Walking aid for joint movement medical rehabilitation

Abstract

The invention provides a walking aid for joint movement medical rehabilitation, and relates to the technical field of medical instruments. The walking aid for joint movement medical rehabilitation comprises a back plate, two waist driving mechanisms and two knee driving mechanisms, thigh connecting rods are hinged to the positions, close to the lower ends, of the two sides of the back plate through the waist driving mechanisms, and shank connecting rods are hinged to the lower ends of the two thigh connecting rods through the knee driving mechanisms. A double-shaft motor is used for providing precise assistance, the physical burden of a patient during walking is relieved, the assistance force is intelligently adjusted according to strides and vital signs, filtering and fusion are conducted in combination with data of various sensors, personalized assistance is provided by using a gait recognition algorithm, the rehabilitation requirements of different patients are met, the heart rate, blood pressure and blood oxygen saturation are monitored in real time, and the rehabilitation effect is improved. And the driving mechanism can be adjusted, the legs are supported to be suspended and fixed, and the device is suitable for injury conditions and rehabilitation stages of different patients.

Description

Technical Field

[0001] The present invention relates to the technical field of medical devices, and particularly to a walking aid for joint sports medicine rehabilitation. Background Art

[0002] Medical rehabilitation walking aids are auxiliary devices specifically designed to help people with motor function disorders or disabilities regain their walking ability. They are commonly used during the rehabilitation treatment process, aiming to help patients gradually recover their ability to walk independently by providing support and stability. These walking aids can include simple canes, four-legged crutches, walking frames, wheelchairs, and more complex electric walking devices, etc. Their designs take into account the user's physical condition, rehabilitation needs, and convenience in daily life. Medical rehabilitation walking aids not only play a role in physical rehabilitation but also give patients confidence and a sense of independence psychologically, and are an indispensable part of the rehabilitation treatment.

[0003] During the process of medical rehabilitation walking assistance, the traditional method is for patients to rely on crutches for walking exercises. However, this method requires patients to expend a large amount of physical effort, which not only increases the difficulty of the exercise but may also exacerbate the injury due to over-reliance or improper use. Although there are some electric walking devices on the market, such as wheelchairs, etc., these devices are often designed in a structure similar to a small vehicle and may not be suitable for all patients, especially those who need specific assistance to regain their walking ability. In addition, these devices may not be flexible enough to match the patient's gait and movement needs, so there are limitations in improving the rehabilitation efficiency. Summary of the Invention

[0004] Aiming at the deficiencies of the prior art, the present invention provides a walking aid for joint sports medicine rehabilitation, which solves the problems of large physical consumption and low matching efficiency existing in the existing walking aids.

[0005] To achieve the above purposes, the present invention is realized through the following technical solutions: A walking aid for joint sports medicine rehabilitation, including a back plate, two waist driving mechanisms, and two knee driving mechanisms. At the lower ends of both sides of the back plate, thigh connecting rods are hingedly arranged through the waist driving mechanisms. At the lower ends of the two thigh connecting rods, calf connecting rods are hingedly arranged through the knee driving mechanisms. At the lower ends of the two calf connecting rods, touch ground feet are hingedly arranged through the ankle driving mechanisms. A controller and a storage battery are respectively fixedly arranged at the rear side of the back plate. An upper thigh restraint belt and a lower thigh restraint belt are respectively fixedly connected to the upper part and the lower part of the thigh connecting rod. A vital sign detection group is embedded inside the inner ring of the upper thigh restraint belt;

[0006] The waist drive mechanism includes a fine-tuning shaft seat and a waist dual-axis motor. The waist dual-axis motor is movably arranged inside the fine-tuning shaft seat. Both sides of the fine-tuning shaft seat are respectively fixedly connected with a connecting side plate and a fixed side plate through a plurality of bolts. The lower end of the waist dual-axis motor is fixedly connected with a lower connecting head;

[0007] The knee drive mechanism includes a knee shaft seat. A protective cover plate is fixedly arranged at the upper end of the knee shaft seat. A knee dual-axis motor is movably arranged inside the knee shaft seat. On both sides of the knee shaft seat, a first knee side plate and a second knee side plate are respectively fixedly arranged through a plurality of bolts.

