A control method, device and medium of an anti-gravity rehabilitation treadmill

Abstract

The application discloses a control method, device and medium of an anti-gravity rehabilitation treadmill, relates to the field of lower limb rehabilitation, and adopts a least square method to fit a trend slope after collecting time sequence data of supporting gravity in a preset sliding time window and eliminating transient spikes; when the trend slopes in a continuously set number of windows are all higher than a threshold value, the application predicts future supporting gravity according to the trend slopes, obtains a predicted weight loss deviation, and calculates a feedforward air pressure pre-adjustment amount, so that pneumatic adjustment is prior to deviation expansion, and response delay is compensated. Meanwhile, a weight loss deviation and a feedback air pressure correction amount are calculated according to a difference between a target weight loss gravity and an actual weight loss gravity, the feedforward air pressure pre-adjustment amount and the feedback air pressure correction amount are superposed to obtain a total air pressure adjustment amount to control air bag inflation and exhaust, and the actual weight loss gravity converges to the target weight loss gravity; the feedforward air pressure pre-adjustment amount compensates for a trend change in advance, the feedback air pressure correction amount eliminates residual errors and sudden disturbances, the two are cooperated to suppress weight center swing and weight loss gravity fluctuation caused by feedback lag, and weight loss stability is improved.

Description

Technical Field

[0001] This application relates to the field of lower limb rehabilitation, and in particular to a control method, device and medium for an anti-gravity rehabilitation treadmill. Background Technology

[0002] With the continuous increase in the number of patients suffering from neurological and musculoskeletal diseases such as stroke, spinal cord injury, and lower limb fracture surgery, the demand for rehabilitation treatment of lower limb motor dysfunction is increasing. Antigravity rehabilitation treadmills, as an important lower limb rehabilitation training device, apply an upward lifting force to the body through an airbag weight-reduction mechanism, partially offsetting the patient's weight. This allows patients to perform walking training in a weight-reduced state, effectively reducing lower limb joint load, improving gait symmetry, and promoting neuromuscular function reconstruction. It has been widely used in the field of clinical rehabilitation.

[0003] The existing closed-loop control strategy for weight reduction in equipment is based solely on the feedback adjustment of the instantaneous deviation between the actual weight reduction force and the target weight reduction force. This approach is essentially a post-event compensation mechanism: the controller only starts to act after the deviation has occurred and been measured, making it impossible to predict and compensate in advance. The weight reduction fluctuations caused by the back-and-forth swing of the center of gravity when a person walks are periodic disturbances, and the response of the pneumatic system itself has an inherent delay. Instantaneous deviation feedback control not only fails to effectively suppress fluctuations, but may also introduce additional weight reduction fluctuations due to lag, resulting in poor weight reduction stability.

[0004] Therefore, how to improve the stability of weight reduction is a problem that needs to be solved by those skilled in the art. Summary of the Invention

[0005] The purpose of this application is to provide a control method, device, and medium for an anti-gravity rehabilitation treadmill, which solves the problem that the existing weight reduction closed-loop control strategy only relies on feedback adjustment based on the instantaneous deviation between the actual weight reduction force and the target weight reduction force, resulting in poor weight reduction stability.

[0006] To solve the above-mentioned technical problems, this application provides a control method for an anti-gravity rehabilitation treadmill, comprising: Acquire the calibrated body weight, target weight reduction force, and supporting gravity detected by the tread pressure sensor; The actual weight reduction force is calculated based on the human body's calibrated weight and the supporting gravity. Calculate the weight reduction deviation based on the target weight reduction force and the actual weight reduction force; The time-series data of the supporting gravity is continuously collected in a sliding time window of a preset duration. Instantaneous spikes in the time-series data that exceed a preset impact threshold are removed. The least squares method is used to fit the trend slope of the supporting gravity changing with time on the removed time-series data. When the trend slope exceeds a set threshold within a set number of consecutive sliding time windows, the predicted support gravity after a preset time period is predicted based on the trend slope, and the predicted weight reduction deviation is obtained. The feedback pressure correction amount is calculated based on the weight reduction deviation, and the feedforward pressure pre-adjustment amount is calculated based on the predicted weight reduction deviation. The total pressure regulation is obtained by superimposing the feedback pressure correction amount with the feedforward pressure pre-adjustment amount. The airbag is inflated or deflated according to the total air pressure regulation amount, so that the actual weight reduction force converges to the target weight reduction force.

[0007] In one optional embodiment, calculating the actual weight reduction force based on the calibrated body weight and the supporting gravity includes: The product of the human body's calibrated weight and gravitational acceleration is taken as the human body's calibrated gravity; The actual weight reduction force is obtained by subtracting the supporting weight from the human body's calibrated weight.

[0008] In one optional embodiment, calculating the feedback pressure correction based on the weight reduction deviation includes: The weight loss deviation is subjected to dead zone determination. The weight loss deviations whose absolute value is less than or equal to the preset dead zone threshold are set to zero, and the weight loss deviations whose absolute value is greater than the preset dead zone threshold are retained. The weight loss deviation after dead zone determination is filtered by mean filtering to obtain the filtered weight loss deviation. The rate of change of weight reduction deviation is calculated based on the filtered weight reduction deviation between the current cycle and the previous cycle. Using the filtered weight reduction deviation and the rate of change of the weight reduction deviation corresponding to the current cycle as inputs, a fuzzy sliding mode controller is used to calculate the feedback air pressure correction.

[0009] In one optional embodiment, calculating the feedforward pressure pre-adjustment based on the predicted weight reduction deviation includes: The product of the predicted weight reduction deviation and the feedforward control coefficient is calculated to obtain the feedforward pressure pre-adjustment amount; The feedforward control coefficient is calibrated based on the relationship between air pressure and gravity reduction.

[0010] In one optional embodiment, controlling the inflation or deflation of the airbag according to the total air pressure regulation includes: Obtain the initial weight loss distribution coefficients for the waist, left leg, and right leg; The initial weight loss distribution coefficients of the waist, left leg, and right leg are updated based on the pressure distribution data collected by the plantar pressure sensor array to obtain the current weight loss distribution coefficients of the waist, left leg, and right leg. Based on the current weight loss distribution coefficient of the waist, the current weight loss distribution coefficient of the left leg, and the current weight loss distribution coefficient of the right leg, the total air pressure regulation is divided into waist air pressure regulation, left leg air pressure regulation, and right leg air pressure regulation. The inflation or deflation of the lumbar airbag is controlled according to the lumbar air pressure adjustment amount, the inflation or deflation of the left leg airbag is controlled according to the left leg air pressure adjustment amount, and the inflation or deflation of the right leg airbag is controlled according to the right leg air pressure adjustment amount.

[0011] In an optional embodiment, it further includes: Obtain pre-set target training parameters; wherein, the target training parameters include target velocity, target damping, target weight reduction force, and target buffer stiffness; Acquire plantar pressure data collected by the plantar pressure sensor array, and calculate the plantar pressure change rate based on the plantar pressure data; When the rate of change of plantar pressure is greater than a preset rate of change threshold, it is determined to be the impact phase of foot landing; During the landing impact phase, the running speed of the corresponding track is reduced to less than the target speed, the damping actuator is controlled to reduce the damping of the corresponding track to less than the target damping, the buffer actuator is controlled to reduce the buffer stiffness of the corresponding track to less than the target buffer stiffness, and the actual weight reduction force is adjusted to be greater than the target weight reduction force. When the rate of change of plantar pressure is less than or equal to a preset rate of change threshold, and the plantar pressure is greater than or equal to a preset pressure threshold, it is determined to be the support phase; During the support phase, the actual contact area of ​​the sole is determined based on the plantar pressure data. When the actual contact area of ​​the sole is less than a preset area threshold, the actual weight reduction force is adjusted to be greater than the target weight reduction force, and / or the running speed of the corresponding track is reduced to be less than the target speed. During the support phase, when the actual contact area of ​​the foot is greater than or equal to the preset area threshold, the training parameters are restored to the preset target training parameters.

[0012] In one optional embodiment, restoring each training parameter to a preset target training parameter includes: Calculate the weight reduction force deviation between the current actual weight reduction force and the target weight reduction force, the speed deviation between the current actual speed and the target speed, the damping deviation between the current actual damping and the target damping, and the buffer stiffness deviation between the current actual buffer stiffness and the target buffer stiffness, respectively. PID closed-loop calculations are performed based on the gravity reduction deviation, the velocity deviation, the damping deviation, and the buffer stiffness deviation to obtain the gravity reduction recovery adjustment amount, the velocity recovery adjustment amount, the damping recovery adjustment amount, and the buffer stiffness recovery adjustment amount. The gravity recovery adjustment amount, the velocity recovery adjustment amount, the damping recovery adjustment amount, and the buffer stiffness recovery adjustment amount are respectively limited to ensure that the absolute value of each recovery adjustment amount does not exceed its corresponding preset single-cycle limit value. The airbag is inflated or deflated according to the weight reduction force recovery adjustment amount after the limit, the drive motor of the corresponding runway is controlled according to the speed recovery adjustment amount after the limit, the damping actuator of the corresponding runway is controlled according to the damping recovery adjustment amount after the limit, and the buffer actuator at the bottom of the runway is controlled according to the buffer stiffness recovery adjustment amount after the limit, until the actual weight reduction force is restored to the target weight reduction force, the actual speed is restored to the target speed, the actual damping is restored to the target damping, and the buffer stiffness is restored to the target buffer stiffness.

