplitude lower than the interference threshold to the user to verify the user's response stability and movement intention. Figure 1After the consistent perturbation torque, it also includes: S6720: Continuously monitor the user's mechanical response data and extract the user's response start time, maximum torque response time, and maximum perturbation torque time point; S6730: Calculate the time difference between the maximum torque response moment and the maximum value time point, and determine whether the time difference exceeds a preset response hysteresis threshold; S6740: If the time difference is determined to exceed the response hysteresis threshold, a hysteresis response event is recorded; if the cumulative number of hysteresis response events exceeds the preset number threshold during the preset number of perturbation torque application operations, the current user is determined to be a hysteresis response mode user; S6750: The user is determined to be a delayed response mode user, and the start time of the response determination time window is corrected by delaying the preset delay compensation time length.

[0048] Among them, continuously monitoring the user's mechanical response data means that after the perturbation torque is applied, the dynamic feedback information of the user's limbs to the torque is continuously collected. Specifically, it can be obtained in real time through devices such as force sensors and angle sensors. The response start time refers to the initial moment when the user's limbs produce a recognizable response to the perturbation torque. Specifically, it can be determined by analyzing the point where the torque or position change rate in the mechanical response data first exceeds the background noise threshold. Its purpose is to mark the exact time point when the user begins to respond to the stimulus. The maximum torque response moment refers to the moment when the torque amplitude in the user's mechanical response data reaches its peak value. Specifically, it can be identified by performing peak detection on the collected torque data. Its purpose is to determine the time when the user's response intensity reaches the highest point. The maximum value time point of the perturbation torque refers to the moment when the perturbation torque applied by the rehabilitation device reaches its preset peak amplitude on the time axis. Specifically, it can be determined by the torque output control signal of the rehabilitation device or the built-in sensor record.

[0049] The response hysteresis threshold refers to the preset time limit for determining whether a user's response is slow. It can be pre-set based on clinical experience, user group characteristics, or rehabilitation goals. Its purpose is to quantitatively determine whether the user's response exceeds the acceptable delay range. A hysteresis response event refers to a situation where the difference in the user's response time to a single perturbation torque exceeds the preset response hysteresis threshold. Its purpose is to mark each case of untimely response. The preset number of perturbation torque application operations refers to the total number of times the rehabilitation device applies perturbation torque to the user within a period of time or a specific training cycle. The preset number threshold refers to the cumulative upper limit of hysteresis response events used to determine whether a user belongs to the hysteresis response mode. It can be set based on clinical guidelines or the user's rehabilitation stage. A user in the hysteresis response mode refers to a user whose cumulative number of hysteresis response events exceeds the preset number threshold in a preset number of perturbation torque application operations. The starting time of the response determination time window refers to the starting time point when the rehabilitation device system begins to evaluate whether the user's response to the perturbation torque meets expectations. The preset hysteresis compensation time refers to the preset amount of time used to delay the start of the response determination time window after the user is determined to be a user in the hysteresis response mode. It can be set according to the user's hysteresis level or clinical experience.

[0050] The solution of this application verifies the user's response stability and movement intention by applying an amplitude lower than the interference threshold to the user. Figure 1 After obtaining consistent perturbation torque, the evaluation of the user's mechanical response is further refined.