[0008] Preferably, waist restraint belts are fixedly arranged at the centers near both sides of the back plate. The two output ends of the waist dual-axis motor are respectively fixedly connected with the connecting side plate and the fixed side plate. The rear end of the connecting side plate is hinged to the back plate. The lower end of the lower connecting head is fixedly connected with the thigh connecting rod.

[0009] Preferably, a calf restraint belt is fixedly connected to the middle of the calf connecting rod. The vital sign detection group includes a heart rate sensor, a blood pressure sensor, and a blood oxygen saturation sensor. The controller is provided with an initial acceleration sensor and a gyroscope.

[0010] Preferably, the two output ends of the knee dual-axis motor are respectively fixedly connected with the first knee side plate and the second knee side plate. The lower end of the knee dual-axis motor is fixedly provided with a knee connecting head. The lower end of the knee connecting head is fixedly connected with the upper end of the calf connecting rod. The upper end of the knee shaft seat is fixedly connected with the lower end of the thigh connecting rod. The ankle drive mechanism has the same structure as the knee drive mechanism. The dual-axis motors inside the waist drive mechanism, the knee drive mechanism, and the ankle drive mechanism are all equipped with encoders.

[0011] A walking assistance control method for joint sports medicine rehabilitation specifically includes the following steps:

[0012] S1. Motion detection

[0013] Collect the motor rotation data through the encoder on the motor, collect the posture and motion data of the wearer using the acceleration sensor and the gyroscope, perform filtering processing on the collected data to remove noise and outliers, fuse the multi-source sensor data to improve the accuracy and reliability of the data, use the gait recognition algorithm, judge the gait of the wearer according to the encoder data and the sensor data, calculate the stride length, stride frequency and other motion parameters, determine the amplitude and direction of each step through the data of the motor encoder, and use the machine learning models such as support vector machine or neural network to identify different motion patterns;

[0014] S2. Auxiliary decision-making

[0015] Obtain the wearer's basic health data, such as heart rate, blood pressure, and blood oxygen saturation, from the vital sign monitoring unit, and continuously obtain the stride and step frequency motion parameters from the motion detection algorithm module. When a large stride of the wearer is detected, appropriately increase the auxiliary force of the motor to reduce the wearer's burden. When a small stride is detected, appropriately reduce the auxiliary force of the motor to avoid excessive assistance. When the wearer's heart rate is too high or blood pressure is abnormal, increase the auxiliary force of the motor to help the wearer reduce physical exertion. When the wearer's heart rate and blood pressure indicators are normal, maintain the normal auxiliary force. Determine whether the wearer has stopped moving through the sensor and encoder data. If it is detected that the wearer has stopped moving for more than a certain period of time, automatically stop the auxiliary output of the motor;

[0016] S3. Data Management and Analysis

[0017] Store the collected motion data and vital sign data in a local or cloud database, use data analysis algorithms to perform real-time and historical analysis on the stored data, predict the wearer's physical exertion trend through a machine learning model, provide scientific exercise suggestions, evaluate the wearer's health status based on the vital sign data, generate a health report, including heart rate variability and blood pressure change indicators, and continuously monitor the wearer's vital sign data to detect abnormal conditions. When an abnormality is detected, remind the wearer through the voice prompt function and recommend stopping exercise or seeking medical attention.

[0018] Preferably, the calculation methods for stride and step frequency in the motion detection of step S1 are as follows:

[0019]

[0020] Among them, Δencoder_angle is the change in the angle turned by the motor between two steps, encoder_resolution is the resolution of the encoder, expressed as the number of pulses per revolution, and stride_factor is the stride factor, which converts the rotation angle of the encoder into the actual stride;

[0021]

[0022] Among them, peaks_count is the number of detected peaks, and time_window is the length of the time window, in seconds.