[0013] This application also provides a control device for an anti-gravity rehabilitation treadmill, comprising: The first acquisition module is used to acquire the human body's calibrated weight, target weight reduction force, and support gravity detected by the tread pressure sensor; The first calculation module is used to calculate the actual weight reduction force based on the human body's calibrated weight and the supporting gravity. The second calculation module is used to calculate the weight reduction deviation based on the target weight reduction force and the actual weight reduction force. The fitting module is used to continuously collect time-series data of the supporting gravity in a sliding time window of a preset duration, remove instantaneous spikes in the time-series data that exceed a preset impact threshold, and use the least squares method to fit the trend slope of the supporting gravity changing with time on the removed time-series data. The prediction module is used to predict the predicted support gravity after a preset time period based on the trend slope when the trend slope exceeds a set threshold for a set number of consecutive sliding time windows, and to obtain the predicted weight reduction deviation. The third calculation module is used to calculate the feedback pressure correction amount based on the weight reduction deviation and to calculate the feedforward pressure pre-adjustment amount based on the predicted weight reduction deviation. The fourth calculation module is used to superimpose the feedback pressure correction amount with the feedforward pressure pre-adjustment amount to obtain the total pressure adjustment amount; The first control module is used to control the inflation or deflation of the airbag according to the total air pressure adjustment amount, so that the actual weight reduction force converges to the target weight reduction force.

[0014] This application also provides a control device for an anti-gravity rehabilitation treadmill, including a memory for storing computer programs; A processor is used to implement the steps of the control method for the anti-gravity rehabilitation treadmill when executing the computer program.

[0015] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the control method for the anti-gravity rehabilitation treadmill.

[0016] The control method for an anti-gravity rehabilitation treadmill provided in this application continuously collects time-series data of supporting gravity through a preset sliding time window. Instantaneous spikes exceeding a preset impact threshold are removed from the time-series data. The least squares method is used to fit the trend slope of the supporting gravity changing with time on the removed time-series data. When the trend slope in a consecutive set number of sliding time windows exceeds the set threshold, it is determined that there is a real drift trend. The predicted supporting gravity after a preset time period is predicted based on the trend slope to obtain the predicted weight reduction deviation. The feedforward air pressure pre-adjustment amount is calculated based on the predicted weight reduction deviation. This allows the aerodynamic adjustment action to precede the deviation expansion, effectively compensating for the inherent response delay of the aerodynamic system and realizing the transformation from ex-post compensation to ex-ante prediction. The weight reduction deviation is calculated based on the difference between the target weight reduction and the actual weight reduction. The feedback pressure correction is then calculated based on this deviation. The feedback pressure correction is superimposed with the feedforward pressure pre-adjustment to obtain the total pressure adjustment. The inflation or deflation of the airbag is controlled based on the total pressure adjustment to bring the actual weight reduction to the target weight reduction. The feedforward pressure pre-adjustment is responsible for anticipating predictable trend changes, while the feedback pressure correction is responsible for eliminating residual errors and sudden disturbances. The two work together to suppress the periodic fluctuations in weight reduction caused by the oscillation of the center of gravity and to avoid introducing additional weight reduction fluctuations due to feedback lag, thus effectively improving weight reduction stability.

[0017] The beneficial effects and methods of the control device and medium for the anti-gravity rehabilitation treadmill provided in this application are as described above. Attached Figure Description

[0018] To more clearly illustrate the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1A flowchart illustrating a control method for an anti-gravity rehabilitation treadmill provided in this application embodiment; Figure 2 A structural diagram of a control device for an anti-gravity rehabilitation treadmill provided in an embodiment of this application; Figure 3 A structural diagram of a control device for another anti-gravity rehabilitation treadmill provided in an embodiment of this application. Detailed Implementation

[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of this application.

[0021] The core of this application is to provide a control method, device, and medium for an anti-gravity rehabilitation treadmill to improve weight loss stability.

[0022] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0023] Figure 1 A flowchart of a control method for an anti-gravity rehabilitation treadmill provided in this application embodiment is shown below. Figure 1 As shown, a control method for an anti-gravity rehabilitation treadmill includes: S10: Acquire the calibrated body weight, target weight reduction force, and support gravity detected by the tread pressure sensor.

[0024] S11: Calculate the actual weight reduction force based on the human body's calibrated weight and supporting gravity.

[0025] S12: Calculate the weight reduction deviation based on the target weight reduction force and the actual weight reduction force.

[0026] S13: Continuously collect time series data of supporting gravity with a sliding time window of preset duration, remove instantaneous spikes in the time series data that exceed the preset impact threshold, and use the least squares method to fit the trend slope of the supporting gravity changing with time on the removed time series data.

[0027] S14: When the trend slope within a set number of consecutive sliding time windows exceeds the set threshold, predict the predicted support gravity after a preset time period based on the trend slope, and obtain the predicted weight reduction deviation.

[0028] S15: Calculate the feedback pressure correction amount based on the weight reduction deviation, and calculate the feedforward pressure pre-adjustment amount based on the predicted weight reduction deviation.

[0029] S16: The feedback pressure correction amount is superimposed with the feedforward pressure pre-adjustment amount to obtain the total pressure adjustment amount.

[0030] S17: Control the inflation or deflation of the airbag according to the total air pressure regulation amount, so that the actual weight reduction force converges to the target weight reduction force.

[0031] To make it easier to understand, the overall structure of the anti-gravity rehabilitation treadmill will be introduced below.

[0032] The anti-gravity rehabilitation treadmill includes a segmented airbag lifting and weight reduction body assembly, a split dual-track walking mechanism, a foot pressure sensing tread assembly, a landing adaptive cushioning mechanical module, and the whole machine frame and electrical hardware carrier.

[0033] The segmented airbag lifting and weight reduction machine assembly includes a waist ring-shaped sealed airbag, a left leg ring-shaped sealed airbag, a right leg ring-shaped sealed airbag, flexible sealing lips, a proportional servo pneumatic valve assembly, air pipelines, and a whole-machine weighing sensor (tread pressure sensor). The three airbags operate independently, corresponding to the waist and abdomen area, left leg area, and right leg area respectively, and are sealed by flexible sealing lips. The whole-machine weighing sensor collects the supporting gravity, providing mechanical input for the control algorithm.

[0034] Split-type dual-track walking mechanism: Two completely independent running belt modules, left and right. Each module includes a servo drive motor, reduction unit, running belt body, electromagnetic damping assembly (damping actuator), and a single-sided lead screw lifting mechanism. The left and right tracks are mechanically independent, allowing for separate speed adjustment, adjustment of applied motion damping, and support for single-sided locking. The speed, gradient, and damping of the left and right tracks can be set separately.

[0035] Foot pressure sensor tread assembly: A distributed foot pressure sensor array laid on the tread of the two running belts, covering the entire walking area, used to collect peak foot landing pressure and force distribution, providing foot mechanical feedback signals for control algorithms.

[0036] Adaptive landing buffer mechanical module: The buffer actuator is located under the runway and consists of an adjustable preload spring assembly and a displacement sensor; it receives commands from the control system to change the spring preload, thereby dynamically adjusting the landing buffer stiffness, and at the same time outputs a buffer compression displacement feedback signal.

[0037] The complete machine frame and electrical hardware carrier include the main equipment frame, electrical control cabinet, and host computer operation terminal; the electrical control cabinet carries the main control unit, completes the electrical connection with all pneumatic, motor, and sensor components, and executes the control method of the anti-gravity rehabilitation treadmill mentioned above.

[0038] In step S10, the calibrated body weight is the patient's weight value. The target weight loss force is preset by the doctor through a touch-screen terminal. The desired weight loss percentage or the target weight loss force can be input on the interface based on the patient's condition, rehabilitation stage, and training goals. The support gravity detected by the tread pressure sensor is real-time dynamic data, reflecting the actual pressure exerted on the tread belt by the patient's weight at any given moment.

[0039] In step S11, calculating the actual weight reduction based on the body's calibrated weight and the supporting gravity includes: multiplying the body's calibrated weight by the gravitational acceleration to obtain the body's calibrated weight; and subtracting the supporting gravity from the body's calibrated weight to obtain the actual weight reduction. This can be expressed as: Actual weight reduction = Body calibrated weight - Supporting gravity.

[0040] In step S12, the target weight reduction force is subtracted from the actual weight reduction force at the current moment in each control cycle to obtain the weight reduction deviation. The weight reduction deviation is a signed value; a positive value indicates that the current actual weight reduction force is insufficient, and the airbag lifting force needs to be increased; a negative value indicates that the current actual weight reduction force is excessive, and the airbag lifting force needs to be decreased. It is expressed by the formula: Weight Reduction Deviation = Target Weight Reduction Force - Actual Weight Reduction Force. This weight reduction deviation serves as the input to the feedback control and participates in the closed-loop calculation of the subsequent fuzzy sliding mode controller. Simultaneously, the weight reduction deviation undergoes dead-zone determination and mean filtering preprocessing to eliminate the interference of gait impacts and minor oscillations on the control decision.

