Self-adaptive movement method and system based on intelligent rehabilitation device

Abstract

The invention relates to a self-adaptive movement method and system based on an intelligent rehabilitation device, and relates to the field of intelligent rehabilitation devices.The method comprises the steps that biomechanical data are collected; extracting multi-dimensional motion features from the biomechanical data, and combining a preset multi-stage standard motion template library to obtain a current motion state; identifying a current motion phase based on the current motion state; matching a standard motion template of a corresponding stage from a multi-stage standard motion template library according to the current motion stage; combining the current motion state with a standard motion template to obtain a motion parameter deviation risk value; when the motion parameter deviation risk value exceeds a preset risk threshold value, generating a motion adjustment strategy according to the standard motion template and the motion parameter deviation risk value; and performing adaptive motion adjustment on the intelligent rehabilitation device based on the motion adjustment strategy. The method has the effect of adapting to the actual motion states of the user in different stages.

Description

Technical Field

[0001] This invention relates to the field of intelligent rehabilitation devices, and in particular to an adaptive movement method and system based on an intelligent rehabilitation device. Background Technology

[0002] Intelligent rehabilitation devices refer to auxiliary equipment that integrates sensor technology, drive control, and human-computer interaction to help users conduct scientific and effective rehabilitation training.

[0003] Currently, existing rehabilitation devices have basic motion data acquisition capabilities. Some products can collect single-dimensional information such as joint angles and movement speeds from users through sensors, and perform simple monitoring of the movement process based on preset fixed thresholds.

[0004] However, existing devices rely on a single evaluation index when judging movement status, which cannot extract movement characteristics from multiple dimensions. They also do not consider the differences in movement needs at different stages of rehabilitation training. They can only judge the movement process according to a uniform standard, which is difficult to adapt to the actual movement status of users at different stages and needs to be improved. Summary of the Invention

[0005] To adapt to the actual movement status of users at different stages, this invention provides an adaptive movement method and system based on an intelligent rehabilitation device.

[0006] In a first aspect, the present invention provides an adaptive exercise method based on an intelligent rehabilitation device, employing the following technical solution: An adaptive exercise method based on an intelligent rehabilitation device includes: Collect biomechanical data; Multidimensional motion features are extracted from biomechanical data and combined with a pre-set multi-stage standard motion template library to obtain the current motion state; Identify the current motion stage based on the current motion state; Based on the current stage of the movement, the corresponding standard movement template is matched from the multi-stage standard movement template library; Combine the current motion state with a standard motion template to obtain the risk value of deviation from the motion parameters; When the motion parameters deviate from the risk value beyond the preset risk threshold, a motion adjustment strategy is generated based on the standard motion template and the deviation of the motion parameters from the risk value. Based on motion adjustment strategies, adaptive motion regulation is performed on intelligent rehabilitation devices.

[0007] By employing the above technical solution, biomechanical data is first collected to extract multidimensional movement features. Combined with a multi-stage standard movement template library, the user's current movement state and rehabilitation stage are determined. Subsequently, a standard movement template corresponding to the stage is matched, and the deviation risk value of movement parameters is calculated by comparing the current movement state with the template. When the risk value exceeds the limit, a targeted movement adjustment strategy is generated based on the template and the risk value, and the control device executes adaptive adjustment. This approach comprehensively reflects the user's movement state through multidimensional features and adapts to the different needs of different rehabilitation stages through staged template matching, ensuring a better fit for the actual movement state of users at different stages.

[0008] Optionally, it may also include the specific implementation steps for generating the motion adjustment strategy: Collect user physiological indicators and current environmental parameters; A comprehensive feature space is generated by combining user physiological indicators, current environmental parameters, and biomechanical data. Determine the parameter adjustment dimension based on the comprehensive feature space; The feature deviation value is obtained based on the multidimensional motion features and the standard motion features extracted from the standard motion template. The motion adjustment strategy is generated by combining feature deviation values, motion parameter deviation risk values, parameter adjustment dimensions, and a preset evaluation function.

[0009] By employing the aforementioned technical solution, user physiological indicators and current environmental parameters are collected and integrated to construct a comprehensive feature space. Based on this comprehensive feature space, the parameter adjustment dimensions are precisely determined. Then, by comparing multi-dimensional movement features with standard movement features to obtain feature deviation values, and combining these with movement parameter deviation risk values, adjustment dimensions, and preset evaluation functions, a scientific movement adjustment strategy is calculated and generated. This approach avoids the one-sidedness of strategies caused by single data sources through multi-data fusion, and ensures that the strategy aligns with the user's current physiological tolerance through precise adjustment dimension positioning and a scientific evaluation function. Ultimately, it achieves more personalized and safer adaptive adjustment, further adapting to the actual movement needs of users at different stages and in different physical states.

[0010] Optional, also includes: Collect user electromyography signals, heart rate variability, and exercise speed decay rate; Fatigue level values ​​are generated based on user electromyography signals, heart rate variability, and exercise speed decay rate. When the fatigue level value exceeds the preset fatigue level safety threshold, the fatigue exceedance value is obtained based on the fatigue level value and the fatigue level safety threshold. The target adjustment range is determined based on the fatigue exceedance value; The motion adjustment strategy is modified by adjusting the target amplitude to generate a safe motion adjustment strategy; The intelligent rehabilitation device reduces exercise load based on a safe exercise adjustment strategy.

[0011] By employing the aforementioned technical solution, the system first collects the user's electromyography (EMG) signals, heart rate variability, and exercise speed decay rate. Based on the fusion of these three data points, a quantified fatigue level value is calculated to accurately determine whether the user is in a state of excessive fatigue. When the fatigue level value exceeds a safe threshold, the system calculates the excess fatigue value to determine the target adjustment range for reducing the exercise load. This range is then used to modify the original exercise adjustment strategy, generating a safe exercise adjustment strategy that prioritizes user safety. Ultimately, the system controls the intelligent rehabilitation device to perform the load reduction operation. This avoids the risk of training injuries caused by focusing solely on adjusting exercise parameters while neglecting user fatigue, and further adapts to the actual exercise needs of users under different fatigue states.

[0012] Optional, wearable recognition methods may also be included: Collect exercise resistance values ​​and rehabilitation device models; The standard resistance value is obtained based on the model of the rehabilitation device and the current stage of exercise; A standard resistance range is generated based on the standard resistance value and the preset resistance range value; When the resistance value exceeds the standard resistance range, the wearing adjustment is performed using the preset wearing adjustment method.