[0023] Preferably, the auxiliary decision-making in step S2 includes heart rate adjustment, blood pressure adjustment, blood oxygen saturation adjustment, and motion stop detection. Specifically:

[0024] Calculate the difference between the current heart rate and the median of the normal heart rate range: Heart rate difference = current heart rate - median of the normal heart rate range, and adjust the motor auxiliary force according to the heart rate difference: Auxiliary force = basic auxiliary force + heart rate difference × proportionality coefficient;

[0025] Calculate the difference between the current blood pressure and the median of the normal blood pressure range: Systolic pressure difference = current systolic pressure - median of the normal systolic pressure range, Diastolic pressure difference = current diastolic pressure - median of the normal diastolic pressure range, and adjust the motor assistive force according to the blood pressure difference: Assistive force = basic assistive force + (systolic pressure difference + diastolic pressure difference) × proportionality coefficient;

[0026] Calculate the difference between the current blood oxygen saturation and the median of the normal blood oxygen saturation range: Blood oxygen saturation difference = median of the normal blood oxygen saturation range - current blood oxygen saturation, and adjust the motor assistive force according to the blood oxygen saturation difference: Assistive force = basic assistive force + blood oxygen saturation difference × proportionality coefficient;

[0027] Detect whether the step frequency is lower than a certain threshold: if step frequency < static step frequency threshold and posture change < static posture change threshold.

[0028] The present invention provides a walking aid for joint sports medicine rehabilitation. It has the following beneficial effects:

[0029] The present invention provides a walking aid for joint sports medicine rehabilitation. Through the driving mechanisms at the waist, knee, and ankle, it uses a dual-axis motor to provide precise assistive force, reducing the physical burden on patients during walking exercises. It intelligently adjusts the motor assistive power according to the patient's step length, step frequency, and vital sign data to ensure that patients can exercise effectively with the least physical consumption. The present invention combines data from multiple sensors such as encoders, accelerometers, and gyroscopes, and through filtering processing and data fusion, improves the accuracy and reliability of the detected data. It uses machine learning models such as support vector machines or neural networks to identify different motion patterns and provide more personalized assistance. It is equipped with a heart rate sensor, a blood pressure sensor, and a blood oxygen saturation sensor to monitor the patient's health status in real time and promptly detect abnormal conditions. By detecting the step frequency and posture change, it automatically determines whether the patient has stopped moving and stops the motor's assistive output after a certain period of time to avoid accidents. The driving mechanisms at the waist, knee, and ankle are all adjustable to adapt to the specific needs of different patients. For patients with relatively severe leg injuries, the injured leg can be suspended and fixed by an upper thigh strap, a lower thigh strap, and a calf strap to avoid contact with the ground, reducing pain and further injury. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 is the front orthographic axonometric schematic diagram of the present invention;

[0031] Figure 2 is the rear orthographic axonometric schematic diagram of the present invention;

[0032] Figure 3 is the exploded axonometric schematic diagram of the waist driving mechanism of the present invention;

[0033] Figure 4 Exploded axonometric view of the knee drive mechanism of the present invention.

[0034] Among them, 1. Waist drive mechanism; 2. Knee drive mechanism; 3. Rear backplate; 4. Waist restraint strap; 5. Vital sign detection group; 6. Upper thigh restraint strap; 7. Thigh link; 8. Lower thigh restraint strap; 9. Calf restraint strap; 10. Calf link; 11. Touching foot; 12. Ankle drive mechanism; 13. Controller; 14. Battery; 101. Fine-tuning shaft seat; 102. Connecting side plate; 103. Waist double-axis motor; 104. Fixed side plate; 105. Lower connecting head; 201. Knee shaft seat; 202. Protective cover plate; 203. First knee side plate; 204. Knee connecting head; 205. Knee double-axis motor; 206. Second knee side plate. Specific embodiments