[0041] In step S13, to predict the patient's center of gravity shift trend in advance, time-series analysis is performed on the support gravity data collected by the tread pressure sensor. The preset duration of the sliding time window can be 2 to 5 seconds. The time-series support gravity data includes instantaneous pressure spikes generated at the moment the heel strikes the ground. These instantaneous spikes are normal gait phenomena and do not represent the true center of gravity shift trend. The time-series data within the window is preprocessed to remove data points with instantaneous spikes exceeding a preset impact threshold. The preset impact threshold is set based on the statistical value of peak plantar pressure during normal human walking, for example, it can be set to 120% of the human body's calibrated weight. When the support gravity at a certain sampling point exceeds this threshold, it is determined to be a landing impact spike, and the data point is marked as invalid and not included in subsequent fitting calculations. After removing instantaneous spikes, the remaining valid data points within the window are linearly fitted using the least squares method. Suppose the window contains n valid data points, where the x-coordinate of each data point is the sampling time ti, and the y-coordinate is the corresponding supporting gravity Gi. The fitted line G = k·t + b is calculated using the least squares algorithm. The sign of the slope k reflects the direction of change of the supporting gravity within the preset duration of the sliding time window. A positive value indicates that the supporting gravity increases with time, while a negative value indicates that the supporting gravity decreases with time. The absolute value of the slope k reflects the rate of change of the supporting gravity.

[0042] In step S14, the trend slope k calculated for each sliding time window is continuously monitored. When the trend slope exceeds a set threshold for a set number of consecutive sliding time windows, it is determined that the patient has a genuine support gravity drift trend, rather than normal gait periodic fluctuations. The set number can be 3 to 5 windows. Multiple consecutive sliding time windows are used to avoid false triggering due to noise or abnormal gait in a single window, ensuring that feedforward adjustment is only activated when there is a continuous and stable drift trend. The set threshold is divided into a positive threshold and a negative threshold, for example, ±5 N / s. A trend slope exceeding the set threshold includes: a trend slope greater than the positive threshold and a trend slope less than the negative threshold. If the trend slope is greater than the positive threshold, it is determined that the support gravity is in a continuous upward trend; if the trend slope is less than the negative threshold, it is determined that the support gravity is in a continuous downward trend. Once a genuine drift trend is determined, the support gravity value is predicted after a preset time period based on the latest support gravity at the end of the current sliding time window and the fitted trend slope. The preset time period can be from 0.5 seconds to 2 seconds and can be dynamically adjusted according to the response delay characteristics of the pneumatic system. The prediction formula is: Predicted support gravity = Current measured support gravity + Trend slope × Preset time period. After obtaining the predicted support gravity, the predicted support gravity is subtracted from the human body's calibrated weight to obtain the predicted actual weight reduction; then, the predicted actual weight reduction is subtracted from the target weight reduction to obtain the predicted weight reduction deviation. This predicted value is used as the input to the feedforward control channel to calculate the feedforward pressure pre-adjustment amount for advance compensation.

[0043] In step S15, calculating the feedback pressure correction based on the weight reduction deviation includes: performing dead zone determination on the weight reduction deviation, setting weight reduction deviations whose absolute value is less than or equal to a preset dead zone threshold to zero, and retaining weight reduction deviations whose absolute value is greater than the preset dead zone threshold; performing mean filtering on the weight reduction deviation after dead zone determination to obtain the filtered weight reduction deviation; calculating the weight reduction deviation change rate based on the filtered weight reduction deviation of the current cycle and the previous cycle; and using the filtered weight reduction deviation and the weight reduction deviation change rate corresponding to the current cycle as inputs, and employing a fuzzy sliding mode controller to calculate the feedback pressure correction.

[0044] Dead zone determination is performed on weight reduction deviation. The purpose of dead zone determination is to eliminate control output jitter caused by sensor noise, small fluctuations or quantization errors, avoid frequent opening and closing of air valves near the target value, extend the working life of proportional servo air valves and improve system stability.

[0045] Specifically, a dead zone threshold is preset. This threshold is a positive number in Newtons, preferably ranging from 1% to 3% of the human body's calibrated weight. In each control cycle, the weight reduction deviation of the current cycle is read, and its absolute value is compared with the preset dead zone threshold. If the absolute value of the weight reduction deviation is less than or equal to the preset dead zone threshold, it indicates that the actual weight reduction force is very close to the target weight reduction force, and the deviation between the two is too small to trigger an effective control action. In this case, the weight reduction deviation is forcibly set to zero, so that it does not participate in any subsequent control calculations. If the absolute value of the weight reduction deviation is greater than the preset dead zone threshold, it indicates that there is a significant deviation between the actual weight reduction force and the target weight reduction force, and adjustment is required. In this case, the original value of the weight reduction deviation is retained, allowing it to enter the subsequent filtering and control calculation stages normally. Dead zone determination effectively suppresses high-frequency oscillations in the control quantity caused by small deviations, avoiding valve wear and energy waste caused by repeated micro-movements of the valve near the target value. The arithmetic mean of the weight reduction deviations after dead zone determination is calculated to obtain the filtered weight reduction deviation. Weight reduction deviation change rate = (current cycle filtered weight reduction deviation - previous cycle filtered weight reduction deviation) / control cycle.

[0046] The calculation of feedback pressure correction using a fuzzy sliding mode controller includes: the fuzzy sliding mode controller first calculates the weight reduction deviation e and the rate of change of weight reduction deviation based on the filtered weight reduction deviation e. Construct the sliding surface, the expression for which is s = c·e + Where c is a preset sliding surface coefficient, a positive number, used to balance the weights of deviation and deviation change rate in the sliding surface; the larger the value of c, the greater the weight of deviation in the sliding surface; the smaller the value of c, the greater the weight of deviation change rate. After constructing the sliding surface, the sliding surface s and the sliding surface change rate are used as the basis for the calculation. As input variables, perform a fuzzy inference process; specifically, for the sliding surface s and the rate of change of the sliding surface... Each value is fuzzified separately, and its precise value is mapped to the membership degree of each fuzzy subset. The internally stored fuzzy rule base contains multiple if-law rules, such as: if s is positive and... If s is positive and large, then the output control quantity is negative and large; if s is positive and small, then the output control quantity is negative and large. If the value of s is negative, then the output control quantity is zero; if s is zero and If the negative value is small, then the output control quantity is small; through three steps of fuzzification, inference, and defuzzification, the precise values ​​of s and s are obtained. The input value is mapped to a precise control output value; the defuzzification uses the centroid method to obtain a normalized fuzzy control quantity. The normalized fuzzy control quantity is multiplied by a preset output gain coefficient to obtain the feedback air pressure correction quantity; the preset output gain coefficient is calibrated based on the conversion relationship between air pressure and weight reduction force; the feedback air pressure correction quantity is a signed value; a positive value indicates that the airbag needs to be inflated to increase the weight reduction force, and a negative value indicates that the airbag needs to be deflated to reduce the weight reduction force.

[0047] The feedforward pressure pre-adjustment is calculated based on the predicted weight reduction deviation, including: calculating the product of the predicted weight reduction deviation and the feedforward control coefficient to obtain the feedforward pressure pre-adjustment; wherein, the feedforward control coefficient is calibrated based on the conversion relationship between air pressure and weight reduction force; the physical meaning of the feedforward control coefficient is: the amount of air pressure change required for each unit change in weight reduction force under the current working state of the airbag.

[0048] The feedforward control coefficient is a proportional coefficient used to convert predicted weight reduction deviation into feedforward air pressure pre-adjustment (air pressure). Its physical meaning is: the amount of air pressure change required to achieve a unit change in weight reduction force under the current operating state of the airbag. The feedforward control coefficient is obtained through actual measurement and calibration before the equipment leaves the factory. Specifically, a series of known air pressure values ​​are applied to the airbag, and the corresponding lifting force (weight reduction force) is measured. A calibration curve showing the conversion relationship between air pressure and weight reduction force is then fitted. During operation, based on the current actual operating air pressure collected by the airbag pressure sensor, linear interpolation is performed on the calibration curve to obtain the feedforward control coefficient at the current operating point.

[0049] In step S16, the feedback pressure correction amount and the feedforward pressure pre-adjustment amount are superimposed to obtain the total pressure adjustment amount; the superposition method is vector addition, if the signs are the same, they are added together to enhance, and if the signs are opposite, they cancel each other out to weaken.

[0050] In step S17, the sign of the total air pressure adjustment determines whether the airbag is inflated or deflated; if the total air pressure adjustment is positive, it means that the airbag lifting force needs to be increased to inflate the airbag; if the total air pressure adjustment is negative, it means that the airbag lifting force needs to be reduced to deflate the airbag.

[0051] The airbags include a lumbar airbag, a left leg airbag, and a right leg airbag. Inflation and deflation of the airbags are controlled based on the total air pressure regulation, including: acquiring the initial weight distribution coefficients for the lumbar, left, and right legs; updating these coefficients based on pressure distribution data collected by a foot pressure sensor array to obtain the current weight distribution coefficients for the lumbar, left, and right legs; dividing the total air pressure regulation into lumbar, left leg, and right leg air pressure regulation amounts based on these coefficients; controlling the inflation and deflation of the lumbar airbag based on the lumbar air pressure regulation amount, the left leg airbag based on the left leg air pressure regulation amount, and the right leg airbag based on the right leg air pressure regulation amount.

[0052] Before training begins, initial weight-loss distribution coefficients for the waist, left leg, and right leg are obtained according to the training plan set by the doctor. The initial weight-loss distribution coefficients determine the proportion of each airbag in the total weight-loss force. The sum of the initial weight-loss distribution coefficients for the waist, left leg, and right leg is 100%.

