Method and system for assessing a motor state

Abstract

The application provides a motion state evaluation method and system, the method is applied to the cloud, the cloud establishes direct communication connection or indirect communication connection through the user terminal with the intelligent wearable device, the method comprises the following steps: receiving physiological data and head motion data of a target object in a motion process collected and provided by the intelligent wearable device, the head motion data at least comprising head motion posture data and absolute orientation data; fusing the head motion posture data and the absolute orientation data to generate head motion feature sequence; coupling and analyzing the change sequence of the physiological data and the head motion feature sequence to generate a motion state evaluation result of the target object and send the result to the intelligent wearable device or the user terminal through the communication connection. The application collects physiological data and head motion data through a single wearable device, analyzes the dynamic correlation between physiology and motion through time sequence coupling, can ensure the isogenic and synchronous collection of signals, and effectively improves the accuracy of motion state evaluation.

Description

Technical Field

[0001] This application relates to the field of motion analysis technology, and in particular to a method and system for assessing motion status. Background Technology

[0002] Smart wearable devices refer to a general term for wearable devices that utilize wearable technology to intelligently design and develop everyday wearables, including various forms such as headphones, watches, bracelets, and glasses. These devices typically have functions such as health monitoring and activity tracking, and are widely used in fields such as sports and health management.

[0003] With the gradual development of sports and health management, sports performance analysis and health monitoring are receiving increasing attention. Traditional sports monitoring devices typically only collect users' routine physiological data (such as heart rate, steps, etc.) for daily physiological monitoring, which is insufficient to meet users' needs for in-depth analysis of their exercise status.

[0004] In existing motion monitoring technologies, some solutions collect physiological and motion data separately using multiple devices (such as a smartwatch with an external camera), and then analyze the data after aligning them by timestamps. However, such solutions have the following problems: reliance on multiple devices leads to complexity in use; the analysis of physiological and motion data often remains at the level of simple numerical comparison, failing to reflect the intrinsic correlation between the two; and the application scenarios are limited, making it impossible to complete high-quality data collection with a single device, thus restricting the reliability of subsequent data analysis. Summary of the Invention

[0005] In view of the above problems, embodiments of this application provide a motion state evaluation method and system that overcomes or at least partially solves the above problems.

[0006] In a first aspect, embodiments of this application provide a motion state assessment method applied in a cloud environment. The cloud environment establishes a communication connection with a smart wearable device. This communication connection can be a direct connection between the smart wearable device and the cloud environment or an indirect connection established via a user terminal. The method includes: The system receives physiological data and head movement data of the target object during its movement process, which are collected and provided by the smart wearable device. The head movement data includes at least head movement posture data and absolute orientation data. The system then fuses the head movement posture data with the absolute orientation data to generate a head movement feature sequence of the target object. The change sequence of the physiological data is coupled with the head motion feature sequence for analysis to generate an assessment result of the motion state of the target object. The motion state assessment results are sent to the smart wearable device or the user terminal via the communication connection.

[0007] Secondly, embodiments of this application provide a motion state assessment system, including a cloud, a smart wearable device, and a user terminal. The smart wearable device establishes a communication connection with the cloud, which can be a direct connection or an indirect connection established through the user terminal. The smart wearable device collects physiological data and head movement data during the movement of the target object, and provides the collected physiological data and head movement data directly or through the user terminal to the cloud. The head movement data includes at least head movement posture data and absolute orientation data. The cloud receives the physiological data and head movement data provided by the smart wearable device and executes the motion state assessment method as described in the first aspect.

[0008] Thirdly, embodiments of this application provide an electronic device, including a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the motion state evaluation method as described in the first aspect above.

[0009] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the motion state evaluation method as described in the first aspect above.