[0035] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0036] Embodiment 1

[0037] As Figures 1-4 shown, the embodiment of the present invention provides a walking aid for arthrokinematic medicine rehabilitation, including a rear backplate 3, two waist drive mechanisms 1, and two knee drive mechanisms 2. Thigh links 7 are hingedly arranged on both sides of the rear backplate 3 near the lower end through the waist drive mechanisms 1. The lower ends of the two thigh links 7 are hingedly arranged with calf links 10 through the knee drive mechanisms 2. The lower ends of the two calf links 10 are hingedly arranged with touching feet 11 through the ankle drive mechanisms 12. A controller 13 and a battery 14 are respectively fixedly arranged on the rear side of the rear backplate 3. An upper thigh restraint strap 6 and a lower thigh restraint strap 8 are respectively fixedly connected to the upper and lower parts of the thigh link 7. A vital sign detection group 5 is embedded inside the inner ring of the upper thigh restraint strap 6;

[0038] The waist drive mechanism 1 includes a fine-tuning shaft seat 101 and a waist double-axis motor 103. The waist double-axis motor 103 is movably arranged inside the fine-tuning shaft seat 101. Both sides of the fine-tuning shaft seat 101 are respectively fixedly connected with a connecting side plate 102 and a fixed side plate 104 through a plurality of bolts. The lower end of the waist double-axis motor 103 is fixedly connected with a lower connecting head 105;

[0039] The knee drive mechanism 2 includes a knee shaft seat 201. A protective cover plate 202 is fixedly arranged at the upper end of the knee shaft seat 201. A knee double-shaft motor 205 is movably arranged inside the knee shaft seat 201. On both sides of the knee shaft seat 201, a first knee side plate 203 and a second knee side plate 206 are respectively fixedly arranged through a plurality of bolts.

[0040] Waist restraint belts 4 are fixedly arranged at the positions near the center on both sides of the back plate 3. The two output ends of the waist double-shaft motor 103 are respectively fixedly connected to the connecting side plate 102 and the fixed side plate 104. The rear end of the connecting side plate 102 is hinged to the back plate 3. The lower end of the lower connecting head 105 is fixedly connected to the thigh connecting rod 7.

[0041] A calf restraint belt 9 is fixedly connected to the middle of the calf connecting rod 10. The vital sign detection group 5 includes a heart rate sensor, a blood pressure sensor, and a blood oxygen saturation sensor. The controller 13 is internally provided with an initial accelerometer and a gyroscope.

[0042] Both output ends of the knee double-shaft motor 205 are respectively fixedly connected to the first knee side plate 203 and the second knee side plate 206. A knee connecting head 204 is fixedly arranged at the lower end of the knee double-shaft motor 205. The lower end of the knee connecting head 204 is fixedly connected to the upper end of the calf connecting rod 10. The upper end of the knee shaft seat 201 is fixedly connected to the lower end of the thigh connecting rod 7. The ankle drive mechanism 12 has the same structure as the knee drive mechanism 2. The double-shaft motors inside the waist drive mechanism 1, the knee drive mechanism 2, and the ankle drive mechanism 12 are all equipped with encoders.

[0043] Specifically:

[0044] When wearing, if the leg is severely injured and difficult to support the ground, then the injured leg is suspended and fixed by the upper thigh restraint belt 6, the lower thigh restraint belt 8, and the calf restraint belt 9. Through the supporting force of the waist, the touching foot 11 is in contact with the ground, and the patient's leg does not contact the touching foot 11, and there is a certain gap between the two. The patient swings the leg to make the device move, realizing walking exercise. If, after exercising for a period of time, the patient can bear force, the patient's foot can contact the touching foot 11, increasing the force on the foot, thereby increasing the intensity of the exercise. The waist drive mechanism 1, the knee drive mechanism 2, and the ankle drive mechanism 12 in the auxiliary power mechanism of this device are all adjusted by using the upward connection of the double-shaft motor. For example, because of the special hinged method of the waist drive mechanism 1, the upward adjustment can only be slightly adjusted. The knee drive mechanism 2 and the ankle drive mechanism 12 respectively provide assistance adjustment for the thigh connecting rod 7 and the calf connecting rod 10. The upper thigh restraint belt 6 is fastened at the root of the patient's thigh for monitoring vital signs.