[0053] The initial weight-reduction distribution coefficient can be determined in one of the following ways: First, based on statistical data of human anatomy and clinical experience. For example, for normal gait training, the initial weight-reduction distribution coefficient for the waist is 40%, for the left leg it is 30%, and for the right leg it is 30%, totaling 100%. This initial weight-reduction distribution coefficient ensures that the weight-reduction force is reasonably distributed between the trunk and lower limbs, avoiding excessive load on any one airbag. Second, it can be manually set by the doctor using a touch-screen terminal based on the patient's specific condition. For example, for a patient with left hemiplegia, the doctor can increase the initial weight-reduction distribution coefficient for the left leg to reduce the load on the affected lower limb; for a patient with uneven strength in both lower limbs, the doctor can adjust the distribution coefficients for the left and right thighs separately.

[0054] During walking training, pressure distribution data from the plantar pressure sensor array is continuously collected, and the initial weight-loss distribution coefficient is dynamically updated accordingly to adapt the output ratio of each airbag to the patient's actual gait. Specifically, the plantar pressure sensor array is laid on the treads of the left and right running belts, fully covering the patient's foot landing area. In each control cycle, the output signals of each sensor unit in the plantar pressure sensor array are analyzed to calculate the total force on the left and right feet respectively. The total force on the left foot is the sum of the pressure values ​​of all sensor units on the left running belt, reflecting the proportion of body weight borne by the patient's left lower limb during the stance phase; the total force on the right foot is the sum of the pressure values ​​of all sensor units on the right running belt, reflecting the proportion of weight borne by the right lower limb. The total forces on the left and right feet are then compared. Based on the pressure distribution data collected by the plantar pressure sensor array, the initial weight reduction distribution coefficients for the waist, left leg, and right leg are updated to obtain the current weight reduction distribution coefficients for the waist, left leg, and right leg. This includes: summing the sensor output values ​​of the left and right tread areas in the plantar pressure sensor array to obtain the total force on the left and right foot; and updating the initial weight reduction distribution coefficients for the left and right legs based on the magnitude relationship between the total force on the left and right foot: when the total force on the left foot is greater than the total force on the right foot... Increase the initial weight-distribution coefficient of the left leg to obtain the current weight-distribution coefficient of the left leg, and decrease the initial weight-distribution coefficient of the right leg to obtain the current weight-distribution coefficient of the right leg. When the total force on the right foot is greater than that on the left foot, increase the initial weight-distribution coefficient of the right leg to obtain the current weight-distribution coefficient of the right leg, and decrease the initial weight-distribution coefficient of the left leg to obtain the current weight-distribution coefficient of the left leg. Adjust the weight-distribution coefficient of the waist based on the updated current weight-distribution coefficients of the left and right legs to obtain the current weight-distribution coefficient of the waist, so that the sum of the updated current weight-distribution coefficients of the waist, the left and right legs equals the initial sum (100%). When the total force on the left foot is greater than that on the right foot, it indicates that the patient's left lower limb bears a larger proportion of weight during walking, while the right lower limb bears relatively less weight. In this case, increase the weight-distribution coefficient of the left leg and correspondingly decrease the weight-distribution coefficient of the right leg so that the left thigh airbag provides greater support to share the load of the left lower limb. Conversely, when the total force on the right foot is greater than that on the left foot, the weight distribution coefficient of the right leg is increased and the weight distribution coefficient of the left leg is decreased accordingly. While adjusting the weight distribution coefficient of the legs, the weight distribution coefficient of the waist can also be adjusted in tandem. The sum of the updated three distribution coefficients is always 100%, ensuring that the total weight reduction force remains constant.The allocation coefficient is updated using a smoothing filter, ensuring that the adjustment range in each control cycle does not exceed the preset maximum step size, thus preventing sudden changes in the allocation coefficient that could cause imbalance or discomfort to the patient. Based on the current weight-loss allocation coefficient, the total pressure regulation is divided into three independent pressure regulation amounts, obtained by superimposing the feedback pressure correction and the feedforward pressure pre-adjustment. The allocation method is multiplicative: lumbar pressure regulation = total pressure regulation × current lumbar weight-loss allocation coefficient; left leg pressure regulation = total pressure regulation × current left leg weight-loss allocation coefficient; right leg pressure regulation = total pressure regulation × current right leg weight-loss allocation coefficient. Since the sum of the three current weight-loss allocation coefficients is 100%, the sum of the three pressure regulation amounts equals the total pressure regulation, ensuring the total regulation remains constant before and after allocation. Each pressure regulation amount is a signed value; a positive value indicates that the corresponding airbag needs to be inflated, and a negative value indicates that the corresponding airbag needs to be deflated. The signs of the three adjustment parameters can be the same (i.e., all three parameters inflate or deflate simultaneously) or different (e.g., the lumbar region inflates while the thigh region deflates), determined by the allocation coefficient and the current operating status of each parameter. The three pressure adjustment parameters are output to their respective airbag control channels, driving the proportional servo valves of each airbag to perform inflation or deflation. The control signals for all three airbags are output synchronously within the same 40ms control cycle, ensuring strict consistency in the timing of the pressure adjustments and preventing a momentary shift in the patient's center of gravity due to one parameter adjusting first. Each of the three airbags is equipped with an independent pressure sensor to collect the current actual operating pressure of each airbag in real time.

[0055] Based on the above embodiments, this application embodiment further includes: acquiring pre-set target training parameters; wherein, the target training parameters include target speed, target damping, target weight reduction force, and target cushioning stiffness; acquiring plantar pressure data collected by the plantar pressure sensor array, and calculating the plantar pressure change rate based on the plantar pressure data; when the plantar pressure change rate is greater than a preset change rate threshold, it is determined to be the landing impact phase; during the landing impact phase, the operating speed of the corresponding track is reduced to less than the target speed, the damping actuator is controlled to reduce the damping of the corresponding track to less than the target damping, and the cushioning actuator is controlled to reduce the cushioning stiffness of the corresponding track to less than the target damping. The target cushioning stiffness is set, and the actual weight reduction force is adjusted to be greater than the target weight reduction force. When the rate of change of plantar pressure is less than or equal to a preset rate of change threshold, and the plantar pressure is greater than or equal to a preset pressure threshold, it is determined to be the support phase. In the support phase, the actual force-bearing contact area of ​​the plantar surface is determined based on the plantar pressure data. When the actual force-bearing contact area of ​​the plantar surface is less than a preset area threshold, the actual weight reduction force is adjusted to be greater than the target weight reduction force, and / or the running speed of the corresponding track is reduced to less than the target speed. In the support phase, when the actual force-bearing contact area of ​​the plantar surface is greater than or equal to the preset area threshold, all training parameters are restored to the preset target training parameters.

[0056] During weight-loss walking training, patients may experience abnormal foot impact or inaccurate foot placement due to insufficient muscle strength, gait instability, or fatigue. Simple weight-loss closed-loop control cannot handle these transient events, requiring additional multi-dimensional adaptive adjustment mechanisms to ensure training safety and comfort. Therefore, based on weight-loss closed-loop control, a multi-dimensional linkage adjustment strategy based on real-time plantar pressure feedback is introduced. This strategy automatically adjusts track speed, damping, cushioning stiffness, and weight-reduction force according to the degree of foot impact and foot contact state; it can also include factors such as incline.

[0057] Before training begins, the doctor pre-sets the target training parameters for the patient using a touch-screen terminal. These parameters include target speed, target damping, target weight reduction, and target cushioning stiffness. Target speed is the desired linear velocity of the treadmill belt, measured in m / s, and is set by the doctor based on the patient's walking ability and rehabilitation stage. For patients requiring passive assisted training, the target speed can be set to a lower, constant value, allowing the treadmill belt to actively pull the patient's lower limbs to complete the stepping motion; for patients with some active walking ability, the target speed can serve as the upper speed limit in follow-up mode. Target damping is the desired resistance of the treadmill belt to the foot's movement, set by the doctor based on the patient's muscle strength level; patients with stronger muscles can have higher damping to increase the training load, while patients with weaker muscles or in the early stages of recovery should have lower damping to reduce training difficulty. Target weight reduction is the desired upward lifting force provided by the airbag weight reduction mechanism, set by the doctor based on the patient's weight and rehabilitation needs, measured in N. The target cushioning stiffness is the stiffness value of the runway's bottom cushioning actuator (which can be an electronically adjustable damping shock absorber). It is set by the physician based on the patient's joint tolerance and gait characteristics. Patients with sensitive joints or high impact upon landing are given a lower cushioning stiffness to achieve a softer landing feel, while patients with better joint stability are given a higher cushioning stiffness to enhance support. These target training parameters serve as the baseline and recovery target for subsequent adaptive adjustments.