[0051] In some preferred embodiments, the present application is specifically implemented as follows: When the rehabilitation device applies a perturbation torque with an amplitude below the perturbation threshold to the user, the device's built-in high-precision torque and angle sensors can continuously collect mechanical response data from the user's limbs, including joint torque and angle changes, at a sampling frequency of, for example, 1000 Hz. An embedded processor, such as a microcontroller based on the ARM Cortex-M4 architecture, can process this data in real time. The processor can run a signal processing algorithm to extract the response start time, for example, by detecting the moment when the slope change of the torque signal first exceeds a preset noise level. Simultaneously, the processor can identify the maximum peak in the torque response curve and determine the time of maximum torque response. The time point of maximum perturbation torque can be obtained from the device's torque output command or an internal timer. The processor then calculates the time difference between the maximum torque response time and the time point of maximum perturbation torque. For example, if the preset response hysteresis threshold is 80 milliseconds, the system can record a hysteresis response event when the calculated time difference exceeds 80 milliseconds. The system can also continuously count the cumulative number of hysteresis response events, for example, over 10 consecutive perturbation torque application operations. If the cumulative number of times exceeds a preset threshold of, for example, 3 times, the user can be determined to be a user in a hysteresis response mode. Once the user is determined to be a user in a hysteresis response mode, the system can immediately adjust the subsequent response determination logic. For example, if the starting time of the original response determination time window is 50 milliseconds after the perturbation torque is applied, the system can delay it by a preset hysteresis compensation time length, for example, by 50 milliseconds, so that the new response determination time window starts to be evaluated from 100 milliseconds after the perturbation torque is applied. In this way, even if there is a certain delay in the user's response, the system can accurately capture its true response within a more ample time window, avoiding misjudgment caused by hysteresis.

[0052] In another embodiment of the present application, after the user is determined to be a user in the delayed response mode, the method further includes: S6751: Obtain the user's mechanical response peak amplitude, response duration, and vibration input delay during the application of the perturbation torque, and calculate the user's vibration tolerance parameter. The vibration tolerance parameter is used to characterize the user's sensitivity to the perturbation torque. S6752: Dynamically adjust the output characteristics of subsequent perturbation torque based on vibration tolerance parameters. The output characteristics include amplitude upper limit, rise time, continuous application duration, and / or start delay duration. S6753: After applying the adjusted perturbation torque, collect the user's mechanical response data, update the vibration tolerance parameters based on the mechanical response data, and use the updated vibration tolerance parameters for adaptive adjustment of the output characteristics of the subsequent perturbation torque.

[0053] Among them, the peak amplitude of mechanical response refers to the instantaneous maximum value of the maximum reaction torque or displacement generated by the user's limbs or joints when the user is stimulated by the perturbation torque. It can be analyzed using data collected by force sensors or angle sensors. Its purpose is to quantify the user's immediate reaction intensity to the stimulus; the response duration refers to the length of time it takes for the user's mechanical response to basically return to a stable state after being stimulated by the perturbation torque. It can be achieved using the duration analysis method in signal processing. Its purpose is to evaluate the user's tolerance and recovery ability to the stimulus; the vibration input delay refers to the time from the start of the perturbation torque. The time interval between the mechanical responses applied to the user that can be recognized can be achieved by timestamp comparison or signal threshold detection methods; the vibration tolerance parameter refers to a comprehensive indicator used to quantify the user's sensitivity and adaptability to the perturbation torque, which can be achieved by weighted combination of the peak amplitude of the mechanical response, the duration of the response and the delay time of the vibration input or by calculation through a machine learning model, with the aim of providing a basis for personalized adjustment of the perturbation torque; the output characteristics of the perturbation torque refer to the physical properties that can be controlled and adjusted during the application of the perturbation torque, which can be achieved by the control of the torque output module of the rehabilitation instrument. The purpose is to finely control the application method of the perturbation torque; the upper limit of the amplitude refers to the maximum torque value allowed to be reached by the perturbation torque during the application process, which can be achieved by software setting or hardware limiting, with the purpose of ensuring user safety and avoiding excessive stimulation; the rising edge time refers to the time required for the perturbation torque to rise from zero value or starting value to its peak value or stable value, which can be achieved by slope control or piecewise linear approximation, with the purpose of controlling the degree of sudden change of stimulation and reducing user discomfort; the continuous application time refers to the length of time the perturbation torque remains in effect after reaching the preset amplitude, which can be achieved by a timer The purpose of the system is to control the total amount of stimulation by means of a control or event triggering mechanism; the start-up delay time refers to the time interval between the system deciding to apply the perturbation torque and the actual start of applying the perturbation torque, which can be achieved by means of a delay device or a waiting instruction, with the purpose of providing the user with psychological preparation time or synchronizing with other actions of the rehabilitation device; adaptive adjustment refers to the process in which the system can automatically and dynamically modify the output characteristics of the perturbation torque according to the user's real-time response data to optimize the stimulation effect, which can be achieved by means of a feedback control algorithm or a reinforcement learning algorithm, with the purpose of achieving personalized and optimized rehabilitation training.