[0045] Embodiment 2

[0046] A walking assistance control method for joint sports medicine rehabilitation, specifically including the following steps:

[0047] S1. Motion detection

[0048] Collect the motor rotation data through the encoder on the motor, use the accelerometer and gyroscope to collect the posture and motion data of the wearer, filter the collected data to remove noise and outliers, fuse the multi-source sensor data to improve the accuracy and reliability of the data, use the gait recognition algorithm, judge the gait of the wearer according to the encoder data and sensor data, calculate the stride and step frequency motion parameters, determine the amplitude and direction of each step through the data of the motor encoder, and use the machine learning models support vector machine or neural network to identify different motion patterns (such as walking, running, going up and down stairs);

[0049] S2. Auxiliary decision-making

[0050] Obtain the basic health data of the wearer, such as heart rate, blood pressure, and blood oxygen saturation, from the vital sign monitoring unit, continuously obtain the stride and step frequency motion parameters from the motion detection algorithm module. When it is detected that the wearer has a large stride, appropriately increase the auxiliary force of the motor to reduce the burden on the wearer. When it is detected that the stride is small, appropriately reduce the auxiliary force of the motor to avoid over-assistance. When it is monitored that the wearer's heart rate is too high or the blood pressure is abnormal, increase the auxiliary force of the motor to help the wearer reduce physical consumption. When it is monitored that the wearer's heart rate and blood pressure indicators are normal, maintain the normal auxiliary force. Judge whether the wearer stops moving through the sensor and encoder data. If it is detected that the wearer stops moving for more than a certain time (such as 5 seconds), automatically stop the auxiliary output of the motor;

[0051] S3. Data management and analysis

[0052] Store the collected motion data and vital sign data in a local or cloud database, use data analysis algorithms to perform real-time and historical analysis on the stored data, predict the physical consumption trend of the wearer through machine learning models, provide scientific exercise suggestions, evaluate the health status of the wearer according to the vital sign data, generate a health report, including heart rate variability and blood pressure change indicators, and continuously monitor the vital sign data of the wearer to detect abnormal situations. When an abnormality is detected, remind the wearer through the voice prompt function and recommend stopping exercise or seeking medical attention.

[0053] The calculation methods of stride and step frequency in step S1 motion detection are as follows:

[0054]

[0055] Among them, Δencoder_angle is the change in the angle that the motor rotates between two steps, encoder_resolution is the resolution of the encoder, expressed as the number of pulses per revolution, and stride_factor is the stride factor that converts the rotation angle of the encoder into the actual stride;

[0056]

[0057] Among them, peaks_count is the number of detected peaks, and time_window is the length of the time window, in seconds.

[0058] Step S2 auxiliary decision-making includes heart rate adjustment, blood pressure adjustment, blood oxygen saturation adjustment, and exercise stop detection. Specifically:

[0059] Calculate the difference between the current heart rate and the median of the normal heart rate range: Heart rate difference = current heart rate - median of the normal heart rate range. Adjust the motor assistance force according to the heart rate difference: Assistance force = basic assistance force + heart rate difference × proportionality coefficient;

[0060] Calculate the difference between the current blood pressure and the median of the normal blood pressure range: Systolic blood pressure difference = current systolic blood pressure - median of the normal systolic blood pressure range, Diastolic blood pressure difference = current diastolic blood pressure - median of the normal diastolic blood pressure range. Adjust the motor assistance force according to the blood pressure difference: Assistance force = basic assistance force + (systolic blood pressure difference + diastolic blood pressure difference) × proportionality coefficient;

[0061] Calculate the difference between the current blood oxygen saturation and the median of the normal blood oxygen saturation range: Blood oxygen saturation difference = median of the normal blood oxygen saturation range - current blood oxygen saturation. Adjust the motor assistance force according to the blood oxygen saturation difference: Assistance force = basic assistance force + blood oxygen saturation difference × proportionality coefficient;

[0062] Detect whether the step frequency is lower than a certain threshold: if step frequency < static step frequency threshold and posture change < static posture change threshold.