[0010] The technical solution of this application embodiment can directly obtain high-precision time-aligned data by receiving physiological data and head movement data of the target object during the movement process provided by the smart wearable device, without the need for data synchronization between devices. This avoids the latency and errors introduced by multi-dimensional data reception and provides high-quality input data for subsequent processing. Furthermore, by receiving multi-dimensional data collected by a single smart wearable device, the ease of use in different scenarios can be improved without the need for additional auxiliary equipment, simplifying the structure and reducing usage costs.

[0011] After acquiring head motion data, the head motion posture data and absolute orientation data are fused to generate a head motion feature sequence indicating the dynamic changes of the head during movement. The physiological data change sequence is coupled with the head motion feature sequence for analysis, quantitatively assessing the temporal synchronization relationship between physiological indicators and head movements, and identifying changes in movement state. This fully utilizes the powerful computing and storage resources of the cloud to perform deep coupling analysis of the physiological data and head motion data of the target object, capturing changes that are difficult to reflect by a single indicator from the perspective of physiological and motor coordination, achieving a refined and accurate assessment of movement state, and improving the accuracy and reliability of movement state assessment. Moreover, since the complex analysis tasks are carried out in the cloud, smart wearable devices can focus on data collection, which helps to achieve device lightweighting and extend battery life.

[0012] After obtaining motion status assessment results based on coupling analysis, the results are sent to smart wearable devices or user terminals, which can promptly provide motion status information to the target object, making it easier for the target object to adjust the intensity of exercise and / or prevent exercise risks. Attached Figure Description

[0013] Figure 1 A schematic diagram of the motion state assessment method provided in an embodiment of this application is shown; Figure 2 A schematic diagram illustrating the method for generating head motion feature sequences provided in an embodiment of this application is shown. Figure 3 A schematic diagram illustrating the method for generating motion state evaluation results provided in an embodiment of this application is shown. Figure 4 This illustration shows a method for generating and feeding back motion status reminder information in the cloud, as provided in an embodiment of this application. Figure 5 This application provides a schematic diagram illustrating a method for generating and providing single-move suggestion information in the cloud, as illustrated in an embodiment of the present application. Figure 6 This application provides a schematic diagram illustrating a method for generating and providing personalized exercise and health recommendations in the cloud, as illustrated in an embodiment of the present application. Figure 7 A schematic diagram of the motion state assessment system provided in an embodiment of this application is shown; Figure 8 A schematic diagram of the electronic device structure provided in an embodiment of this application is shown. Detailed Implementation

[0014] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0015] It should be understood that the phrase "one embodiment" or "an embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of this application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Multiple embodiments in this application may include two or more.

[0016] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0017] The motion state evaluation method provided in this application will be described in detail below with reference to the accompanying drawings, through specific embodiments and application scenarios.

[0018] The motion state assessment method provided in this application is applied to the cloud. The cloud establishes a communication connection with the smart wearable device. The communication connection between the two is either a direct connection between the smart wearable device and the cloud, or an indirect connection between the smart wearable device and the cloud via a user terminal.

[0019] For cases where a direct communication connection is established between the cloud and the smart wearable device, the smart wearable device has a built-in WIFI module or cellular communication module, which can communicate directly with the cloud; for cases where an indirect communication connection is established between the cloud and the smart wearable device, the smart wearable device establishes a connection with the user terminal through short-range communication methods such as Bluetooth, and the user terminal communicates with the cloud, thereby forming a communication link between the smart wearable device and the cloud.

[0020] Figure 1 A schematic diagram of the motion state assessment method provided in the embodiments of this application is shown, such as... Figure 1 As shown, the method includes the following steps: Step 101: Receive physiological data and head movement data of the target object during its movement process collected and provided by the smart wearable device. The head movement data includes at least head movement posture data and absolute orientation data.