[0054] The solution of the present application overcomes the limitations of merely correcting the response determination time window by further introducing a dynamic optimization mechanism for the perturbation torque application strategy after the user is determined to be a hysteresis response mode user.

[0055] In some preferred embodiments, when the user is determined to be a user in the hysteresis response mode, the control system of the rehabilitation apparatus may specifically implement the following steps: First, during the application of the perturbation torque, the system collects real-time mechanical response data from the user using force sensors and angle sensors integrated into the joints of the rehabilitation device. For example, the force sensor can record the reaction torque curve of the user's limb to the perturbation torque, while the angle sensor can record the displacement curve of the limb under the perturbation torque. From these curves, the system can extract the peak amplitude of the mechanical response (for example, by identifying the maximum instantaneous value on the torque curve); the response duration (for example, by calculating the time from when the torque response first exceeds the baseline threshold to when it falls back below the baseline threshold); and the vibration input delay (for example, by comparing the timestamp of the start of the perturbation torque application with the timestamp when the user's mechanical response first reaches a preset activation threshold). The system then uses these extracted feature values ​​to calculate the user's vibration tolerance parameter using a pre-defined algorithm model. For example, this parameter can be a weighted average, where the peak amplitude of the mechanical response, the response duration, and the vibration input delay are each multiplied by different weighting coefficients to reflect their relative importance in assessing sensitivity. Alternatively, a model based on fuzzy logic or a neural network can be used to input these feature values ​​and output a normalized value representing the user's sensitivity level. Then, based on the calculated vibration tolerance parameters, the control system can dynamically adjust the output characteristics of subsequent perturbation torques. For example, the system can preset a parameter adjustment table to find the corresponding upper amplitude limit, rise time, duration of application, and / or start-up delay duration based on the range of the vibration tolerance parameters. If the vibration tolerance parameters indicate high user sensitivity, the system can automatically select a smaller upper amplitude limit, a longer rise time, a shorter duration of application, or increase the start-up delay to ensure mild stimulation. Conversely, if the parameters indicate good user tolerance, the upper amplitude limit or rise time can be appropriately increased to ensure effective stimulation. Finally, after applying the adjusted perturbation torque, the system will again collect the user's mechanical response data. This new data is input into the same algorithm model to update the user's vibration tolerance parameters. For example, an exponentially weighted moving average method can be used to integrate the new vibration tolerance parameters with historical parameters so that the parameters smoothly reflect changes in the user's state. The updated vibration tolerance parameters will be immediately used to adjust the output characteristics of the next perturbation torque, thus forming a continuous, adaptive feedback loop to ensure that the application of the perturbation torque always matches the user's real-time physiological state.

[0056] In another embodiment of the present application, it is further proposed that the sub-step of S7000: after switching the mode of the rehabilitation device to the force compensation mode, includes: S7200: Continuously collects the user's active force response data under the action of the basic pushing torque within the preset monitoring time window; S7300: Calculate the time-weighted muscle fatigue integral based on the difference between the active force response data and the actual output torque of the rehabilitation device; S7400: Determine whether the muscle fatigue score is lower than the recovery determination threshold; S7500: If the muscle fatigue score is lower than the recovery judgment threshold, the user is determined to have the potential to recover through independent force, and the rehabilitation device mode is switched from force compensation mode to preset path recovery assessment mode. The preset path recovery assessment mode uses a step-by-step reduction of the pushing torque to guide the user to complete the target movement again independently.