[0058] During training, plantar pressure data is continuously acquired from a plantar pressure sensor array. This array is a distributed sensor array covering the left and right running belt treads, outputting a plantar pressure distribution map in each control cycle, containing the real-time pressure values ​​of each sensor unit. The plantar pressure change rate is calculated based on the total plantar pressure values ​​of the current and previous control cycles. The total plantar pressure value is the sum of the pressure values ​​of all sensor units on the left or right running belt tread at the current moment. The formula for calculating the plantar pressure change rate is: Plantar pressure change rate = (Total plantar pressure in the current cycle - Total plantar pressure in the previous cycle) / Control cycle duration. The plantar pressure change rate characterizes the degree of change in plantar pressure over time when the patient lands or pushes off the ground. During normal walking, pressure rises rapidly when the heel strikes the ground and drops rapidly when the toes leave the ground, with the pressure change rate showing a regular variation at different stages of the gait cycle. When the patient experiences an abnormally heavy landing motion, plantar pressure will rise sharply in a very short time, and the pressure change rate will significantly exceed the normal gait range. The calculated rate of change of plantar pressure is compared with a preset threshold value, which is a positive number in N / s. This threshold value is set based on statistical values ​​of plantar pressure change during normal walking; for example, it can be set as the human body's calibrated weight divided by 0.05 seconds. When the plantar pressure change rate is greater than the preset threshold, it indicates that the patient is experiencing an abnormally forceful foot strike, and the current moment is determined to be in the impact phase. When the plantar pressure change rate is less than or equal to the preset threshold, the current moment is determined to be in the non-impact phase.

[0059] Once the rate of change in plantar pressure is less than or equal to a preset threshold, the system further determines whether the patient has entered a stable support phase. Specifically, the current total plantar pressure value is read and compared with a preset pressure threshold. The preset pressure threshold is a positive number in N, typically set as a certain percentage of the body's calibrated weight, such as 20%. When the total plantar pressure is greater than or equal to the preset pressure threshold, it indicates that the patient's foot has firmly planted on the running belt and their weight has been smoothly transferred to that lower limb, thus determining the current moment as the support phase. When the total plantar pressure is less than the preset pressure threshold, it indicates that the patient's foot has not yet firmly planted or is in the swing phase.

[0060] When it is determined that the current moment is in the impact phase, a multi-dimensional linkage adjustment mechanism is immediately activated, simultaneously adjusting four dimensions: track speed, track damping, track cushioning stiffness, and weight reduction. First, the operating speed of the corresponding track is reduced to below the target speed. The corresponding track refers to the track on the same side as the foot that experienced the abnormal impact; if the left foot impacts, the speed of the left track is reduced, and if the right foot impacts, the speed of the right track is reduced. The magnitude of the speed reduction is proportional to the degree to which the plantar pressure change rate exceeds the impact threshold; the more intense the impact, the greater the speed reduction. However, the reduction within a single cycle does not exceed the preset maximum speed change step size to avoid sudden speed drops causing patient imbalance. Second, the damping actuator is controlled to reduce the damping of the corresponding track to below the target damping. The damping actuator can be a magnetic powder brake, which reduces the braking torque on the roller shaft by decreasing the excitation current, thus reducing the resistance of the running belt movement and reducing the resistance to abnormal muscle tension, helping the patient complete the landing action more smoothly. Third, the control system reduces the buffer stiffness of the corresponding track to less than the target buffer stiffness. This buffer actuator is an electronically controlled adjustable damping shock absorber. By changing the drive current, the damping characteristics inside the shock absorber are adjusted, reducing its equivalent stiffness. With reduced buffer stiffness, the track has a greater compression stroke and a longer buffer time when subjected to vertical impact, absorbing more landing impact energy and effectively reducing the impact load transmitted to the patient's lower limb joints. Fourth, the actual weight reduction force is adjusted to be greater than the target weight reduction force. Specifically, the control system applies a positive offset to the target weight reduction force, causing the airbag weight reduction mechanism to provide a greater lifting force than during normal training, to share the excessive load borne by the lower limbs at the moment of impact. The increase in weight reduction force is proportional to the impact intensity; the more severe the impact, the greater the increase in weight reduction force, but it does not exceed a preset safety limit. The adjustments in the above four dimensions are initiated synchronously within the same control cycle. The execution instructions are simultaneously sent to the drive motor, magnetic powder brake, electronically adjustable damping shock absorber, and proportional servo valve to ensure that the actions of each actuator are strictly consistent in timing, avoiding any incoordination felt by the patient due to one actuator acting first.

[0061] Once the current position is determined to be in the support phase, the actual contact area of ​​the foot is further determined based on plantar pressure data. The calculation method for the actual contact area is as follows: count the number of sensor units in the plantar pressure sensor array whose pressure values ​​exceed a preset contact threshold, multiply this number by the tread area represented by a single sensor unit, and obtain the effective contact area between the foot and the running belt at the current moment. When the actual contact area is less than the preset area threshold, it indicates that the patient's foot is not fully committed to the running belt, resulting in insufficient foot contact. The preset area threshold is set according to a certain proportion of the normal foot area, for example, it can be set to 60% of the normal foot area. When the patient's foot is not fully committed to the running belt, corresponding auxiliary adjustment measures are initiated: on the one hand, the actual weight reduction force is adjusted to be greater than the target weight reduction force, increasing the airbag support force to distribute more weight and reduce the load on the patient's legs due to insufficient foot contact; on the other hand, the running speed of the corresponding track is reduced to less than the target speed, limiting the running belt's speed and preventing the running belt from dragging the patient's foot and causing slippage or sprains when the foot is not fully committed. The two adjustment measures mentioned above can be implemented simultaneously or one of them can be chosen to be implemented.

[0062] When the rate of change of plantar pressure is less than or equal to a preset rate of change threshold, the plantar pressure is greater than or equal to a preset pressure threshold, and the actual contact area of ​​the foot is greater than or equal to a preset area threshold, it indicates that the patient has recovered from abnormal foot impact or unstable foot placement and entered a stable, normal support phase. At this point, the training parameters are gradually restored to the target training parameters preset by the doctor, including restoring the speed to the target speed, the damping to the target damping, the cushioning stiffness to the target cushioning stiffness, and the weight reduction force to the target weight reduction force.

[0063] Restoring each training parameter to the preset target training parameters includes: calculating the gravity reduction deviation between the current actual gravity reduction force and the target gravity reduction force, the velocity deviation between the current actual velocity and the target velocity, the damping deviation between the current actual damping and the target damping, and the buffer stiffness deviation between the current actual buffer stiffness and the target buffer stiffness; performing PID closed-loop calculations based on the gravity reduction force deviation, velocity deviation, damping deviation, and buffer stiffness deviation to obtain the gravity reduction force recovery adjustment, velocity recovery adjustment, damping recovery adjustment, and buffer stiffness recovery adjustment; and adjusting the gravity reduction force recovery adjustment, velocity recovery adjustment, and damping recovery adjustment... The magnitude and buffer stiffness recovery adjustment are limited to ensure that the absolute value of each recovery adjustment does not exceed its corresponding preset single-cycle limit value. The airbag inflation or deflation is controlled according to the limited weight reduction force recovery adjustment, the drive motor of the corresponding runway is controlled according to the limited speed recovery adjustment, the damping actuator of the corresponding runway is controlled according to the limited damping recovery adjustment, and the buffer actuator at the bottom of the runway is controlled according to the limited buffer stiffness recovery adjustment, until the actual weight reduction force is restored to the target weight reduction force, the actual speed is restored to the target speed, the actual damping is restored to the target damping, and the buffer stiffness is restored to the target buffer stiffness.

[0064] Calculation of weight reduction deviation: Read the current actual weight reduction value (obtained by subtracting the current support weight from the body's calibrated weight), and subtract it from the target weight reduction value to obtain the weight reduction deviation. This can be expressed as: Weight Reduction Deviation = Target Weight Reduction - Current Actual Weight Reduction. A positive deviation indicates that the current actual weight reduction is lower than the target value, requiring inflation of the airbag to increase the lifting force; a negative deviation indicates that the current actual weight reduction is higher than the target value, requiring deflation of the airbag to reduce the lifting force.

[0065] Speed ​​deviation calculation: The actual operating speed of the track at the current moment is collected by the motor encoder, and then subtracted from the target speed to obtain the speed deviation. Expressed as: Speed ​​Deviation = Target Speed ​​- Current Actual Speed. A positive speed deviation indicates that the current actual speed is lower than the target value, requiring an increase in motor drive torque to increase the speed; a negative deviation indicates that the current actual speed is higher than the target value, requiring a decrease in motor drive torque or the application of brakes to reduce the speed. The speed deviation includes the left and right running belt speed deviations, and correspondingly, the target speed includes the left and right running belt target speeds.

[0066] Calculation of damping deviation: The current actual damping is determined based on the current operating current of the damping actuator (magnetic powder brake) or the feedback value of the force sensor. This actual damping is then subtracted from the target damping to obtain the damping deviation. This can be expressed as: Damping deviation = Target damping - Current actual damping. A positive damping deviation indicates that the current actual damping is lower than the target value, requiring an increase in the excitation current to increase damping. A negative deviation indicates that the current actual damping is higher than the target value, requiring a decrease in the excitation current to decrease damping. The damping deviation includes the left running track damping deviation and the right running track damping deviation; correspondingly, the target damping includes the left running track target damping and the right running track target damping.

[0067] Calculation of buffer stiffness deviation: The actual buffer stiffness is determined by the displacement sensor or internal state estimation of the electronically adjustable damping shock absorber. This actual buffer stiffness is then subtracted from the target buffer stiffness to obtain the buffer stiffness deviation. This can be expressed as: Buffer stiffness deviation = Target buffer stiffness - Actual buffer stiffness. A positive buffer stiffness deviation indicates that the actual buffer stiffness is lower than the target value, requiring adjustment of the shock absorber drive current to increase stiffness. A negative deviation indicates that the actual buffer stiffness is higher than the target value, requiring reduction of the drive current to decrease stiffness. The buffer stiffness deviation includes the left running belt buffer stiffness deviation and the right running belt buffer stiffness deviation; correspondingly, the target buffer stiffness includes the left running belt target buffer stiffness and the right running belt target buffer stiffness.