[0021] The smart wearable device in this embodiment is worn on the head of the target object (e.g., headphones). Wearing on the head means that the device maintains a relatively fixed position relative to the target object's head and can follow head movements. The wearing position includes, but is not limited to, the ear, temple, and other head areas. The smart wearable device has built-in multiple sensing components for simultaneously collecting physiological data and head movement data during the target object's movement. Because the smart wearable device can follow head movements, it can capture the target object's head movement data. In common sports (such as running and cycling), head movement states (such as amplitude of swaying and changes in orientation) are closely related to the overall movement state of the human body (such as movement rhythm and body stability); for example, when a user experiences increased fatigue during a long run, head swaying increases. Therefore, this embodiment uses head movement data as an example to analyze the target object's movement state. All subsequent mentions of movement data refer to head movement data collected by the smart wearable device.

[0022] The various sensing components in the smart wearable device can automatically activate when the target object moves to collect physiological and head movement data during the movement; they can also be activated by the target object via voice commands or manually activated in the user terminal's application. During the target object's non-movement phase, the sensing components can also activate based on commands to collect data, but the data collected during non-movement phases is primarily used for daily health monitoring or physiological state reference and is not related to the motion state assessment scenario focused on in this application's embodiments. Unless otherwise specified, the physiological data and head movement data in this application's embodiments refer to data collected during the target object's movement.

[0023] The smart wearable device has a built-in heart rate sensor to collect heart rate data; if the smart wearable device has a built-in respiration sensor and / or electromyography (EMG) sensor, it collects respiration data and / or EMG data. The smart wearable device can collect head motion posture data of the target object through a built-in 6-axis sensor (such as a 6-axis IMU, i.e., an inertial measurement unit), and absolute orientation data through a 3-axis geomagnetic sensor (i.e., a magnetometer); wherein, the head motion posture data includes, for example, at least angular velocity and / or acceleration, and the absolute orientation data indicates the absolute orientation of the head relative to a preset geographical direction (such as the geographic North Pole).

[0024] By deploying multiple sensing components within a smart wearable device, all data can be collected synchronously from the same device during the data acquisition process. This ensures high-precision time-series alignment of the data and avoids latency and errors introduced by data transmission between multiple devices. Furthermore, the ability to collect multi-dimensional data from a single smart wearable device eliminates the need for additional auxiliary equipment, effectively simplifying the structure, reducing costs, and improving ease of use in various scenarios.

[0025] Specifically, since the cloud and smart wearable devices communicate directly or indirectly, the process of receiving physiological data and head movement data from the smart wearable devices is as follows: Based on a direct connection with smart wearable devices, it receives physiological data and head movement data collected and provided by the smart wearable devices; or, Based on the indirect connection between the user terminal and the smart wearable device, the device receives physiological data and head movement data collected by the smart wearable device and forwarded by the user terminal. The smart wearable device then interacts with the cloud through an application on the user terminal.

[0026] If smart wearable devices are directly connected to the cloud, after the smart wearable devices collect physiological data and head movement data, the cloud can receive the physiological data and head movement data directly provided by the smart wearable devices, which can reduce data transfer and reduce data transmission latency.

[0027] If a smart wearable device is indirectly connected to the cloud, after collecting physiological and head movement data, it sends the collected data to the user terminal. The application installed on the user terminal receives the data and forwards it to the cloud, thus achieving data transmission based on the communication link between the smart wearable device, the user terminal, and the cloud. This data transmission method, based on the user terminal's communication capabilities and a stable network connection, ensures reliable data transmission while reducing the communication power consumption and hardware costs of the smart wearable device itself.

[0028] Step 102: Fuse the head motion posture data with the absolute orientation data to generate a head motion feature sequence of the target object. The received head motion posture data and absolute orientation data are fused. Since both are collected synchronously by a smart wearable device, the timing can be automatically aligned during fusion, eliminating the need for additional timestamp correction, and the alignment error can be controlled at an extremely low level (e.g., less than 1 millisecond). After fusion, multidimensional motion data representing the head motion state is obtained.

[0029] This multidimensional motion data corresponds to the continuous motion process of the target object over a period of time. By extracting features from it, a head motion feature sequence of the target object is generated. The head motion feature sequence indicates the dynamic changes of the head during the motion process. By generating the head motion feature sequence, the dynamic information of the head in three-dimensional space can be structurally represented in time series, thus providing quantitative input for subsequent coupling analysis.