[0057] The preset monitoring time window refers to a pre-set period of time during which the system continuously collects the user's physiological or motion data. The purpose is to obtain sufficient data for stable and reliable evaluation and avoid misjudgments caused by instantaneous data fluctuations. Active force response data refers to the torque or mechanical signal generated by the user's own muscles to complete the target movement when the rehabilitation device provides the basic pushing torque. Specifically, it can be collected by devices such as force sensors, electromyographic sensors, or kinematic sensors. Its purpose is to quantify the user's autonomous movement ability. The time-weighted muscle fatigue score refers to a cumulative indicator. It is obtained by weighted summing the difference between the user's active force response data at different time points and the actual output torque of the rehabilitation device, where the most recent data is given a higher weight. Specifically, it can be achieved through algorithms such as exponential decay weighting or linear decay weighting. The recovery judgment threshold refers to a pre-set numerical standard used to determine whether the user's muscle fatigue score has reached the conditions for switching to rehabilitation mode. It can be set based on clinical experience, individual user differences, or rehabilitation goals. The preset path recovery assessment mode refers to a specific operating mode of the rehabilitation device. In this mode, the rehabilitation device will gradually reduce the pushing torque on the user according to the pre-set path. Specifically, this can be achieved by controlling the motor output torque or adjusting the mechanical damping. Its purpose is to guide the user to gradually increase their autonomous force, complete the target movement independently again, and evaluate their actual recovery level. Gradual reduction of the pushing torque means that in the preset path recovery assessment mode, when the rehabilitation device assists the user to complete the movement, the pushing torque it provides is not completely withdrawn at one time, but is gradually reduced in stages according to the preset step size or gradient. Specifically, the torque output can be controlled by a software algorithm. For example, each time a cycle is completed or a time point is reached, the pushing torque is reduced by a fixed value or a percentage. Its purpose is to gradually challenge the user's autonomous force ability and avoid the user's inability to complete the movement or discomfort due to the sudden disappearance of the auxiliary force.

[0058] The solution of the present application further introduces a dynamic evaluation and mode adjustment mechanism after the rehabilitation device switches to the force compensation mode, thereby solving the problem that simply increasing the constant value of the basic pushing torque cannot determine the user's autonomous force recovery potential.

[0059] In some preferred embodiments, when the rehabilitation device is switched to force compensation mode, the device's control unit can initiate a continuous monitoring program. For example, the control unit can be an embedded processor connected to a force sensor and a motion sensor for collecting user-initiated force response data. Within a preset monitoring time window, for example, lasting 5 minutes, the control unit can continuously acquire torque output data from the user under the action of the basic propulsion torque at a sampling frequency of 100 Hz. Simultaneously, the rehabilitation device's motor drive unit provides real-time feedback on its actual torque output. The control unit can calculate the user's voluntary force contribution based on the difference between the collected voluntary force response data (e.g., the torque applied by the user to the rehabilitation device) and the device's actual output torque. To assess muscle fatigue, the control unit can calculate a muscle fatigue score using a time-weighted approach. For example, an exponentially weighted average algorithm can be used to assign a higher weight to the most recent voluntary force contribution data, so that the score more sensitively reflects the user's current fatigue state. The control unit can then determine whether the calculated muscle fatigue score is below a preset recovery threshold. This threshold can be personalized based on the user's stage of rehabilitation, disease type, or rehabilitation goals. For example, it can be set so that when the muscle fatigue score falls below a certain value, the user is considered to have the potential for self-force recovery. If the judgment result shows that the muscle fatigue score is below the recovery judgment threshold, the control unit can immediately determine that the user has the potential for self-force recovery. At this time, the control unit can send a command to the rehabilitation device's mode switching module to switch the device's mode from force compensation mode to preset path recovery assessment mode. In preset path recovery assessment mode, the rehabilitation device's motor drive unit can be programmed to guide the user to complete the target movement along a preset path by gradually reducing the propulsion torque. For example, after each complete exercise cycle, the rehabilitation device's propulsion torque can be automatically reduced by 5% or a fixed value until the propulsion torque drops to a preset minimum assistance level or the user is able to complete the movement completely independently. This approach can gradually guide the user to rely more on their own strength, thereby promoting muscle strength recovery and improving their motor control ability.