[0068] The four deviations mentioned above are calculated in parallel within the same control cycle, and are uncoupled and independent of each other.

[0069] Proportional-Integral-Derivative (PID) closed-loop control is performed on the four deviations to generate the recovery adjustment for each deviation. Taking the gravity reduction recovery adjustment as an example, its calculation formula is: Gravity Reduction Recovery Adjustment = Kp_f × Gravity Reduction Deviation + Ki_f × Gravity Reduction Deviation Integral + Kd_f × Gravity Reduction Deviation Change Rate; where Kp_f is the proportional coefficient, Ki_f is the integral coefficient, and Kd_f is the derivative coefficient. The gravity reduction deviation integral is calculated by multiplying the current gravity reduction deviation by the control cycle duration in each control cycle and then adding it to the integral term. However, the integral term needs to be limited to prevent integral saturation. The gravity reduction deviation change rate is calculated by subtracting the gravity reduction deviation of the previous cycle from the current cycle's gravity reduction deviation and then dividing by the control cycle duration. The PID calculation formulas for speed recovery adjustment, damping recovery adjustment, and buffer stiffness recovery adjustment are the same as those described above, but their proportional, integral, and derivative coefficients are independently tuned and different from each other to adapt to the response characteristics of different actuators.

[0070] Each recovery adjustment value obtained from the PID closed-loop calculation is subjected to amplitude limiting. The purpose of amplitude limiting is to limit the output change of each actuator within a single cycle, preventing sudden parameter changes during the recovery process from causing shock or imbalance to the patient.

[0071] Taking the gravity recovery adjustment as an example, a single-cycle limit value is preset for the gravity recovery channel. This limit value represents the maximum allowable air pressure change of the airbag system in each control cycle. The absolute value of the theoretically calculated gravity recovery adjustment is compared with this limit value: if the absolute value of the theoretical recovery adjustment is less than or equal to the preset single-cycle limit value, it indicates that the adjustment is within the safe range, and the theoretical value is directly used as the gravity recovery adjustment. If the theoretical recovery adjustment is positive and greater than the preset single-cycle limit value, it indicates that the adjustment exceeds the safe range in the positive direction, and the gravity recovery adjustment is limited to a positive limit value, with the excess adjustment being output gradually in subsequent control cycles. If the theoretical recovery adjustment is negative and its absolute value is greater than the preset single-cycle limit value, it indicates that the adjustment exceeds the safe range in the negative direction, and the gravity recovery adjustment is limited to a negative limit value, with the excess adjustment being output gradually in subsequent control cycles. The velocity recovery adjustment, damping recovery adjustment, and buffer stiffness recovery adjustment are each subject to the same limiting treatment using their own independent single-cycle limit values.

[0072] The limited-amplitude recovery adjustment values ​​are converted into corresponding actuator drive signals and sent synchronously within the same control cycle. The limited-amplitude weight reduction recovery adjustment value is converted into a drive signal for the proportional servo air valve via a digital-to-analog converter or pulse width modulation output module. If the adjustment value is positive, a drive signal is output to the inflation valve of the corresponding airbag, inflating the airbag and causing the weight reduction force to recover towards the target value; if the adjustment value is negative, a drive signal is output to the deflation valve, deflating the airbag and causing the weight reduction force to fall back towards the target value. The limited-amplitude speed recovery adjustment value is sent to the servo drive motor of the corresponding track. If the adjustment value is positive, the motor increases the output torque to accelerate the running belt; if the adjustment value is negative, the motor decreases the output torque to decelerate the running belt. The limited-amplitude damping recovery adjustment value is sent to the magnetic powder brake of the corresponding track via an analog output or pulse width modulation output module; if the adjustment value is positive, the excitation current is increased to increase damping; if the adjustment value is negative, the excitation current is decreased to decrease damping. The buffer stiffness recovery adjustment amount after amplitude limiting is sent to the electronically controlled adjustable damping shock absorber at the bottom of the runway through the analog output or pulse width modulation output module; if the adjustment amount is positive, the shock absorber drive current is adjusted to increase the buffer stiffness; if the adjustment amount is negative, the shock absorber drive current is adjusted to decrease the buffer stiffness.

[0073] After one control cycle of adjustment, each actual value moves closer to the target value by a certain step. In the next control cycle, the deviation between the current actual value and the target value is recalculated, and the entire process of deviation calculation, PID operation, amplitude limiting, and recovery action is repeated. Due to the amplitude limiting process, if the entire recovery cannot be completed in a single cycle, the remaining deviation is gradually corrected by subsequent control cycles. This process is repeated until the actual values ​​of all training parameters are restored to within the allowable error range of the target value.

[0074] The control method for an anti-gravity rehabilitation treadmill disclosed in this application continuously collects time-series data of the supporting gravity within a preset sliding time window. Instantaneous spikes exceeding a preset impact threshold are removed from the time-series data. The least squares method is used to fit the trend slope of the supporting gravity changing over time in the removed time-series data. When the trend slope exceeds the preset threshold within a consecutive set number of sliding time windows, a true drift trend is determined. Based on the trend slope, the predicted supporting gravity after a preset time period is predicted, resulting in a predicted weight reduction deviation. The feedforward air pressure pre-adjustment is calculated based on the predicted weight reduction deviation. This allows the aerodynamic adjustment action to precede the deviation expansion, effectively compensating for the inherent response delay of the aerodynamic system and realizing a shift from post-event compensation to pre-event prediction. The weight reduction deviation is calculated based on the difference between the target weight reduction and the actual weight reduction. The feedback pressure correction is then calculated based on this deviation. The feedback pressure correction is superimposed with the feedforward pressure pre-adjustment to obtain the total pressure adjustment. The inflation or deflation of the airbag is controlled based on the total pressure adjustment to bring the actual weight reduction to the target weight reduction. The feedforward pressure pre-adjustment is responsible for anticipating predictable trend changes, while the feedback pressure correction is responsible for eliminating residual errors and sudden disturbances. The two work together to suppress the periodic fluctuations in weight reduction caused by the oscillation of the center of gravity and to avoid introducing additional weight reduction fluctuations due to feedback lag, thus effectively improving weight reduction stability.

[0075] Based on the above embodiments, the embodiments of this application also include the determination of abnormal situations.

[0076] Air pressure anomaly detection: Collect the measured values ​​of the pressure sensors of each airbag. If the measured value exceeds the preset upper and lower limits of the hardware pressure for multiple sampling cycles, or if the pressure does not change for a long time after the inflation / deflation adjustment command is output, or if the sensor signal is stuck / abnormally jumps, then an air pressure anomaly is determined, and the abnormal airbag channel, pressure data and timestamp are recorded.

[0077] Motor overload detection: Collect the real-time operating current of the motor and the encoder speed feedback, and combine them with the thermal accumulation model; when the motor current continuously exceeds the overload threshold, or when the output drive command is almost zero and the current is high, and the thermal accumulation value exceeds the heat threshold, the motor is determined to be overloaded, and the motor number, current, command speed and actual speed are recorded.

[0078] Foot pressure exceeding limit detection: Acquire real-time force data from left and right foot pressure sensor arrays, and filter out instantaneous spikes caused by foot impact; when the steady-state foot pressure exceeds the safe pressure threshold for a certain period of time, it is determined that the foot pressure exceeds the limit, and the corresponding foot channel and the magnitude of the force are recorded.

[0079] Communication disconnection detection: Perform periodic communication queries on each peripheral module. If there are no valid response messages for several consecutive cycles, or if the data packet fails to pass multiple verifications, the corresponding peripheral is determined to be disconnected, and the disconnected device is marked with an identifier and a frame loss count is recorded.

[0080] Furthermore, anti-conflict settings are included: When the device is in walking training mode, if it receives operations such as calibration, mechanism self-test, factory reset, switching debugging modes, modifying hardware protection thresholds, or modifying sensor zero-point parameters, it is considered a mode conflict and is directly rejected. When the device is in calibration mode, if it receives a walking training start command, it is considered a conflict and is rejected. When the device is in fault self-lock or emergency stop lock state, all operations except fault query and manual reset are considered conflicts and rejected. During the operation of gait following real-time weight reduction mode, if it receives a forced switch to constant pressure weight reduction mode, it is considered a conflict and is directly rejected; if one side is in medium overload protection state, if it receives a command to forcibly increase the rotation speed of that side or forcibly pressurize the airbag, it is considered a conflict and is directly rejected. If the upper and lower limit logic of the new input parameter itself is contradictory, it is considered a parameter conflict and is directly rejected.

[0081] In the above embodiments, the control method for the anti-gravity rehabilitation treadmill has been described in detail. This application also provides embodiments corresponding to the control device for the anti-gravity rehabilitation treadmill. It should be noted that this application describes the embodiments of the device from two perspectives: one based on functional modules and the other based on hardware.