[0063] Illustrate specifically as follows:

[0064] Data acquisition:

[0065] Collect accelerometer data: accel_x, accel_y, accel_z. Collect motor encoder data: encoder_angle, encoder_speed.

[0066] Data preprocessing:

[0067] Filter the acceleration data using a low-pass filter: filtered_accel_x = low_pass_filter(accel_x). Fuse the accelerometer and encoder data using a Kalman filter: fusion_data = kalman_filter(filtered_accel_x, encoder_angle).

[0068] Stride detection:

[0069] Detect the start and end points of the gait cycle through acceleration peaks. Calculate the stride length of each step through encoder data: step_length = encoder_angle_change / encoder_resolution * stride_factor.

[0070] Step frequency detection:

[0071] Calculate the step frequency through the frequency of the acceleration signal: step_frequency = count_peaks(fusion_data) / time_window.

[0072] Motion mode recognition:

[0073] Construct a feature vector: features = [step_length, step_frequency, filtered_accel_x_mean, filtered_accel_x_var, encoder_speed_mean, encoder_speed_var]. Input the trained model for classification: mode = model.predict(features).

[0074] State judgment:

[0075] Perform state judgment according to the recognized mode: If mode is "walking", start the corresponding assist mode. If mode is "stationary", stop the motor assist.

[0076] Code example:

[0077] import numpy as np

[0078] import scipy.signal as signal

[0079] # Assume sensor data

[0080] accel_x = np.array([...]) # Accelerometer x-axis data

[0081] encoder_angle = np.array([...]) # Motor encoder angle data

[0082] # Data preprocessing

[0083] def low_pass_filter(data, cutoff, fs, order = 5):

[0084] nyquist = 0.5 * fs

[0085] normal_cutoff = cutoff / nyquist

[0086] b, a = signal.butter(order, normal_cutoff, btype='low', analog = False)

[0087] filtered_data = signal.filtfilt(b, a, data)

[0088] return filtered_data

[0089] filtered_accel_x = low_pass_filter(accel_x, cutoff = 5, fs = 100)

[0090] # Stride detection

[0091] def detect_step_length(accel_x, encoder_angle, encoder_resolution, stride_factor):

[0092] peaks, _ = signal.find_peaks(filtered_accel_x, height = 0.5)

[0093] # Detect acceleration peaks

[0094] step_lengths = []

[0095] for i in range(len(peaks) - 1):

[0096] angle_change = np.abs(encoder_angle[peaks[i + 1]] - encoder_angle[peaks[i]])

[0097] step_length = angle_change / encoder_resolution * stride_factor

[0098] step_lengths.append(step_length)

[0099] return np.mean(step_lengths)

[0100] average_step_length = detect_step_length(filtered_accel_x, encoder_angle, encoder_resolution = 0.01, stride_factor = 0.5)

[0101] # Step frequency detection

[0102] def detect_step_frequency(accel_x, time_window = 5):

[0103] peaks, _ = signal.find_peaks(filtered_accel_x, height = 0.5)

[0104] step_count = len(peaks)

[0105] step_frequency = step_count / time_window

[0106] return step_frequency

[0107] step_frequency = detect_step_frequency(filtered_accel_x)

[0108] # Motion mode recognition

[0109] from sklearn.ensemble import RandomForestClassifier

[0110] from sklearn.model_selection import train_test_split

[0111] # Assume we already have a trained model

[0112] model = RandomForestClassifier()