[0060] Reference Figure 2 In another embodiment of the present application, a rehabilitation apparatus motion mode control system is further proposed, comprising: Motion and torque acquisition module 1, used to obtain the user's motion angle data and resistance torque data; Athletic performance evaluation module 2 is used to determine whether the deviation between the actual motion trajectory and the preset path exceeds a preset threshold based on the motion angle data, and to determine whether the torque actively exerted by the user shows a continuous attenuation trend based on the counteracting torque data, thereby determining whether the user's athletic performance has declined; Diagnostic stimulation application module 3 is used to change the auxiliary force output form of the rehabilitation device and apply diagnostic mechanical stimulation if it is determined that the user's athletic performance has declined. The diagnostic mechanical stimulation includes superimposing a vibration component of a preset frequency and preset amplitude on the original basic driving torque of the rehabilitation device; Response data acquisition module 4, used to collect the user's mechanical response data during the application of diagnostic mechanical stimulation; The resistance source determination module 5 is used to determine, based on the mechanical response data, whether the source of the user's resistance is a reduction in active force or an involuntary muscle stretch reflex; a vibration assistance mode switching module 6, configured to switch the mode of the rehabilitation device to a vibration assistance and low-speed motion mode if the source of the antagonistic force is determined to be an involuntary muscle stretch reflex. The vibration assistance and low-speed motion modes include continuously superimposing a vibration component in the auxiliary force output, reducing the motion speed of the task target, and simplifying the complexity of the preset path; The force compensation mode switching module 7 is used to switch the mode of the rehabilitation device to the force compensation mode if the source of the antagonistic force is determined to be a reduction in active force. The force compensation mode includes increasing the constant value of the basic driving torque.

[0061] The solution of the present application implements the various functional steps in the rehabilitation instrument movement mode control method into a series of mutually cooperating modules, thereby realizing the perception, evaluation, diagnosis and intervention of the user's rehabilitation training process.

[0062] In one embodiment, the rehabilitation apparatus movement mode control system can be specifically implemented as follows: The motion and torque acquisition module 1 can consist of a rotary encoder and a torque sensor installed at the joints of the rehabilitation device. For example, the rotary encoder is used to measure the movement angle of the user's limb, and the torque sensor is used to monitor the output torque of the rehabilitation device and the user's opposing torque in real time. These sensors are connected to the central processing unit via wired or wireless means.

[0063] The sports performance evaluation module 2 can be integrated into an embedded controller of the rehabilitation device, which runs a preset algorithm and continuously receives sports angle data and resistance torque data.

[0064] The diagnostic stimulation applying module 3 may be composed of a driving motor of the rehabilitation apparatus and its supporting driving circuit.

[0065] The response data acquisition module 4 may utilize the same torque sensor as the motion and torque acquisition module to continuously acquire the user's torque response data at a higher sampling rate (eg, 1000 Hz) during the application of diagnostic mechanical stimulation.

[0066] The antagonistic force source determination module 5 can be a software module running on the central processing unit, which receives the torque response data provided by the response data acquisition module 4. This module can use Fourier transform-based analysis of the phase and amplitude characteristics of the vibration response, combined with a preset physiological model or machine learning model, to determine whether the antagonistic force is caused by a reduction in active force (for example, the response amplitude is linearly related to the stimulus amplitude and has no significant phase lag) or an involuntary muscle stretch reflex (for example, the response amplitude increases at a specific frequency and is accompanied by a phase advance).

[0067] Both the vibration assistance mode switching module 6 and the force compensation mode switching module 7 can serve as control logic units in the central processing unit.

[0068] The above description is merely an embodiment of the present application and is not intended to limit the scope of protection of the present application. For those skilled in the art, various modifications and variations of the present application are possible. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A method for controlling a rehabilitation apparatus movement mode, characterized in that: include: Obtain the user's motion angle data and resistance torque data; Determining whether the deviation between the actual motion trajectory and the preset path exceeds a preset threshold based on the motion angle data, and determining whether the torque actively exerted by the user shows a continuous attenuation trend based on the counteracting torque data, thereby judging whether the user's athletic performance has declined; If it is determined that the user's athletic performance has declined, the auxiliary force output form of the rehabilitation device is changed and diagnostic mechanical stimulation is applied. The diagnostic mechanical stimulation includes superimposing a vibration component of a preset frequency and preset amplitude on the original basic driving torque of the rehabilitation device; collecting mechanical response data of the user during application of the diagnostic mechanical stimulation; determining, based on the mechanical response data, whether a source of the user's resistance force is a reduction in active force or an involuntary muscle stretch reflex; If the source of the antagonistic force is determined to be the involuntary muscle stretch reflex, the mode of the rehabilitation device is switched to a vibration-assisted and low-speed motion mode, wherein the vibration-assisted and low-speed motion modes include continuously superimposing a vibration component in the auxiliary force output, reducing the motion speed of the task target, and simplifying the complexity of the preset path; If the source of the antagonistic force is determined to be a reduction in the active force, the mode of the rehabilitation apparatus is switched to a force compensation mode, wherein the force compensation mode includes increasing a constant value of the basic propulsion torque.