[0013] By adopting the above technical solution, real-time exercise resistance values ​​and rehabilitation device model are first collected, and the corresponding standard resistance value is retrieved from preset data based on the current exercise stage. Then, a standard resistance range is generated based on the standard resistance value and preset resistance range values, which serves as the basis for judging whether the wearing is appropriate. When the real-time exercise resistance value exceeds this range, the system initiates the wearing adjustment method to adjust the wearing, avoiding the impact of improper wearing on the accuracy of exercise data collection or causing user discomfort, further ensuring the effectiveness and comfort of rehabilitation training, and indirectly improving the accuracy of exercise adjustment strategies.

[0014] Optionally, the wearing adjustment method includes: Combine the motion resistance value with the standard resistance range to generate a resistance difference value; The wearing difference rate is derived based on the resistance difference value; Determine whether the wearing difference rate exceeds the preset adjustment difference rate; When the wearing difference rate exceeds the adjustment difference rate, a prompt to stop using the device should be reported; When the wearing difference rate does not exceed the adjustment difference rate, the intelligent rehabilitation device will automatically adjust itself using a preset automatic adjustment method.

[0015] By adopting the above technical solution, a resistance difference value is first generated by comparing the exercise resistance value with the standard resistance range, and then the wearing difference rate is calculated. Then, based on the adjustment difference rate as a threshold, wearing abnormalities are graded and handled: when the wearing difference rate exceeds the adjustment difference rate, a stop-use prompt is immediately reported to avoid forced training causing damage to the user or aggravating device abnormalities; when the difference rate does not exceed the threshold, the automatic adjustment method is activated to allow the device to correct its wearing status automatically. This not only prevents the automatic adjustment function from operating ineffectively in cases of severe abnormalities, but also achieves automatic optimization of the wearing status within a safe range, reducing the tediousness of manual adjustment by the user and ensuring that the rehabilitation device always participates in training with a suitable fit.

[0016] Optionally, the automatic adjustment method includes: Collect user recovery plans; The subsequent exercise phase is determined based on the user's recovery plan and current exercise phase. The motion parameters of the device are obtained based on the subsequent motion stages; Based on the wearing difference rate and device motion parameters, the fixation device offset parameters are generated; The device adjustment parameters are generated by combining the device offset parameters, wearing difference rate, and device motion parameters. The intelligent rehabilitation device is controlled to continue exercising according to the user's recovery plan, while the preset wearing adjustment device is controlled to operate according to the device adjustment parameters.

[0017] By employing the above technical solution, the user's complete recovery plan is first collected, and the subsequent exercise stages are predicted based on the current exercise stage. The corresponding device motion parameters are then determined according to the needs of the subsequent stages. Based on this, the device offset parameters are generated by combining the real-time calculated wearing difference rate with the device motion parameters for the subsequent stages. Then, the offset parameters, wearing difference rate, and current device motion parameters are integrated to generate device adjustment parameters that balance current wearing correction with adaptation for the subsequent stages. Finally, the rehabilitation device is controlled to continue training according to the recovery plan, while simultaneously driving the wearing adjustment device to operate according to the adjustment parameters. This not only solves the problem of motion resistance deviation caused by current wearing abnormalities but also prepares for the device parameter requirements of the subsequent exercise stages in advance, avoiding the problem of repeated device adjustments during training stages.

[0018] Optionally, the method of operating the adjustment device also includes: The driving amplitude value is obtained by adjusting the device parameters. The device contact strength value is generated based on the driving amplitude value; The airbag inflation volume is determined by measuring the contact strength of the device. Based on the airbag inflation volume, the preset inflation airbag is inflated to establish a physical contact state between the wearing adjustment device and the intelligent rehabilitation device. While maintaining physical contact, the intelligent rehabilitation device is controlled to move according to the user's recovery plan, thereby driving the wearing adjustment device to adjust its fit.

[0019] By adopting the above technical solution, the specific driving amplitude value is first determined based on the device's adjustment parameters. Then, the contact strength value between the device and the human body is calculated based on this amplitude value, and the strength value is converted into the corresponding airbag inflation volume. By quantitatively inflating the airbag, a stable physical contact state is formed between the wearing adjustment device, the intelligent rehabilitation device, and the human body. While maintaining this contact state, the rehabilitation device continues to move according to the user's recovery plan. Utilizing the relative displacement between the device and the human body during movement, the wearing adjustment device naturally completes the wearing adjustment, thus balancing the stability of the adjustment and the user's comfort. At the same time, the adjustment process is integrated into normal training exercises, avoiding interruptions caused by separate shutdowns for adjustment. This ensures the continuity of training while making the adjusted wearing state more closely match the actual exercise needs, further improving the wearing adaptability and user experience of the intelligent rehabilitation device during dynamic training.

[0020] Optional, also includes: Collect user facial expression data and voice feedback information; Generate a user's emotional state assessment value based on user facial expression data and voice feedback information; Combine user emotional state assessment values ​​and fatigue level values ​​to generate a comprehensive user experience index; The training comfort evaluation results are obtained based on comprehensive user experience metrics; The safe exercise adjustment strategy is optimized based on the training comfort assessment results, and the exercise parameters of the intelligent rehabilitation device are adjusted based on the optimized safe exercise adjustment strategy.

[0021] Optional, also includes: An exercise quality score is generated based on training comfort assessment results and risk values ​​of deviation from exercise parameters; Collect user electromyography signal recovery rate, joint range of motion improvement indicators, and user recovery plan; The exercise effect value is generated by combining exercise quality score, electromyographic signal recovery rate and joint range of motion improvement index. An optimized recovery plan is generated based on the exercise effect value and the user's recovery plan, and training is planned according to the optimized recovery plan.

[0022] Secondly, this application provides an adaptive motion system based on an intelligent rehabilitation device, employing the following technical solution: An adaptive motion system based on an intelligent rehabilitation device includes: The data acquisition module is used to collect biomechanical data; A memory for storing a program that implements an adaptive movement method based on an intelligent rehabilitation device; The processor is used to load and execute programs stored in memory.