[0113] X_train, X_test, y_train, y_test = train_test_split(features, labels, test_size = 0.2, random_state = 42)

[0114] model.fit(X_train, y_train)

[0115] # Real-time data input

[0116] current_features = [average_step_length, step_frequency, np.mean(filtered_accel_x), np.var(filtered_accel_x), np.mean(encoder_angle), np.var(encoder_angle)]

[0117] predicted_mode = model.predict([current_features])[0]

[0118] print(f"Average step length: {average_step_length} meters")

[0119] print(f"Step frequency: {step_frequency} steps per second")

[0120] print(f"Predicted motion mode: {predicted_mode}")

[0121] Although the embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A walking aid for joint sports medical rehabilitation, comprising a back plate (3), two waist drive mechanisms (1), and two knee drive mechanisms (2), characterized in that: The lower ends of both sides of the back plate (3) are hingedly provided with thigh connecting rods (7) through a waist driving mechanism (1), the lower ends of the two thigh connecting rods (7) are hingedly provided with calf connecting rods (10) through a knee driving mechanism (2), the lower ends of the two calf connecting rods (10) are hingedly provided with ground contact feet (11) through an ankle driving mechanism (12), a controller (13) and a storage battery (14) are fixedly provided on the rear side of the back plate (3), an upper thigh restraint belt (6) and a lower thigh restraint belt (8) are fixedly connected to the upper and lower parts of the thigh connecting rods (7), and a vital sign detection group (5) is embedded in the inner ring of the upper thigh restraint belt (6); The waist driving mechanism (1) comprises a fine-tuning shaft seat (101) and a waist dual-axis motor (103); the waist dual-axis motor (103) is movably arranged inside the fine-tuning shaft seat (101); both sides of the fine-tuning shaft seat (101) are respectively fixedly connected to a connecting side plate (102) and a fixed side plate (104) by a plurality of bolts; and the lower end of the waist dual-axis motor (103) is fixedly connected to a lower connecting head (105); The knee drive mechanism (2) comprises a knee shaft seat (201), a protective cover plate (202) is fixedly provided at the upper end of the knee shaft seat (201), a knee dual-axis motor (205) is movably provided inside the knee shaft seat (201), and a first knee side plate (203) and a second knee side plate (206) are respectively fixedly provided on both sides of the knee shaft seat (201) by a plurality of bolts.

2. A walking aid for joint sports medical rehabilitation according to claim 1, characterized in that: Waist restraint belts (4) are fixedly arranged at the center of both sides of the back plate (3); the two output ends of the waist dual-axis motor (103) are fixedly connected to the connecting side plate (102) and the fixed side plate (104) respectively; the rear end of the connecting side plate (102) is hinged to the back plate (3); and the lower end of the lower connecting head (105) is fixedly connected to the thigh connecting rod (7).

3. A walking aid for joint sports medical rehabilitation according to claim 1, characterized in that: The middle part of the calf connecting rod (10) is fixedly connected to a calf restraint belt (9); the vital sign detection group (5) comprises a heart rate sensor, a blood pressure sensor and a blood oxygen saturation sensor; and the controller (13) is provided with an initialization accelerometer and a gyroscope.

4. A walking aid for joint sports medical rehabilitation according to claim 1, characterized in that: The two output ends of the knee dual-axis motor (205) are respectively fixedly connected to the first knee side plate (203) and the second knee side plate (206); the lower end of the knee dual-axis motor (205) is fixedly provided with a knee connector (204); the lower end of the knee connector (204) is fixedly connected to the upper end of the calf connecting rod (10); the upper end of the knee shaft seat (201) is fixedly connected to the lower end of the thigh connecting rod (7); the ankle driving mechanism (12) has the same structure as the knee driving mechanism (2); and the dual-axis motors inside the waist driving mechanism (1), the knee driving mechanism (2) and the ankle driving mechanism (12) all carry encoders.