2. The rehabilitation apparatus movement mode control method according to claim 1, characterized in that: The collecting of the user's mechanical response data during the application of the diagnostic mechanical stimulation comprises: Obtaining the natural frequency and natural phase of the user's intrinsic limb tremor; applying an anti-phase vibration torque having the same frequency as the natural frequency and opposite phase to the natural phase to suppress the natural limb tremor; The user's torque data is collected as the mechanical response data.

3. The rehabilitation apparatus movement mode control method according to claim 2, characterized in that: Also includes: Continuously updating the natural frequency and the natural phase in a rolling time window manner; After each update of the natural frequency and the natural phase, the frequency and phase of the anti-phase vibration torque are synchronously adjusted so that the frequency is continuously consistent with the natural frequency and the phase is opposite to the natural phase; determining whether the intrinsic limb tremor is in a stable state according to a change trend of the intrinsic frequency and the intrinsic phase in the rolling time window; If it is determined to be a stable state, the current parameters of the anti-phase vibration torque are kept unchanged; If it is determined to be an unstable state, the natural frequency and the natural phase are continuously updated by means of the rolling time window, and the frequency and phase of the anti-phase vibration torque are synchronously adjusted based on the updated natural frequency and the natural phase.

4. The rehabilitation apparatus movement mode control method according to claim 3, characterized in that: Also includes: determining whether there is a residual tremor component in the torque response of the user after applying the anti-phase vibration torque; If the residual tremor component exists, constructing a compensation signal based on frequency and phase information of the residual tremor component; Based on the compensation signal, the output amplitude and phase of the anti-phase vibration torque are adaptively adjusted.

5. The rehabilitation apparatus movement mode control method according to claim 1, characterized in that: After switching the mode of the rehabilitation device to the vibration assistance and low-speed exercise mode, the method further includes: Monitoring the actual movement deviation value of the user along the preset path; Determining whether the actual motion deviation value is continuously in a deviation stable state below a first deviation threshold; If it is determined that the deviation is in a stable state, the complexity of the preset path and the movement speed of the task target are gradually increased according to the preset adjustment rules; If it is determined that the deviation has not reached a stable state, the current settings of the preset path and the movement speed are maintained.

6. The method for controlling the exercise mode of a rehabilitation apparatus according to claim 5, wherein: After switching the mode of the rehabilitation device to the vibration assistance and low-speed exercise mode, the method further includes: Determining, within a preset response determination time window, based on the actual motion trajectory, whether the user exhibits a continuous deviation trajectory or a periodic disturbance response that is inconsistent with the task target direction while in the vibration-assisted and low-speed motion modes; If it is determined that the continuous offset trajectory or periodic disturbance response exists, it is determined that there is non-goal-oriented compensatory movement or abnormal behavior affected by external interference, and a prompt signal is sent to the rehabilitation personnel, and a perturbation torque is applied to the user with an amplitude lower than the interference threshold to verify the consistency of the user's response stability and movement intention. The interference threshold is a preset upper limit torque amplitude of the user's safe response; If the user produces an unpredictable response to the perturbation torque, the current mechanical response data is marked as a key evaluation interval, and the corresponding abnormal trajectory characteristic parameters are recorded.