[0082] Figure 2 A structural diagram of a control device for an anti-gravity rehabilitation treadmill provided in an embodiment of this application is shown below. Figure 2 As shown, a control device for an anti-gravity rehabilitation treadmill includes: The first acquisition module 10 is used to acquire the human body's calibrated weight, target weight reduction force, and support gravity detected by the tread pressure sensor; The first calculation module 11 is used to calculate the actual weight reduction force based on the human body's calibrated weight and supporting gravity. The second calculation module 12 is used to calculate the weight reduction deviation based on the target weight reduction force and the actual weight reduction force. The fitting module 13 is used to continuously collect time series data of supporting gravity in a sliding time window of preset duration, remove instantaneous spikes in the time series data that exceed the preset impact threshold, and use the least squares method to fit the trend slope of the supporting gravity changing with time on the removed time series data. Prediction module 14 is used to predict the predicted support gravity after a preset time period based on the trend slope when the trend slope exceeds a set threshold for a set number of consecutive sliding time windows, and to obtain the predicted weight reduction deviation. The third calculation module 15 is used to calculate the feedback pressure correction amount based on the weight reduction deviation and to calculate the feedforward pressure pre-adjustment amount based on the predicted weight reduction deviation. The fourth calculation module 16 is used to superimpose the feedback pressure correction amount and the feedforward pressure pre-adjustment amount to obtain the total pressure adjustment amount. The first control module 17 is used to control the inflation or deflation of the airbag according to the total air pressure adjustment amount, so that the actual weight reduction force converges to the target weight reduction force.

[0083] Based on the above embodiments, in an optional embodiment, the first computing module includes: The first calculation unit is used to calculate the product of the human body's calibrated weight and gravitational acceleration as the human body's calibrated gravity. The second calculation unit is used to subtract the supporting weight from the human body's calibrated weight to obtain the actual weight reduction.

[0084] Based on the above embodiments, in an optional embodiment, the third computing module includes: The dead zone determination unit is used to determine the dead zone of the weight loss deviation. It sets the weight loss deviation with an absolute value less than or equal to the preset dead zone threshold to zero, and retains the weight loss deviation with an absolute value greater than the preset dead zone threshold. The mean filtering unit is used to perform mean filtering on the weight reduction deviation after dead zone determination to obtain the filtered weight reduction deviation. The third calculation unit is used to calculate the rate of change of weight reduction deviation based on the filtered weight reduction deviation between the current cycle and the previous cycle. The fourth calculation unit is used to calculate the feedback air pressure correction amount using a fuzzy sliding mode controller, taking the filtered weight reduction deviation and the weight reduction deviation change rate corresponding to the current cycle as input.

[0085] Based on the above embodiments, in an optional embodiment, the third computing module includes: The fifth calculation unit is used to calculate the product of the predicted weight reduction deviation and the feedforward control coefficient to obtain the feedforward air pressure pre-adjustment amount; wherein, the feedforward control coefficient is calibrated based on the conversion relationship between air pressure and weight reduction force.

[0086] Based on the above embodiments, in one optional embodiment, the first control module includes: The acquisition unit is used to acquire the initial weight loss distribution coefficient of the waist, the initial weight loss distribution coefficient of the left leg, and the initial weight loss distribution coefficient of the right leg; The update unit is used to update the initial weight reduction distribution coefficients of the waist, left leg, and right leg based on the pressure distribution data collected by the plantar pressure sensor array, so as to obtain the current weight reduction distribution coefficients of the waist, left leg, and right leg. The splitting unit is used to split the total air pressure regulation into the air pressure regulation of the waist, the air pressure regulation of the left leg, and the air pressure regulation of the right leg based on the current weight distribution coefficient of the waist, the current weight distribution coefficient of the left leg, and the current weight distribution coefficient of the right leg. The first control unit is used to control the inflation or deflation of the lumbar airbag according to the lumbar air pressure adjustment, to control the inflation or deflation of the left leg airbag according to the left leg air pressure adjustment, and to control the inflation or deflation of the right leg airbag according to the right leg air pressure adjustment.

[0087] Based on the above embodiments, in an optional embodiment, it further includes: The second acquisition module is used to acquire pre-set target training parameters; wherein, the target training parameters include target velocity, target damping, target weight reduction force and target buffer stiffness; The fifth calculation module is used to acquire plantar pressure data collected by the plantar pressure sensor array and calculate the plantar pressure change rate based on the plantar pressure data. The first judgment module is used to determine the impact phase when the change rate of plantar pressure is greater than the preset change rate threshold. The second control module is used to reduce the running speed of the corresponding runway to less than the target speed during the landing impact phase, control the damping actuator to reduce the damping of the corresponding runway to less than the target damping, control the buffer actuator to reduce the buffer stiffness of the corresponding runway to less than the target buffer stiffness, and adjust the actual weight reduction force to be greater than the target weight reduction force. The second determination module is used to determine the support phase when the rate of change of plantar pressure is less than or equal to a preset rate of change threshold and the plantar pressure is greater than or equal to a preset pressure threshold. The third control module is used to determine the actual force contact area of ​​the sole based on the plantar pressure data during the support phase. When the actual force contact area of ​​the sole is less than the preset area threshold, the actual weight reduction force is adjusted to be greater than the target weight reduction force, and / or the running speed of the corresponding track is reduced to be less than the target speed. The fourth control module is used to restore each training parameter to the preset target training parameter when the actual force contact area of ​​the sole is greater than or equal to the preset area threshold during the support phase.

[0088] Based on the above embodiments, in one optional embodiment, the fourth control module includes: The sixth calculation unit is used to calculate the gravity reduction deviation between the current actual gravity reduction force and the target gravity reduction force, the velocity deviation between the current actual speed and the target speed, the damping deviation between the current actual damping and the target damping, and the buffer stiffness deviation between the current actual buffer stiffness and the target buffer stiffness. The seventh calculation unit is used to perform PID closed-loop calculations based on the gravity reduction deviation, velocity deviation, damping deviation, and buffer stiffness deviation, respectively, to obtain the gravity reduction recovery adjustment amount, velocity recovery adjustment amount, damping recovery adjustment amount, and buffer stiffness recovery adjustment amount. The amplitude limiting processing unit is used to limit the gravity recovery adjustment amount, velocity recovery adjustment amount, damping recovery adjustment amount and buffer stiffness recovery adjustment amount respectively, so that the absolute value of each recovery adjustment amount does not exceed the corresponding preset single-cycle amplitude limiting value. The second control unit is used to control the inflation or deflation of the airbags according to the adjustment amount of the weight reduction force after the limit, to control the drive motor of the corresponding runway according to the adjustment amount of the speed after the limit, to control the damping actuator of the corresponding runway according to the adjustment amount of the damping after the limit, and to control the buffer actuator at the bottom of the runway according to the adjustment amount of the buffer stiffness after the limit, until the actual weight reduction force is restored to the target weight reduction force, the actual speed is restored to the target speed, the actual damping is restored to the target damping, and the buffer stiffness is restored to the target buffer stiffness.

[0089] Since the embodiments of the apparatus and the embodiments of the method correspond to each other, please refer to the description of the embodiments of the method for the embodiments of the apparatus, which will not be repeated here.

[0090] Figure 3 A structural diagram of another anti-gravity rehabilitation treadmill control device provided in this application embodiment is shown below. Figure 3 As shown, the control device of the anti-gravity rehabilitation treadmill includes: a memory 20 for storing computer programs; The processor 21 is used to execute computer programs to implement the steps of the control method for the anti-gravity rehabilitation treadmill as described in the above embodiments.

[0091] The control device for the anti-gravity rehabilitation treadmill provided in this embodiment may include, but is not limited to, smartphones, tablets, laptops, or desktop computers.

[0092] The processor 21 may include one or more processing cores, such as a quad-core processor or an octa-core processor. The processor 21 may be implemented using at least one of the following hardware forms: Digital Signal Processor (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 21 may also include a main processor and a coprocessor. The main processor, also known as the Central Processing Unit (CPU), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, the processor 21 may integrate a Graphics Processing Unit (GPU), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, the processor 21 may also include an Artificial Intelligence (AI) processor, which handles computational operations related to machine learning.

[0093] The memory 20 may include one or more computer-readable storage media, which may be non-transitory. The memory 20 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In this embodiment, the memory 20 is used to store at least the following computer program 201, which, after being loaded and executed by the processor 21, is capable of implementing the relevant steps of the control method for the anti-gravity rehabilitation treadmill disclosed in any of the foregoing embodiments. In addition, the resources stored in the memory 20 may also include an operating system 202 and data 203, and the storage method may be temporary or permanent storage. The operating system 202 may include Windows, Unix, Linux, etc. The data 203 may include, but is not limited to, human body calibration weight, target weight loss force, etc.

[0094] In some embodiments, the control device of the antigravity rehabilitation treadmill may further include a display screen 22, an input / output interface 23, a communication interface 24, a power supply 25, and a communication bus 26.

[0095] Those skilled in the art will understand that Figure 3 The structure shown does not constitute a limitation on the control device of the antigravity rehabilitation treadmill and may include more or fewer components than shown.

[0096] This application provides a control device for an anti-gravity rehabilitation treadmill, including a memory and a processor. When the processor executes a program stored in the memory, it can perform the following methods: acquire a calibrated body weight, a target weight reduction force, and a support weight detected by a tread pressure sensor; calculate the actual weight reduction force based on the calibrated body weight and the support weight; calculate the weight reduction deviation based on the target weight reduction force and the actual weight reduction force; continuously collect time-series data of the support weight using a preset sliding time window; remove instantaneous spikes in the time-series data that exceed a preset impact threshold; and process the removed time-series data... The data is fitted using the least squares method to determine the trend slope of the supporting gravity over time. When the trend slope exceeds a set threshold within a set number of consecutive sliding time windows, the predicted supporting gravity after a preset time period is predicted based on the trend slope, and the predicted weight reduction deviation is obtained. The feedback air pressure correction is calculated based on the weight reduction deviation, and the feedforward air pressure pre-adjustment is calculated based on the predicted weight reduction deviation. The feedback air pressure correction and the feedforward air pressure pre-adjustment are superimposed to obtain the total air pressure adjustment. The airbag is inflated or deflated based on the total air pressure adjustment to bring the actual weight reduction force to the target weight reduction force.