[0023] In summary, this application includes at least one of the following beneficial technical effects: 1. By first collecting biomechanical data and extracting multidimensional movement features, and combining this with a multi-stage standard movement template library, the system determines the user's current movement status and identifies their rehabilitation stage. Then, it matches the corresponding standard movement template to the stage and calculates the risk value of movement parameter deviation by comparing the current movement status with the template. When the risk value exceeds the limit, a targeted movement adjustment strategy is generated based on the template and the risk value, and the control device performs adaptive adjustment. This approach comprehensively reflects the user's movement status through multidimensional features and adapts to the different needs of different rehabilitation stages through staged template matching, ensuring a better fit for the actual movement status of users at different stages. 2. First, a resistance difference value is generated by comparing the exercise resistance value with the standard resistance range, and then the wearing difference rate is calculated. Then, based on the adjustment difference rate as a threshold, wearing abnormalities are graded and handled: when the wearing difference rate exceeds the adjustment difference rate, a stop-use prompt is immediately reported to avoid forced training causing damage to the user or aggravating device abnormalities; when the difference rate does not exceed the threshold, the automatic adjustment method is activated to allow the device to correct its wearing status automatically. This not only prevents the automatic adjustment function from operating ineffectively in cases of severe abnormalities, but also achieves automatic optimization of the wearing status within a safe range, reducing the tediousness of manual adjustment by the user and ensuring that the rehabilitation device always participates in training with an appropriate fit. 3. By first collecting the user's complete recovery plan, predicting subsequent exercise stages based on the current stage, and determining the corresponding device motion parameters according to the needs of the subsequent stages, the system generates fixed device offset parameters by combining the real-time calculated wearing difference rate with the device motion parameters of the subsequent stages. Then, by integrating the offset parameters, wearing difference rate, and current device motion parameters, device adjustment parameters that balance current wearing correction with adaptation for the subsequent stages are generated. Finally, the rehabilitation device is controlled to continue training according to the recovery plan, while simultaneously driving the wearing adjustment device to operate according to the adjustment parameters. This not only solves the motion resistance deviation caused by current wearing abnormalities but also prepares for the device parameter requirements of subsequent exercise stages in advance, avoiding the problem of repeatedly adjusting the device during training stages. Attached Figure Description

[0024] Figure 1 This is a flowchart of an adaptive exercise method based on an intelligent rehabilitation device; Figure 2 This is a simplified diagram illustrating the inflatable airbag before and after inflation. Figure 3 This is a flowchart of the automatic adjustment method; Figure 4 This is a flowchart illustrating the operation of wearing the adjustment device.

[0025] The parts referred to by the numbers in the above attached diagrams are as follows: 1. Intelligent rehabilitation device; 2. Wearing adjustment device; 3. Inflatable airbag; 4. Turntable assembly. Detailed Implementation

[0026] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0027] Reference Figure 1 and Figure 2 This application discloses an adaptive exercise method based on an intelligent rehabilitation device, comprising the following steps: S10: Collect biomechanical data.

[0028] Biomechanical data refers to quantitative data that reflects the mechanical state and interactions during human movement. Specifically, it includes the following: Joint biomechanics data: such as the interaction torque at the exoskeleton joints (torque sensor), joint range of motion (angle sensor), and angular velocity (angular velocity sensor).

[0029] Foot biomechanical data: Pressure distribution and center of gravity offset of a single foot support device collected by a foot pressure sensor array.

[0030] Trunk balance data: Parameters reflecting the human body's balance state, such as trunk tilt angle and angular velocity, collected by a trunk inertial measurement unit (IMU).

[0031] S11: Extract multidimensional motion features from biomechanical data and combine them with a pre-set multi-stage standard motion template library to obtain the current motion state.

[0032] Multidimensional motion features refer to quantified feature vectors extracted from collected biomechanical data, covering multiple dimensions of motion attributes, used to accurately describe the detailed state of human movement. By preprocessing the biomechanical data (Kalman filtering for noise reduction) in 5-second sliding time windows, 15 statistical features such as mean, variance, and kurtosis are calculated. These are ultimately integrated into a multidimensional feature set including "joint biomechanical features, plantar pressure features, and trunk balance features," thus obtaining the multidimensional motion features. The specific 15 statistical features are pre-defined by those skilled in the art and will not be elaborated upon here.

[0033] A multi-stage standard exercise template library refers to a database of standard exercise parameters categorized by exercise stage, established based on a large amount of exercise data from healthy individuals and clinical standards in rehabilitation medicine. The multi-stage standard exercise template library is pre-established by those skilled in the art and will not be elaborated upon here.

[0034] The current motion state refers to the real-time quantitative description of the user's motion at the current moment. The current motion state can be obtained by matching a multi-stage standard motion template library, which stores different current motion states corresponding to different multi-dimensional motion features.

[0035] S12: Identify the current motion phase based on the current motion state.

[0036] The current exercise phase refers to a time segment in the training process that has clear exercise characteristics, corresponding to the target segment of the natural cycle of human movement.

[0037] The current motion stage can be found by looking up the core features of the current motion state through the preset stage comparison table. The table records the different core features corresponding to different motion stages. The comparison content in the stage comparison table is preset by those skilled in the art to form the different core features corresponding to different motion stages, which will not be elaborated here.

[0038] S13: Match the standard motion template corresponding to the current motion stage from the multi-stage standard motion template library.

[0039] A standard motion template refers to a set of quantitative benchmarks in a multi-stage standard motion template library that corresponds one-to-one with the current motion stage and contains the standard range of key motion parameters for that stage.

[0040] The multi-stage standard motion template library can be used to match the standard motion template corresponding to the current motion stage. The template library also records different standard motion templates corresponding to different current motion stages.

[0041] S14: Combine the current motion state with the standard motion template to obtain the risk value of motion parameter deviation.

[0042] The risk value of deviation of motion parameters refers to the degree to which current motion parameters deviate from the standard and the comprehensive quantitative indicator that may lead to safety risks or deviations in training effects.

[0043] A multi-dimensional feature distance calculation method is employed to calculate the Euclidean distances between the current joint biomechanical features, plantar pressure features, trunk balance features, and corresponding features in the standard template. Then, based on preset weights, these distances are weighted and summed to obtain a comprehensive risk value for deviation from the motion parameters. The specific weights are pre-set by those skilled in the art and will not be elaborated upon here.

[0044] The method for calculating multidimensional feature distance is common knowledge in this field and will not be elaborated here.

[0045] S15: When the motion parameters deviate from the risk value by more than the preset risk threshold, a motion adjustment strategy is generated based on the standard motion template and the deviation of the motion parameters from the risk value.

[0046] A risk threshold refers to a pre-defined safety boundary value. Risk thresholds are set in advance by those skilled in the art and will not be elaborated upon here.

[0047] The motion adjustment strategy refers to the specific execution plan used to correct the motion parameters of the intelligent rehabilitation device 1. It includes adjustment instructions for key parameters such as joint drive intensity, motion trajectory range, and auxiliary force. The purpose is to make the patient's motion state return to the standard motion template, reduce potential risks, and ensure the effectiveness of rehabilitation training.

[0048] By understanding the standard motion template and the risk value of deviation from the motion parameters, a motion adjustment strategy can be obtained. The specific implementation steps of the motion adjustment strategy will be explained in detail in S20 to S24, and will not be repeated here.