[0030] Step 103: Perform coupling analysis on the physiological data change sequence and head motion feature sequence to generate motion state assessment results for the target object.

[0031] The physiological data change sequence is derived from the changes in physiological data over time, indicating the dynamic changes of physiological indicators such as heart rate and respiration in the target subject during exercise. During exercise, the dynamic changes of physiological indicators over time are synchronized with the changes in head movement dynamics over time (quantified through head movement feature sequences). When the exercise state is good, the two are highly synchronized. For example, the slight head sway caused by the landing of feet while running is regularly synchronized with the corresponding fluctuations in the heart rate waveform, indicating that their rhythms are highly consistent. When fatigue accumulates or control ability declines, the synchronization relationship between the two will show a detectable weakening or shift. For example, head swaying intensifies due to fatigue, but the rhythm of heart rate changes no longer maintains regular synchronization, and the temporal alignment between the two weakens or shifts.

[0032] By coupling the sequence of physiological data changes with the sequence of head movement features, this synchronization relationship can be quantitatively assessed, thereby enabling the identification of subtle changes in movement status at the physiological-motor coordination level. Compared with relying on a single indicator, this can effectively improve the accuracy and reliability of movement status assessment.

[0033] The cloud-based system includes a data analysis engine, which is a functional unit that performs data analysis and processing. The data analysis engine can execute steps 102 and 103 above. Its data processing is equivalent to the cloud-based data processing.

[0034] Step 104: Send the motion status assessment results to the smart wearable device or user terminal via the communication connection.

[0035] After coupling the physiological data change sequence with the head motion feature sequence to obtain the motion status assessment result of the target object, the motion status assessment result can be sent to a smart wearable device via a direct communication connection. The smart wearable device will then broadcast the result in voice form through its audio output unit. Alternatively, the motion status assessment result can be sent to a user terminal, which will display it in text or chart form in its application interface for the target object to view. The user terminal can also provide the motion status assessment result to the smart wearable device, which will then broadcast it in voice.

[0036] The above-described implementation scheme of this application can directly obtain high-precision time-aligned data by receiving physiological data and head movement data of the target object during the movement process provided by the smart wearable device, without the need for data synchronization between devices. This avoids the latency and errors introduced by multi-dimensional data reception and provides high-quality input data for subsequent processing. Furthermore, by receiving multi-dimensional data collected by a single smart wearable device, the ease of use in different scenarios can be improved without the need for additional auxiliary equipment, simplifying the structure and reducing usage costs.

[0037] After acquiring head motion data, the head motion posture data and absolute orientation data are fused to generate a head motion feature sequence indicating the dynamic changes of the head during movement. The physiological data change sequence is coupled with the head motion feature sequence for analysis, quantitatively assessing the temporal synchronization relationship between physiological indicators and head movements, and identifying changes in movement state. This fully utilizes the powerful computing and storage resources of the cloud to perform deep coupling analysis of the physiological data and head motion data of the target object, capturing changes that are difficult to reflect by a single indicator from the perspective of physiological and motor coordination, achieving a refined and accurate assessment of movement state, and improving the accuracy and reliability of movement state assessment. Moreover, since the complex analysis tasks are carried out in the cloud, smart wearable devices can focus on data collection, which helps to achieve device lightweighting and extend battery life.

[0038] After obtaining motion status assessment results based on coupling analysis, the results are sent to smart wearable devices or user terminals, which can promptly provide motion status information to the target object, making it easier for the target object to adjust the intensity of exercise and / or prevent exercise risks.