[0097] Finally, this application also provides an embodiment corresponding to a computer-readable storage medium. The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps described in the control method for the anti-gravity rehabilitation treadmill of the above method embodiment.

[0098] It is understood that if the methods in the above embodiments are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and executes all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0099] The control method, apparatus, and medium for an anti-gravity rehabilitation treadmill provided in this application have been described in detail above. The various embodiments in the specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section. It should be noted that those skilled in the art can make several improvements and modifications to this application without departing from the principles of this application, and these improvements and modifications also fall within the protection scope of the claims of this application.

[0100] It should also be noted that, in this specification, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

Claims

1. A control method for an anti-gravity rehabilitation treadmill, characterized in that, include: Acquire the calibrated body weight, target weight reduction force, and supporting gravity detected by the tread pressure sensor; The actual weight reduction force is calculated based on the human body's calibrated weight and the supporting gravity. Calculate the weight reduction deviation based on the target weight reduction force and the actual weight reduction force; The time-series data of the supporting gravity is continuously collected in a sliding time window of a preset duration. Instantaneous spikes in the time-series data that exceed a preset impact threshold are removed. The least squares method is used to fit the trend slope of the supporting gravity changing with time on the removed time-series data. When the trend slope exceeds a set threshold within a set number of consecutive sliding time windows, the predicted support gravity after a preset time period is predicted based on the trend slope, and the predicted weight reduction deviation is obtained. The feedback pressure correction amount is calculated based on the weight reduction deviation, and the feedforward pressure pre-adjustment amount is calculated based on the predicted weight reduction deviation. The total pressure regulation is obtained by superimposing the feedback pressure correction amount with the feedforward pressure pre-adjustment amount. The airbag is inflated or deflated according to the total air pressure regulation amount, so that the actual weight reduction force converges to the target weight reduction force.

2. The control method for the anti-gravity rehabilitation treadmill according to claim 1, characterized in that, The actual weight reduction force is calculated based on the calibrated body weight and the supporting gravity, including: The product of the human body's calibrated weight and gravitational acceleration is taken as the human body's calibrated gravity; The actual weight reduction force is obtained by subtracting the supporting weight from the human body's calibrated weight.

3. The control method for the anti-gravity rehabilitation treadmill according to claim 1, characterized in that, The feedback pressure correction amount is calculated based on the weight reduction deviation, including: The weight loss deviation is subjected to dead zone determination. The weight loss deviations whose absolute value is less than or equal to the preset dead zone threshold are set to zero, and the weight loss deviations whose absolute value is greater than the preset dead zone threshold are retained. The weight loss deviation after dead zone determination is filtered by mean filtering to obtain the filtered weight loss deviation. The rate of change of weight reduction deviation is calculated based on the filtered weight reduction deviation between the current cycle and the previous cycle. Using the filtered weight reduction deviation and the rate of change of the weight reduction deviation corresponding to the current cycle as inputs, a fuzzy sliding mode controller is used to calculate the feedback air pressure correction.

4. The control method for the anti-gravity rehabilitation treadmill according to claim 1, characterized in that, The feedforward pressure pre-adjustment is calculated based on the predicted weight reduction deviation, including: The product of the predicted weight reduction deviation and the feedforward control coefficient is calculated to obtain the feedforward pressure pre-adjustment amount; The feedforward control coefficient is calibrated based on the relationship between air pressure and gravity reduction.

5. The control method for the anti-gravity rehabilitation treadmill according to claim 1, characterized in that, Controlling the inflation or deflation of the airbag according to the total air pressure regulation includes: Obtain the initial weight loss distribution coefficients for the waist, left leg, and right leg; The initial weight loss distribution coefficients of the waist, left leg, and right leg are updated based on the pressure distribution data collected by the plantar pressure sensor array to obtain the current weight loss distribution coefficients of the waist, left leg, and right leg. Based on the current weight loss distribution coefficient of the waist, the current weight loss distribution coefficient of the left leg, and the current weight loss distribution coefficient of the right leg, the total air pressure regulation is divided into waist air pressure regulation, left leg air pressure regulation, and right leg air pressure regulation. The inflation or deflation of the lumbar airbag is controlled according to the lumbar air pressure adjustment amount, the inflation or deflation of the left leg airbag is controlled according to the left leg air pressure adjustment amount, and the inflation or deflation of the right leg airbag is controlled according to the right leg air pressure adjustment amount.

6. The control method for the anti-gravity rehabilitation treadmill according to claim 1, characterized in that, Also includes: Obtain pre-set target training parameters; wherein, the target training parameters include target velocity, target damping, target weight reduction force, and target buffer stiffness; Acquire plantar pressure data collected by the plantar pressure sensor array, and calculate the plantar pressure change rate based on the plantar pressure data; When the rate of change of plantar pressure is greater than a preset rate of change threshold, it is determined to be the impact phase of foot landing; During the landing impact phase, the running speed of the corresponding track is reduced to less than the target speed, the damping actuator is controlled to reduce the damping of the corresponding track to less than the target damping, the buffer actuator is controlled to reduce the buffer stiffness of the corresponding track to less than the target buffer stiffness, and the actual weight reduction force is adjusted to be greater than the target weight reduction force. When the rate of change of plantar pressure is less than or equal to a preset rate of change threshold, and the plantar pressure is greater than or equal to a preset pressure threshold, it is determined to be the support phase; During the support phase, the actual contact area of ​​the sole is determined based on the plantar pressure data. When the actual contact area of ​​the sole is less than a preset area threshold, the actual weight reduction force is adjusted to be greater than the target weight reduction force, and / or the running speed of the corresponding track is reduced to be less than the target speed. During the support phase, when the actual contact area of ​​the foot is greater than or equal to the preset area threshold, the training parameters are restored to the preset target training parameters.

7. The control method for the anti-gravity rehabilitation treadmill according to claim 6, characterized in that, Restoring each training parameter to the preset target training parameters, including: Calculate the weight reduction force deviation between the current actual weight reduction force and the target weight reduction force, the speed deviation between the current actual speed and the target speed, the damping deviation between the current actual damping and the target damping, and the buffer stiffness deviation between the current actual buffer stiffness and the target buffer stiffness, respectively. PID closed-loop calculations are performed based on the gravity reduction deviation, the velocity deviation, the damping deviation, and the buffer stiffness deviation to obtain the gravity reduction recovery adjustment amount, the velocity recovery adjustment amount, the damping recovery adjustment amount, and the buffer stiffness recovery adjustment amount. The gravity recovery adjustment amount, the velocity recovery adjustment amount, the damping recovery adjustment amount, and the buffer stiffness recovery adjustment amount are respectively limited to ensure that the absolute value of each recovery adjustment amount does not exceed its corresponding preset single-cycle limit value. The airbag is inflated or deflated according to the weight reduction force recovery adjustment amount after the limit, the drive motor of the corresponding runway is controlled according to the speed recovery adjustment amount after the limit, the damping actuator of the corresponding runway is controlled according to the damping recovery adjustment amount after the limit, and the buffer actuator at the bottom of the runway is controlled according to the buffer stiffness recovery adjustment amount after the limit, until the actual weight reduction force is restored to the target weight reduction force, the actual speed is restored to the target speed, the actual damping is restored to the target damping, and the buffer stiffness is restored to the target buffer stiffness.

8. A control device for an anti-gravity rehabilitation treadmill, characterized in that, include: The first acquisition module is used to acquire the human body's calibrated weight, target weight reduction force, and support gravity detected by the tread pressure sensor; The first calculation module is used to calculate the actual weight reduction force based on the human body's calibrated weight and the supporting gravity. The second calculation module is used to calculate the weight reduction deviation based on the target weight reduction force and the actual weight reduction force. The fitting module is used to continuously collect time-series data of the supporting gravity in a sliding time window of a preset duration, remove instantaneous spikes in the time-series data that exceed a preset impact threshold, and use the least squares method to fit the trend slope of the supporting gravity changing with time on the removed time-series data. The prediction module is used to predict the predicted support gravity after a preset time period based on the trend slope when the trend slope exceeds a set threshold for a set number of consecutive sliding time windows, and to obtain the predicted weight reduction deviation. The third calculation module is used to calculate the feedback pressure correction amount based on the weight reduction deviation and to calculate the feedforward pressure pre-adjustment amount based on the predicted weight reduction deviation. The fourth calculation module is used to superimpose the feedback pressure correction amount with the feedforward pressure pre-adjustment amount to obtain the total pressure adjustment amount; The first control module is used to control the inflation or deflation of the airbag according to the total air pressure adjustment amount, so that the actual weight reduction force converges to the target weight reduction force.

9. A control device for an anti-gravity rehabilitation treadmill, characterized in that, Includes memory used to store computer programs; A processor, configured to execute the computer program to implement the steps of the control method for the antigravity rehabilitation treadmill as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the control method for the anti-gravity rehabilitation treadmill as described in any one of claims 1 to 7.

Images