[0049] When the exercise parameters deviate from the risk value and exceed the risk threshold, it indicates that the deviation between the current patient's exercise status and the standard exercise template for the corresponding stage has exceeded the critical range of safe training and effective rehabilitation. An exercise adjustment strategy needs to be obtained first for subsequent steps.

[0050] S16: Adaptive motion adjustment of the intelligent rehabilitation device 1 based on motion adjustment strategies.

[0051] Once the exercise adjustment strategy is obtained, the intelligent rehabilitation device 1 needs to be controlled to perform adaptive exercise adjustment using the exercise adjustment strategy, thereby reducing potential risks and ensuring the effectiveness of rehabilitation training.

[0052] It also includes the specific implementation steps for generating motion adjustment strategies: S20: Collect user physiological indicators and current environmental parameters.

[0053] User physiological indicators refer to real-time physiological parameter data that reflect the user's physical function status, including but not limited to indicators that characterize the user's physiological load, such as heart rate, blood oxygen saturation, and skin conductance.

[0054] Wearable devices such as smart wristbands can collect users' physiological indicators.

[0055] Current environmental parameters refer to the external environmental conditions data during the operation of the intelligent rehabilitation device 1, including external factors that affect the training effect, such as treadmill speed, treadmill incline, and ambient temperature.

[0056] The rehabilitation device can collect current environmental parameters through its built-in encoder and environmental sensors.

[0057] S21: Combine user physiological indicators, current environmental parameters, and biomechanical data to generate a comprehensive feature space.

[0058] The comprehensive feature space refers to the unified feature representation space formed by standardizing and fusing features of multi-source heterogeneous data (physiological indicators, environmental parameters, biomechanical data).

[0059] By employing a feature-level fusion method, the data of each dimension are normalized and then combined into a high-dimensional comprehensive feature vector through vector concatenation, thereby constructing a comprehensive feature space.

[0060] S22: Determine the parameter adjustment dimension based on the comprehensive feature space.

[0061] The parameter adjustment dimension refers to the specific direction and type of the motion parameter that needs to be adjusted, such as the adjustment methods of different dimensions such as joint torque adjustment, motion speed adjustment, and range of motion adjustment.

[0062] Principal component analysis (PCA) is used to reduce the dimensionality of the comprehensive feature space, identifying the main feature components that have the greatest impact on the motion state. Based on these main components, the dimensions of the motion parameters that need to be adjusted first are determined. The method for identifying the main feature components is common knowledge in this field and will not be elaborated here.

[0063] S23: Obtain the feature deviation value based on the standard motion features extracted from the multidimensional motion features and the standard motion template.

[0064] Standard motion characteristics refer to the ideal motion characteristic values ​​extracted from the standard motion template, representing the standard motion performance expected to be achieved in the current motion stage.

[0065] Standard motion features can be extracted from standard motion templates. The specific extraction method is common knowledge in this field and will not be elaborated here.

[0066] Feature deviation value refers to the quantitative value of the difference between the current multidimensional motion feature and the standard motion feature in each feature dimension.

[0067] A weighted Euclidean distance method is used to calculate the distance between the current multidimensional motion feature and the standard motion feature in each feature dimension. The weights of each dimension are then combined to obtain a comprehensive feature deviation value. The weights of each dimension are pre-defined by those skilled in the art and will not be elaborated upon here.

[0068] S24: Combine feature deviation values, motion parameter deviation risk values, parameter adjustment dimensions, and preset evaluation functions to generate motion adjustment strategies.

[0069] An evaluation function is a mathematical function used to assess the effectiveness of different adjustment strategies. Its inputs are feature deviation values, motion parameter deviation risk values, and parameter adjustment dimensions, and its output is a strategy score. The evaluation function is pre-defined by those skilled in the art and will not be elaborated upon here.

[0070] By constructing an evaluation function based on multi-objective optimization, which comprehensively considers multiple optimization objectives such as human-machine synchronization, energy consumption efficiency, and risk control, the final motion adjustment strategy is generated by solving the optimal solution of this function under a given parameter adjustment dimension.

[0071] Also includes: S30: Collects user electromyography signals, heart rate variability, and exercise speed decay rate.

[0072] User electromyography (EMG) signals refer to the electrophysiological activity signals of muscles collected by surface electromyography (EMG) sensors.

[0073] Heart rate variability refers to the variation in the differences between consecutive heartbeat cycles.

[0074] The velocity decay rate refers to the gradient of velocity decrease within a specific time window.

[0075] Heart rate variability is obtained by collecting electrocardiogram (ECG) signals using an ECG sensor and calculating the heart rate variability.

[0076] The motion velocity decay rate is obtained by collecting motion velocity data through a joint encoder and calculating its time gradient.

[0077] S31: Generates fatigue level values ​​based on user electromyography signals, heart rate variability, and exercise speed decay rate.

[0078] Fatigue level value is a quantitative indicator that comprehensively reflects the fatigue state of a user's neuromuscular and cardiovascular systems.

[0079] By extracting the weighted combination of the decrease rate of median frequency of electromyography signal, the increase slope of the low-frequency to high-frequency power ratio of heart rate variability, and the rate of decrease of exercise speed, a fuzzy logic system is used to perform multi-feature fusion to output a comprehensive fatigue score, i.e., fatigue level value.

[0080] The method of using fuzzy logic systems for multi-feature fusion is common knowledge in this field and will not be elaborated here.

[0081] S32: When the fatigue level value exceeds the preset fatigue level safety threshold, the fatigue exceedance value is obtained based on the fatigue level value and the fatigue level safety threshold.

[0082] The fatigue level safety threshold refers to the critical value of a user's fatigue level. The fatigue level safety threshold is set in advance by those skilled in the art and will not be elaborated here.

[0083] The fatigue exceedance value refers to the positive deviation between the fatigue level value and the fatigue level safety threshold, and is used to quantify the degree to which the current fatigue state exceeds the safe range.

[0084] The fatigue exceedance value can be obtained by calculating the difference between the fatigue level value and the fatigue level safety threshold.

[0085] When the fatigue level value exceeds the safe threshold for fatigue, it indicates that the user is fatigued. The value exceeding the fatigue threshold needs to be calculated first for subsequent steps.

[0086] S33: Determine the target adjustment range based on the fatigue exceedance value.

[0087] The target adjustment range refers to the amount of adjustment of motion parameters determined based on the degree of fatigue, including specific adjustment parameters such as the percentage increase in support force and the percentage decrease in motion speed.