[0039] The following describes the head motion feature sequences and their generation process. The head motion feature sequences include at least one of the following: head orientation drift sequence, head angular velocity change sequence, and head posture stability sequence. The head angular velocity change sequence indicates the rate of head rotation during movement, reflecting the intensity of head rotation. The head posture stability sequence indicates the stability of the head in maintaining a specific posture; higher stability indicates stronger motion control. The changes in head motion dynamics over time can be quantified using the head angular velocity change sequence and / or the head posture stability sequence. The head orientation drift sequence indicates the shift in the head's absolute orientation over time, providing a data basis for subsequent determination of orientation dwell patterns or orientation change patterns. The head motion state includes head motion dynamic features and head spatial orientation features, which are quantified using the head orientation drift sequence.

[0040] This involves fusing head motion posture data with absolute orientation data to generate a head motion feature sequence of the target object, including... Figure 2 The following steps are shown: Step 201: Process the head motion posture data and absolute orientation data through a sliding window, and extract multiple feature values ​​representing the head motion state in each window. The multiple feature values ​​include: head orientation drift value, and at least one of the following: head angular velocity change value and head posture stability value.

[0041] After fusing head motion posture data and absolute orientation data to obtain multidimensional motion data representing the head motion state, feature extraction is performed on this multidimensional motion data. During feature extraction, an adaptive sliding window mechanism can be used to divide the data temporally, and feature extraction is performed on the data within each window separately, obtaining multiple feature values ​​for that window. The obtained multiple feature values ​​include head orientation drift, head angular velocity change, and at least one of the following: head orientation drift: the absolute value of the absolute orientation angle change from the start to the end of the window, reflecting the net rotation of the head in the horizontal direction; head angular velocity change: the root mean square of the composite vector of the three-axis angular velocities within the window, reflecting the severity of head swaying; head posture stability: the reciprocal of the variance of the pitch and roll angles within the window, with larger values ​​indicating more stable head posture.

[0042] The window size can be adaptively adjusted according to actual needs. For example, initial window parameters can be set as follows: window length of 2 seconds and sliding step size of 1 to 1.5 seconds, ensuring appropriate data overlap between adjacent windows to guarantee the continuity and smoothness of feature extraction. The sliding step size can also be adjusted according to the actual application scenario. During the adaptive window size adjustment process, the change value of head angular velocity in each window can be calculated and compared with a preset threshold to automatically adjust the length of subsequent windows. For example, when the change value of head angular velocity is lower than the first threshold (e.g., 0.5 rad / s), it can be determined that the target object is currently in a low-intensity exercise state such as walking or jogging, and the window length can be increased to 3 seconds; when the change value of head angular velocity is higher than the second threshold (e.g., 1.0 rad / s), it can be determined that the target object is currently in a high-intensity exercise state such as running, and the window length can be shortened to 1.5 seconds. Furthermore, the window length can be dynamically adjusted based on the detected exercise type or short-term energy changes in head motion data. For example, the variance of the head angular velocity change sequence can be used to automatically distinguish between different types of exercise, such as running and walking, and the window length can be adjusted accordingly. Alternatively, changes in exercise intensity can be detected based on the short-term fluctuations in acceleration amplitude within the window, and the window length can be dynamically adjusted. The adjusted window length is typically set to cover 2 to 4 complete motion cycles (for example, for running, the corresponding window length can cover 4 to 8 steps). Through these adaptive adjustments, each window can adapt to the changes in the dynamic characteristics of the target object's head movement in the current motion state, thereby helping to extract more representative feature values ​​within each window and providing more stable and reliable input data for subsequent temporal coupling analysis.

[0043] Using a sliding window mechanism for data partitioning can divide continuous motion data into data units with fixed time lengths. By extracting features from these data units and replacing the original data with a small number of feature values, the data size can be effectively reduced and the computational complexity can be decreased.

[0044] Step 202: Arrange the feature values ​​extracted from each window in chronological order to form a head motion feature sequence.

[0045] After extracting features from the data within a window to obtain multiple feature values, the extracted feature values ​​from each window are arranged in chronological order to generate a head motion feature sequence. The head motion feature sequence includes a head orientation drift sequence, a head angular velocity change sequence, and a head posture stability sequence, at least one of these. The head orientation drift sequence is generated based on the head orientation drift values ​​within each window, the head angular velocity change sequence is generated based on the head angular velocity change values ​​within each window, and the head posture stability sequence is generated based on the head posture stability values ​​within each window.