5. A walking aid control method for joint movement medical rehabilitation according to any one of claims 1 to 4, characterized in that: The specific steps include: S1. Motion detection The encoder on the motor collects the motor rotation data, and the accelerometer and gyroscope are used to collect the wearer's posture and motion data. The collected data is filtered to remove noise and outliers, and the multi-source sensor data is fused to improve the accuracy and reliability of the data. The gait recognition algorithm is used to determine the wearer's gait based on the encoder data and sensor data, and the stride and frequency motion parameters are calculated. The amplitude and direction of each step are determined by the motor encoder data, and the machine learning model support vector machine or neural network is used to identify different motion patterns; S2. Decision support Obtain the wearer's basic health data from the vital signs monitoring unit, such as heart rate, blood pressure, and blood oxygen saturation, and continuously obtain the stride and frequency motion parameters from the motion detection algorithm module. When the wearer's stride is detected to be large, the motor's auxiliary strength is appropriately increased to reduce the wearer's burden. When the stride is detected to be small, the motor's auxiliary strength is appropriately reduced to avoid excessive assistance. When the wearer's heart rate is detected to be too high or blood pressure is abnormal, the motor's auxiliary strength is increased to help the wearer reduce physical exertion. When the wearer's heart rate and blood pressure indicators are normal, the auxiliary strength is maintained at a normal level. The sensor and encoder data are used to determine whether the wearer has stopped exercising. If it is detected that the wearer has stopped exercising for more than a certain period of time, the motor's auxiliary output is automatically stopped. S3. Data Management and Analysis The collected motion data and vital signs data are stored in a local or cloud database. The stored data is analyzed in real time and historically using data analysis algorithms. The wearer's physical exertion trend is predicted through machine learning models, and scientific exercise suggestions are provided. The wearer's health status is evaluated based on the vital signs data, and a health report is generated, including heart rate variability and blood pressure change indicators. The wearer's vital signs data is monitored in real time and abnormal conditions are detected. When an abnormality is detected, the wearer is reminded through the voice prompt function, and it is recommended to stop exercising or seek medical treatment.

6. A walking aid control method for joint movement medical rehabilitation according to claim 5, characterized in that: The calculation method of the stride length and the stride frequency in the motion detection in step S1 is: Where Δencoder_angle is the angle change between two steps, encoder_resolution is the resolution of the encoder, expressed in pulses per revolution, and stride_factor is the step factor, which converts the encoder's rotation angle into the actual step length. Where peaks_count is the number of peaks detected and time_window is the length of the time window in seconds.

7. A walking aid control method for joint sports medical rehabilitation according to claim 5, characterized in that: The auxiliary decision-making in step S2 includes heart rate adjustment, blood pressure adjustment, blood oxygen saturation adjustment, and exercise stop detection, specifically: Calculate the difference between the current heart rate and the median of the normal heart rate range: heart rate difference = current heart rate - median of the normal heart rate range, and adjust the motor assist force according to the heart rate difference: assist force = basic assist force + heart rate difference × proportional coefficient; Calculate the difference between the current blood pressure and the median of the normal blood pressure range: systolic pressure difference = current systolic pressure - median of the normal systolic pressure range, diastolic pressure difference = current diastolic pressure - median of the normal diastolic pressure range, and adjust the motor assist force according to the blood pressure difference: assist force = basic assist force + (systolic pressure difference + diastolic pressure difference) × proportional coefficient; Calculate the difference between the current blood oxygen saturation and the median of the normal blood oxygen saturation range: blood oxygen saturation difference = normal blood oxygen saturation range median - current blood oxygen saturation; adjust the motor auxiliary force according to the blood oxygen saturation difference: auxiliary force = basic auxiliary force + blood oxygen saturation difference × proportional coefficient; Check if the cadence is below a certain threshold: if cadence < static cadence threshold and posture change < static posture change threshold.

Images