[0088] The target adjustment range corresponding to the fatigue excess value can be found by using the preset fatigue adjustment comparison table. The table records different target adjustment ranges corresponding to different fatigue excess values. The fatigue adjustment comparison table is generated by a person skilled in the art by collecting a large amount of matching data between the fatigue excess value and the corresponding safe adjustment range of users in the rehabilitation training process, establishing a quantitative relationship model between the two through regression analysis, and generating the comparison table content based on the model calculation results. This will not be elaborated here.

[0089] S34: Modify the motion adjustment strategy by adjusting the target adjustment range to generate a safe motion adjustment strategy.

[0090] The safe exercise adjustment strategy refers to a comprehensive control strategy that combines the original exercise adjustment strategy with fatigue compensation measures, prioritizing the safety of training while maintaining rehabilitation effects.

[0091] By adjusting the parameters of the original motion adjustment strategy and adding a compensation term based on the target adjustment range, the motion trajectory and torque output curve are replanned, generating an optimized strategy that balances training effectiveness and safety, namely a safe motion adjustment strategy.

[0092] S35: Based on the safe movement adjustment strategy, the intelligent rehabilitation device 1 reduces the exercise load.

[0093] Once a safe exercise adjustment strategy is obtained, the intelligent rehabilitation device 1 needs to be controlled to operate according to the safe exercise adjustment strategy, thereby reducing the exercise load.

[0094] It also includes wearable recognition methods: S40: Collect exercise resistance values ​​and rehabilitation device model.

[0095] The resistance value refers to the quantified value of the actual resistance experienced by the actuator of the intelligent rehabilitation device 1 during exercise.

[0096] The motion resistance value is obtained by collecting the motor operating current through a high-precision current sensor installed in the joint drive unit, and calculating the real-time resistance torque in combination with the motor torque constant.

[0097] The model number of the rehabilitation device refers to the specific identification code of the intelligent rehabilitation equipment used, which includes key information such as equipment type, specifications, and hardware configuration.

[0098] The model number of the rehabilitation device is obtained by reading the model identification code pre-stored in the device firmware by the main controller of the rehabilitation device.

[0099] S41: Obtain the standard resistance value based on the model of the rehabilitation device and the current stage of exercise.

[0100] The standard resistance value refers to the ideal resistance reference value for a specific rehabilitation device model under standard working conditions, corresponding to different exercise stages.

[0101] The standard resistance value is obtained by consulting a pre-set standard resistance value reference table, which records the standard resistance values ​​corresponding to different rehabilitation device models at each exercise stage. The standard resistance value reference table was generated by a person skilled in the art by collecting a large amount of resistance characteristic data of different rehabilitation device models at each exercise stage under standard working conditions, establishing a mapping relationship between device model, exercise stage and standard resistance value through statistical analysis methods, and generating the reference table content based on the model calculation results, which will not be elaborated here.

[0102] S42: Generate a standard resistance range based on the standard resistance value and the preset resistance range value.

[0103] The resistance range refers to the allowable deviation range set around the standard resistance value. The resistance range is set in advance by those skilled in the art and will not be elaborated here.

[0104] The standard resistance range refers to the normal fluctuation range of resistance determined by the standard resistance value and the resistance range value. Using the standard resistance value as a benchmark, and incorporating the upper and lower limit deviations of the resistance range value, a standard resistance range containing both the minimum and maximum values ​​is generated.

[0105] S43: When the resistance value exceeds the standard resistance range, adjust the wearing method according to the preset wearing adjustment method.

[0106] The wearing adjustment method refers to the method of adjusting the wearing of the intelligent rehabilitation device 1 when abnormal resistance is detected. The specific wearing adjustment method will be described in detail in S50 to S54, and will not be repeated here.

[0107] When the resistance value exceeds the standard resistance range, it indicates that there is an abnormality in the wearing of the intelligent rehabilitation device 1, and it needs to be adjusted by the wearing adjustment method.

[0108] Wearing adjustment methods include: S50: Combines motion resistance values ​​and standard resistance ranges to generate resistance difference values.

[0109] The resistance difference value refers to the absolute deviation between the motion resistance value and the median of the standard resistance range, and is used to quantify the degree of difference between the current resistance state and the ideal state.

[0110] The resistance difference value can be obtained by calculating the absolute difference between the motion resistance value and the midpoint of the standard resistance range.

[0111] S51: The wearing difference rate is derived based on the resistance difference value.

[0112] Wearing difference rate is a percentage of the resistance difference value relative to the standard resistance range, used to characterize the degree of wearing abnormality.

[0113] Divide the resistance difference value by the standard resistance range (the difference between the maximum and minimum values), and then multiply by 100% to obtain the wearing difference rate.

[0114] S52: Determine whether the wearing difference rate exceeds the preset adjustment difference rate.

[0115] The adjustment difference rate refers to the maximum allowable wearing difference rate threshold for automatic adjustment; manual intervention is required when this threshold is exceeded. The adjustment difference rate is preset by those skilled in the art and will not be elaborated upon here.

[0116] By determining whether the wearing difference rate exceeds the adjustment difference rate, we can determine whether artificial intervention is needed.

[0117] S53: When the wearing difference rate exceeds the adjustment difference rate, report a prompt to stop using the device.

[0118] When the wearing difference rate exceeds the adjustment difference rate, it indicates that the intelligent rehabilitation device 1 cannot complete the wearing adjustment on its own and manual intervention is required. Therefore, a prompt to stop using the device and a prompt for manual intervention adjustment should be reported.

[0119] S54: When the wearing difference rate does not exceed the adjustment difference rate, the intelligent rehabilitation device 1 is controlled to adjust itself using a preset automatic adjustment method.

[0120] The automatic adjustment method refers to the method by which the intelligent rehabilitation device 1 automatically adjusts its wearing status. The specific automatic adjustment method will be explained in detail in subsequent sections S60 to S65, and will not be repeated here.

[0121] When the wearing difference rate does not exceed the adjustment difference rate, it indicates that the intelligent rehabilitation device 1 can complete the wearing adjustment on its own. The intelligent rehabilitation device 1 needs to be controlled to adjust itself using the automatic adjustment method.

[0122] Reference Figure 2 and Figure 3 Automatic adjustment methods include: S60: Collect user recovery plans.

[0123] A user recovery plan is a personalized training program designed for a user, including key parameters such as phased training goals, exercise intensity gradients, and expected recovery periods. User recovery plans are pre-input by those skilled in the art and will not be elaborated upon here.

[0124] S61: Based on the user's recovery plan and the current exercise stage, the subsequent exercise stage is obtained.

[0125] The next training phase refers to the next training phase that should be entered according to the phase sequence of the user's recovery plan after the completion of the current training phase.

[0126] By analyzing the phase sequence in the user's recovery plan, the direct successor phase is determined as the subsequent phase based on the position of the current phase in the sequence.