7. The method for controlling the exercise mode of a rehabilitation apparatus according to claim 6, wherein: After applying a perturbation torque to the user with an amplitude lower than the interference threshold for verifying the consistency between the user's response stability and the movement intention, the method further includes: Continuously monitoring the user's mechanical response data, extracting the user's response start time, maximum torque response time, and the maximum value time point of the perturbation torque; Calculating a time difference between the maximum torque response moment and the maximum value time point, and determining whether the time difference exceeds a preset response hysteresis threshold; If it is determined that the time difference exceeds the response hysteresis threshold, it is recorded as a hysteresis response event; If the cumulative number of hysteresis response events exceeds a preset threshold value during a preset number of perturbation torque application operations, the current user is determined to be a hysteresis response mode user; The user is determined to be a user in the delayed response mode, and the starting time of the response determination time window is delayed by a preset delay compensation time length for correction.

8. The method for controlling the exercise mode of a rehabilitation apparatus according to claim 7, wherein: After the user is determined to be a user in a delayed response mode, the method further includes: Obtaining the user's mechanical response peak amplitude, response duration, and vibration input delay during the application of the perturbation torque, and calculating the user's vibration tolerance parameter, where the vibration tolerance parameter is used to characterize the user's sensitivity to the perturbation torque; Dynamically adjust the subsequent output characteristics of the perturbation torque according to the vibration tolerance parameter, the output characteristics including the amplitude upper limit, the rising edge time, the continuous application time and / or the start delay time; After applying the adjusted perturbation torque, the user's mechanical response data is collected, and the vibration tolerance parameter is updated based on the mechanical response data. The updated vibration tolerance parameter is used for subsequent adaptive adjustment of the output characteristics of the perturbation torque.

9. The method for controlling the exercise mode of a rehabilitation apparatus according to claim 1, wherein: After the mode of the rehabilitation device is switched to the force compensation mode, the method further includes: within a preset monitoring time window, continuously collecting active force response data of the user under the action of the basic propulsion torque; Calculating a time-weighted muscle fatigue integral according to a difference between the active force response data and an actual output torque of the rehabilitation apparatus; determining whether the muscle fatigue score is lower than a recovery determination threshold; If the muscle fatigue score is lower than the recovery judgment threshold, it is determined that the user has the potential to recover through autonomous force, and the mode of the rehabilitation instrument is switched from the force compensation mode to the preset path recovery assessment mode. The preset path recovery assessment mode uses a step-by-step reduction of the pushing torque to guide the user to complete the target action again autonomously.

10. A rehabilitation apparatus motion mode control system, characterized in that: include: Motion and torque acquisition module, used to obtain the user's motion angle data and resistance torque data; an athletic performance evaluation module, configured to determine, based on the motion angle data, whether the deviation between the actual motion trajectory and the preset path exceeds a preset threshold, and to determine, based on the counteracting torque data, whether the torque actively exerted by the user exhibits a continuous attenuation trend, thereby determining whether the user's athletic performance has declined; a diagnostic stimulation application module, configured to, if it is determined that the user's athletic performance has declined, change the auxiliary force output form of the rehabilitation device and apply diagnostic mechanical stimulation, wherein the diagnostic mechanical stimulation comprises superimposing a vibration component of a preset frequency and preset amplitude on the original basic driving torque of the rehabilitation device; a response data acquisition module, configured to acquire the user's mechanical response data during application of the diagnostic mechanical stimulation; an antagonistic force source determination module, configured to determine, based on the mechanical response data, whether the source of the user's antagonistic force is a reduction in active force or an involuntary muscle stretch reflex; a vibration assistance mode switching module, configured to switch the mode of the rehabilitation device to a vibration assistance and low-speed motion mode if the source of the antagonistic force is determined to be the involuntary muscle stretch reflex, wherein the vibration assistance and low-speed motion modes include continuously superimposing a vibration component in the auxiliary force output, reducing the motion speed of the task target, and simplifying the complexity of the preset path; The force compensation mode switching module is used to switch the mode of the rehabilitation device to the force compensation mode if the source of the antagonistic force is determined to be the reduction of the active force. The force compensation mode includes increasing the constant value of the basic driving torque.

Images