[0046] By generating head motion feature sequences, the original multidimensional motion data can be transformed into structured and quantifiable temporal feature sequences, thereby providing standardized motion data input for subsequent coupled analysis of physiological data and head motion data.

[0047] Since the head motion feature sequence includes a head orientation drift sequence, the method also includes: Based on the head orientation drift sequence, determine the orientation dwell mode and / or orientation change mode of the target object's head during the movement process. The orientation dwell mode indicates that the head orientation remains within a stable directional range, and the orientation change mode indicates that the head orientation switches from one stable direction to another stable direction. The coupling analysis of the physiological data change sequence and the head motion feature sequence includes: performing coupling analysis in the orientation dwell mode, and / or performing coupling analysis in the orientation change mode.

[0048] Head orientation drift sequences indicate the shift in the absolute orientation of the head over time. Based on these sequences, the orientation dwell patterns and / or orientation change patterns of the target object's head during movement can be identified. Orientation dwell patterns indicate that the head orientation remains within a stable directional range, reflecting that the target object is in a relatively stable directional state during movement. Orientation change patterns indicate that the head orientation switches from one stable direction to another, reflecting that the target object is actively changing its orientation (such as turning or observing the environment) during movement.

[0049] Before performing coupling analysis, the cloud-based system first uses the K-means clustering algorithm to identify orientation dwell segments and orientation change segments based on the head orientation drift sequence. When the orientation angle center obtained from the clustering is relatively stable and the window orientation drift value is less than a set value (e.g., 15°), it is marked as an orientation dwell mode; when the orientation angle center changes significantly and the window orientation drift value is greater than another set value (e.g., 30°), it is marked as an orientation change mode.

[0050] Data in the orientation dwell mode constitutes a large proportion and is relatively stable, allowing coupling analysis results to purely reflect the evolution trend of the motion state. Data in the orientation change mode, however, is instantaneous and fluctuates wildly; mixing the two modes for analysis would cause these violent fluctuations to interfere with the extraction of steady-state coupling relationships. Therefore, it is necessary to distinguish between the orientation dwell mode and the orientation change mode, performing coupling analysis independently on the data windows of each mode to avoid data interference caused by orientation switching. That is, after determining the orientation dwell mode and / or orientation change mode of the target object's head during motion based on the head orientation drift sequence, when performing coupling analysis on the physiological data change sequence and the head motion feature sequence, coupling analysis should be performed in the orientation dwell mode, and / or in the orientation change mode.

[0051] By dividing the motion process into directional dwell segments and directional change segments, and performing coupling analysis separately, the interference caused by the drastic fluctuations in head movement due to directional switching on the calculation of the physiological-motor coupling relationship can be avoided. For example, during running, when the target object turns on a curve (directional change mode), the head angular velocity will briefly increase, and there will be a time delay in the heart rate response. If the turning segment data is mixed with the straight running segment for unified analysis, it will lower the evaluation result of the overall motion stability. By distinguishing the modes, the data of the directional change segment can be labeled separately and analyzed independently, avoiding its interference with the extraction of the steady-state coupling relationship of the directional dwell segment.

[0052] The coupling analysis results in the directional dwell mode reflect the evolution trend of the target object's motion state during steady-state travel, serving as the primary basis for motion state assessment. The coupling analysis results in the directional change mode can be used to evaluate the target object's steering control capability and stress response coordination, serving as a supplementary reference. By distinguishing between the directional dwell mode and the directional change mode and performing coupling analysis separately, interference from non-steady-state motion data such as steering on stability assessment is avoided, making the assessment results more targeted and accurate.