[0127] S62: Obtain the motion parameters of the device based on the subsequent motion phases.

[0128] The device motion parameters refer to the standard motion control parameters that the intelligent rehabilitation device 1 needs to execute during a specific motion phase, including joint range of motion, motion speed, resistance level, etc.

[0129] The device's motion parameters are obtained by consulting a pre-set device motion parameter lookup table, which records the standard device motion parameters corresponding to each motion stage. The device motion parameter lookup table is generated by a person skilled in the art by collecting a large amount of standard motion parameter data from rehabilitation training, establishing a mapping relationship between motion stages and device motion parameters through kinematic analysis, and generating the lookup table content based on the model calculation results, which will not be elaborated here.

[0130] S63: Generate fixation device offset parameters based on wearing difference rate and device motion parameters.

[0131] The offset parameter of the fastener refers to the amount of physical adjustment, such as the adjustment distance of the Velcro, required to compensate for differences in wearing conditions.

[0132] The offset parameters of the fixation device are obtained by consulting a pre-set offset parameter lookup table. This lookup table records the offset parameters of the fixation device corresponding to different wearing difference rates under various device motion parameters. The offset parameter lookup table was generated by a person skilled in the art by collecting a large amount of optimized parameter data on wearing adjustment, establishing the mapping relationship between wearing difference rate, device motion parameters and fixation device offset parameters through regression analysis, and generating the lookup table content based on the model calculation results, which will not be elaborated here.

[0133] S64: Combine the device offset parameters, wearing difference rate, and device motion parameters to generate device adjustment parameters.

[0134] Device adjustment parameters refer to the specific instruction parameters that control the wearing adjustment device 2 to perform adjustment operations, including adjustment amplitude, speed, duration, etc.

[0135] The device adjustment parameters are obtained by consulting a pre-set adjustment parameter lookup table. This table records the device adjustment parameters corresponding to different combinations of fixation device offset parameters, wearing difference rates, and device motion parameters. The adjustment parameter lookup table was generated by a person skilled in the art by collecting a large amount of optimized parameter data of the device adjustment process, establishing the mapping relationship between fixation device offset parameters, wearing difference rates, device motion parameters, and device adjustment parameters through multivariate regression analysis, and generating the lookup table content based on the model calculation results. Details will not be elaborated here.

[0136] S65: Control the intelligent rehabilitation device 1 to continue exercising according to the user's recovery plan, while controlling the preset wearing adjustment device 2 to operate according to the device adjustment parameters.

[0137] The wearing adjustment device 2 refers to the actuator integrated in the intelligent rehabilitation device 1 for automatically adjusting the wearing status. The wearing adjustment device 2 includes an inflatable airbag 3 and a turntable assembly 4 connected to the inflatable airbag 3. After the inflatable airbag 3 is inflated to establish a physical contact between the wearing adjustment device 2 and the intelligent rehabilitation device 1, the turntable assembly 4 can drive the wearing adjustment device 2 to move in all directions while the intelligent rehabilitation device 1 is moving.

[0138] While controlling the intelligent rehabilitation device 1 to continue exercising according to the user's recovery plan, it is also necessary to control the wearing adjustment device 2 to operate according to the device adjustment parameters in order to complete the wearing adjustment.

[0139] Reference Figure 2 and Figure 4 It also includes the operation method of wearing adjustment device 2: S70: The drive amplitude value is obtained based on the device adjustment parameters.

[0140] The driving amplitude value refers to the physical displacement and rotation angle that the actuator of the wearing adjustment device 2 needs to move.

[0141] The drive amplitude value is obtained by consulting a preset drive amplitude lookup table, which records the drive amplitude values ​​corresponding to different device adjustment parameters. The drive amplitude lookup table is generated by those skilled in the art by collecting a large amount of motion data of drive mechanisms, establishing a mapping relationship between device adjustment parameters and drive amplitude values ​​through kinematic analysis, and generating the lookup table content based on the model calculation results, which will not be elaborated here.

[0142] S71: Generate the contact strength value of the device based on the driving amplitude value.

[0143] The contact strength value refers to the contact pressure strength that needs to be established between the wearing adjustment device 2 and the intelligent rehabilitation device 1 to ensure stable contact during the adjustment process.

[0144] The contact strength value of the device is obtained by consulting a preset contact strength reference table, which records the contact strength values ​​of the device corresponding to different drive amplitude values. The contact strength reference table is generated by a person skilled in the art by collecting a large amount of contact strength optimization data, establishing a mapping relationship between drive amplitude values ​​and device contact strength values ​​through mechanical analysis, and generating the reference table content based on the model calculation results, which will not be elaborated here.

[0145] S72: The airbag inflation volume is determined by the contact strength value of the device.

[0146] The airbag inflation volume refers to the volume of gas injected into the airbag 3 to achieve the contact strength of a specific device.

[0147] The airbag inflation volume is obtained by consulting a pre-set inflation reference table, which records the airbag inflation volume corresponding to different device contact strength values. The inflation reference table was generated by a person skilled in the art by collecting a large amount of airbag inflation experimental data, establishing a mapping relationship between device contact strength values ​​and airbag inflation volume through pneumatic pressure analysis, and generating the reference table content based on the model calculation results, which will not be elaborated here.

[0148] S73: Based on the airbag inflation volume, inflate the preset airbag 3 to establish a physical contact state between the wearing adjustment device 2 and the intelligent rehabilitation device 1.

[0149] The inflatable airbag 3 refers to a flexible, inflatable container placed between the wearing adjustment device 2 and the intelligent rehabilitation device 1. It is made of medical-grade silicone material and generates a resistance force through inflation.

[0150] After obtaining the airbag inflation volume, the inflation device preset in the wearing adjustment device 2 needs to be controlled to inflate the airbag 3 so that the wearing adjustment device 2 and the intelligent rehabilitation device 1 can establish a physical contact state for subsequent steps.

[0151] S74: While maintaining physical contact, control the intelligent rehabilitation device 1 to move according to the user's recovery plan, thereby driving the wearing adjustment device 2 to adjust the wearing.

[0152] While maintaining physical contact, the intelligent rehabilitation device 1 can be controlled to move according to the user's recovery plan, thereby driving the wearing adjustment device 2 to adjust its wearing position.

[0153] Also includes: S80: Collects user facial expression data and voice feedback information.

[0154] User facial expression data refers to the facial muscle movement feature data collected by visual sensors.

[0155] Voice feedback information refers to the user's voice content and acoustic feature data acquired through audio acquisition devices.

[0156] S81: Generate an assessment value of the user's emotional state based on the user's facial expression data and voice feedback information.

[0157] User emotional state assessment refers to a quantitative index of emotional state obtained by combining facial expression features and voice features, and is represented using a three-dimensional emotion model. The three-dimensional emotion model is common knowledge in this field and will not be elaborated upon here.

[0158] By inputting facial motion unit features into a pre-trained expression recognition neural network and simultaneously inputting speech acoustic features into a speech emotion analysis model, a decision-level fusion algorithm is used to generate the final emotion state coordinates, i.e., the user's emotional state assessment value.

[0159] The facial expression recognition neural network and the speech emotion analysis model were established in advance by those skilled in the art and will not be described in detail here.

[0160] Decision-level fusion algorithms are common knowledge in this field and will not be elaborated upon here.

[0161] S82: Combine user emotional state assessment values ​​and fatigue level values ​​to generate comprehensive user experience metrics.

[0162] The comprehensive user experience index refers to an overall experience score that takes into account the user's emotional state and physical fatigue level, and is used to evaluate the comfort and sustainability of the training process.

[0163] The comprehensive user experience index is obtained by querying a pre-defined experience index lookup table. This table records the comprehensive user experience index corresponding to different combinations of user emotional state assessment values ​​and fatigue levels. The experience index lookup table was created by experts who collected a large amount of user experience assessment data, established the mapping relationship between user emotional state assessment values, fatigue levels, and comprehensive user experience indexes through multiple regression analysis, and generated the lookup table content based on the model calculation results; details are omitted here. Based on the input user emotional state assessment values ​​and fatigue levels, the corresponding comprehensive user experience index is obtained by looking up the table.

[0164] S83: Training comfort assessment results are obtained based on comprehensive user experience metrics.

[0165] Training comfort assessment results refer to the classification and evaluation results of the user experience comfort level during the current training process.

[0166] The training comfort assessment results are obtained by querying a pre-set comfort level reference table, which records the comfort levels corresponding to different comprehensive user experience indicators. The comfort level reference table was created by a person skilled in the art by collecting a large amount of user experience assessment data, establishing a mapping relationship between comprehensive user experience indicators and comfort levels through cluster analysis, and generating the reference table content based on the model calculation results; details will not be elaborated here.

[0167] S84: Optimize the safe movement adjustment strategy based on the training comfort assessment results, and adjust the movement parameters of the intelligent rehabilitation device 1 based on the optimized safe movement adjustment strategy.

[0168] The strategy optimization parameters are obtained by querying a pre-defined strategy optimization reference table, which records the strategy adjustment coefficients corresponding to different training comfort assessment results. The strategy optimization reference table was created by experts who collected extensive data on strategy optimization effects, established a mapping relationship between training comfort assessment results and strategy optimization parameters through regression analysis, and generated the table content based on model calculations; details are omitted here. The optimization parameters are then applied to a safe movement adjustment strategy to dynamically adjust the joint torque output curve and movement speed parameters.

[0169] Also includes: S90: Generates a motion quality score based on training comfort assessment results and motion parameter deviation risk values.

[0170] The exercise quality score is a quantitative evaluation index that integrates two dimensions: training comfort and exercise safety.

[0171] The exercise quality score is obtained by querying a pre-defined exercise score lookup table. This lookup table records the exercise quality scores corresponding to different combinations of training comfort assessment results and exercise parameter deviation risk values. The exercise score lookup table was generated by a person skilled in the art by collecting a large amount of training quality assessment data, establishing a mapping relationship between training comfort assessment results, exercise parameter deviation risk values, and exercise quality scores through multiple regression analysis, and generating the lookup table content based on the model calculation results. Details are omitted here.

[0172] S91: Collects user electromyography signal recovery rate, joint range of motion improvement indicators, and user recovery plan.

[0173] User electromyography (EMG) signal recovery rate refers to the proportion of EMG signal characteristic parameters that recover to the resting state within a specific time window after training, reflecting the recovery status of the neuromuscular system. The specific time window is set in advance by those skilled in the art and will not be elaborated here.

[0174] The recovery rate of user electromyography (EMG) signals is obtained by collecting surface EMG signals of the target muscle group within a specific time window after training (collected by a surface EMG sensor) and calculating the recovery ratio of the median frequency to the baseline value before training.

[0175] The joint range of motion improvement index refers to the relative improvement in joint range of motion in the current training cycle compared to the previous cycle.

[0176] The joint range of motion improvement index is calculated by measuring the maximum range of motion of the major joints in the current training cycle and comparing it with the recorded value in the previous cycle to calculate the percentage improvement.

[0177] The maximum range of motion is obtained by a high-precision angle sensor installed at the joint of the rehabilitation device.

[0178] The previous cycle's recorded value is obtained by retrieving the user's joint range of motion data from the training record database in the previous training cycle, extracting the verified maximum range of motion angle as a baseline reference value. The training record database adopts a time-series storage structure, indexed by user identifier and training time to ensure the accuracy and timeliness of data retrieval.

[0179] The method for collecting user recovery plans is the same as that for S60 above, and will not be repeated here.

[0180] S92: Combines exercise quality score, electromyographic signal recovery rate and joint range of motion improvement index to generate exercise effect value.

[0181] The exercise effect value refers to the overall training effect evaluation index that takes into account the quality of exercise, physiological recovery and functional improvement.

[0182] The exercise effect value is obtained by querying a pre-set exercise effect comparison table. This table records the exercise effect values ​​corresponding to different combinations of exercise quality scores, electromyographic signal recovery rates, and joint range of motion improvement indicators. The exercise effect comparison table was generated by a person skilled in the art by collecting a large amount of training effect evaluation data, establishing the mapping relationship between exercise quality scores, electromyographic signal recovery rates, joint range of motion improvement indicators, and exercise effect values ​​through principal component analysis, and generating the comparison table content based on the model calculation results, which will not be elaborated here.

[0183] S93: Generates an optimized recovery plan based on exercise effect values ​​and user recovery plans, and performs training planning based on the optimized recovery plan.

[0184] An optimized recovery plan refers to a personalized rehabilitation training program that is dynamically adjusted based on the actual training results.

[0185] The recovery plan adjustment parameters are first retrieved from a preset plan optimization comparison table, and then applied to the original user recovery plan to generate an optimized recovery plan that includes updated training intensity, frequency, and content.

[0186] The recovery plan optimization comparison table records the adjustment parameters of the recovery plan corresponding to different exercise effect values. The recovery plan optimization comparison table was generated by a person skilled in the art by collecting a large amount of plan optimization data, establishing a mapping relationship between exercise effect values ​​and recovery plan adjustment parameters through machine learning algorithms, and generating the comparison table content based on the model calculation results, which will not be elaborated here.

[0187] Once the optimized recovery plan is obtained, it needs to replace the original user recovery plan as the new user recovery plan, and training should be conducted according to the new user recovery plan.

[0188] Based on the same inventive concept, embodiments of the present invention provide an adaptive motion system based on an intelligent rehabilitation device, comprising: The data acquisition module is used to collect biomechanical data, user physiological indicators, current environmental parameters, user electromyography signals, heart rate variability, exercise speed decay rate, exercise resistance value, rehabilitation device model, user recovery plan, user facial expression data, voice feedback information, user electromyography signal recovery rate, and joint range of motion improvement indicators. A memory for storing a program that implements an adaptive movement method based on an intelligent rehabilitation device; The processor is used to load and execute programs stored in memory.

[0189] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0190] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for adaptive motion based on a smart rehabilitation device, characterized in that, Comprise: Collecting biomechanical data; Extracting multi-dimensional motion features from the biomechanical data, and combining a pre-set multi-stage standard motion template library to obtain a current motion state; Identifying a current motion stage based on the current motion state; Matching a standard motion template of a corresponding stage from the multi-stage standard motion template library according to the current motion stage; Combining the current motion state and the standard motion template to obtain a motion parameter deviation risk value; When the motion parameter deviation risk value exceeds a pre-set risk threshold, generating a motion adjustment strategy according to the standard motion template and the motion parameter deviation risk value; Based on the motion adjustment strategy, adaptively adjusting the motion of the intelligent rehabilitation device (1).

2. The self-adaptive motion method based on the intelligent rehabilitation device according to claim 1, characterized in that, Further comprising specific implementation steps for generating the motion adjustment strategy: Collecting user physiological indicators and current environmental parameters; Combining the user physiological indicators, the current environmental parameters, and the biomechanical data to generate a comprehensive feature space; Based on the comprehensive feature space, determining a parameter adjustment dimension; According to the multi-dimensional motion features and the standard motion features extracted from the standard motion template, obtaining a feature deviation value; Combining the feature deviation value, the motion parameter deviation risk value, the parameter adjustment dimension, and a pre-set evaluation function to generate the motion adjustment strategy.

3. The adaptive motion method based on the intelligent rehabilitation device according to claim 2, characterized in that, Further comprising: Collecting user electromyographic signals, heart rate variability, and motion speed attenuation rate; Based on the user electromyographic signals, heart rate variability, and motion speed attenuation rate, generating a fatigue degree value; When the fatigue degree value exceeds a pre-set fatigue degree safety threshold, obtaining a fatigue exceeding value according to the fatigue degree value and the fatigue degree safety threshold; Based on the fatigue exceeding value, determining a target adjustment amplitude; Correcting the motion adjustment strategy by the target adjustment amplitude to generate a safe motion adjustment strategy; Based on the safe motion adjustment strategy, controlling the intelligent rehabilitation device (1) to reduce the motion load.

4. The adaptive motion method based on the intelligent rehabilitation device according to claim 1, characterized in that, Further comprising a wearing recognition method: Collecting motion resistance values and rehabilitation device models; According to the rehabilitation device model and the current motion stage, obtaining a standard resistance value; Based on the standard resistance value and a pre-set resistance interval value, generating a standard resistance interval; When the motion resistance value exceeds the standard resistance interval, performing wearing adjustment by a pre-set wearing adjustment method.

5. The adaptive motion method based on the intelligent rehabilitation device according to claim 4, characterized in that, The wearing adjustment method comprises: Combining the motion resistance value and the standard resistance interval to generate a resistance difference value; Based on the resistance difference value, obtaining a wearing difference rate; Determining whether the wearing difference rate exceeds a pre-set adjustment difference rate; When the wearing difference rate exceeds the adjustment difference rate, reporting a stop using device prompt; When the wearing difference rate does not exceed the adjustment difference rate, controlling the intelligent rehabilitation device (1) to adjust itself by a pre-set automatic adjustment method.

6. The self-adaptive motion method based on the intelligent rehabilitation device according to claim 5, characterized in that, The automatic adjustment method comprises: Collecting a user recovery plan; Based on the user recovery plan and the current motion stage, obtaining a subsequent motion stage; According to the subsequent motion stage, obtaining a device motion parameter; Based on the wearing difference rate and the device motion parameter, generating a fixed device offset parameter; Combining the fixed device offset parameter, the wearing difference rate, and the device motion parameter to generate a device adjustment parameter; Controlling the intelligent rehabilitation device (1) to continue to move according to the user recovery plan, and controlling a pre-set wearing adjustment device (2) to operate according to the device adjustment parameter.

7. The adaptive motion method based on the intelligent rehabilitation device according to claim 6, characterized in that, The application also provides a method for operating the adjustment device (2): obtaining a driving amplitude value based on the device adjustment parameter; generating a device abutting strength value according to the driving amplitude value; obtaining an air bag inflation amount through the device abutting strength value; inflating the preset air bag (3) based on the air bag inflation amount, so that the adjustment device (2) and the intelligent rehabilitation device (1) establish a physical abutting state; maintaining the physical abutting state, and controlling the intelligent rehabilitation device (1) to move according to the user recovery plan, so as to drive the adjustment device (2) to perform wearing adjustment.

8. The adaptive motion method based on the intelligent rehabilitation device according to claim 3, characterized in that, The application also provides: collecting user facial expression data and voice feedback information; generating a user emotional state evaluation value based on the user facial expression data and the voice feedback information; combining the user emotional state evaluation value and the fatigue degree value to generate a comprehensive user experience index; obtaining a training comfort evaluation result based on the comprehensive user experience index; optimizing the safe movement adjustment strategy through the training comfort evaluation result, and adjusting the movement parameters of the intelligent rehabilitation device (1) based on the optimized safe movement adjustment strategy.

9. The adaptive motion method based on the intelligent rehabilitation device according to claim 8, characterized in that, The application also provides: generating a movement quality score based on the training comfort evaluation result and the movement parameter deviation risk value; collecting a user electromyographic signal recovery rate, a joint activity improvement index, and a user recovery plan; combining the movement quality score, the electromyographic signal recovery rate, and the joint activity improvement index to generate a movement effect value; generating an optimized recovery plan based on the movement effect value and the user recovery plan, and planning the training according to the optimized recovery plan.

10. An adaptive motion system based on a smart rehabilitation device, characterized in that, The application provides: a collecting module for collecting biomechanical data; a memory for storing a program for implementing the self-adaptive movement method based on the intelligent rehabilitation device according to any one of claims 1 to 9; a processor for loading and executing the program stored in the memory.

Images