[0053] The following describes the process of generating motion state assessment results for a target object through coupling analysis. When generating motion state assessment results for a target object by coupling analysis of physiological data change sequences and head motion feature sequences, it includes, for example... Figure 3 The following steps are shown: Step 301: In response to a preset triggering condition, in orientation dwell mode and / or orientation change mode, perform coupling analysis on the change sequence of physiological data and the head motion feature sequence to obtain a temporal coupling parameter sequence. The temporal coupling parameter sequence includes a set of parameters arranged sequentially in time that characterize the coupling relationship between physiological data and head motion data.

[0054] When a preset trigger condition is detected, a coupling analysis is performed on the sequence of physiological data changes and the sequence of head movement characteristics. The preset trigger condition includes at least one of the following: receiving an evaluation command from a smart wearable device or user terminal, reaching a self-triggered timing point, or the received physiological data and / or head movement data meeting the self-triggered timing condition. For the cloud-based system, the coupling analysis can be passively triggered based on the evaluation command or it can be triggered autonomously. For autonomously triggered coupling analysis, for example, the cloud-based system may trigger the analysis when a trigger timing point is determined based on a preset period or preset time point, or it may trigger the analysis autonomously when the received physiological data and / or head movement data meet preset data conditions (such as abnormal heart rate or poor movement stability).

[0055] Since it is necessary to distinguish between orientation dwell mode and orientation change mode, coupling analysis can be performed on the physiological data change sequence and head motion feature sequence under each of the two modes. That is, after determining the orientation dwell mode and / or orientation change mode of the target object's head during movement based on the head orientation drift sequence, coupling analysis is performed on the physiological data change sequence and head motion feature sequence under the orientation dwell mode, and / or, under the orientation change mode, coupling analysis is performed on the physiological data change sequence and head motion feature sequence.

[0056] By coupling the change sequence of physiological data with the head motion feature sequence, a temporal coupling parameter sequence characterizing the phase synchronization relationship or cooperative stability between physiological data and head motion data is obtained. The temporal coupling parameter sequence indicates changes in the temporal coupling relationship between physiological data and head motion data, and includes a set of parameters arranged sequentially in time to characterize the coupling relationship between physiological data and head motion data. As one implementation method, the temporal coupling parameter sequence can be a sequence of PCI (Phase Coupling Index) values ​​calculated based on the phase difference, arranged in chronological order. The phase difference can be calculated through cross-correlation analysis or Hilbert transform in the time domain.

[0057] The physiological data received from the cloud includes at least heart rate data, and selectively includes respiratory data and / or electromyography data. During coupling analysis, the phase coupling index sequence between the corresponding data change sequence and the head motion feature sequence is calculated based on the data type contained in the physiological data.

[0058] In a specific example, the head motion posture data IMU(t) and the absolute orientation data MAG(t) are fused to generate a head motion feature sequence represented as H(t) = f(IMU(t), MAG(t)). Here, IMU(t) represents the head motion posture data acquired by a six-axis sensor, MAG(t) represents the absolute orientation data acquired by a three-axis geomagnetic sensor, and H(t) includes a head orientation drift sequence, as well as at least one of a head angular velocity change sequence and a head posture stability sequence.

[0059] Based on the heart rate data change sequence and head motion feature sequence, a heart rate-head motion phase coupling index sequence is constructed. Within a single window, the heart rate-head motion phase coupling index is calculated using the following formula: PCI_HR = 1 – |φ(HR, H)| / π; φ(HR, H) is the phase difference between heart rate changes and head motion changes, and π is a normalization factor representing the theoretical maximum value of the phase difference, used to normalize the absolute value of the phase difference. A PCI_HR closer to 1 indicates high synchronization, corresponding to a stable motion state; a PCI_HR closer to 0 indicates strong mismatch, corresponding to fatigue or loss of control.

[0060] Based on the data change sequence corresponding to respiratory data and the head movement feature sequence, a respiratory-head movement phase coupling index sequence is constructed. Within a single window, the respiratory-head movement phase coupling index corresponding to that window is calculated using the following formula: PCI_BR = 1 – |φ(BR,H)| / π; used to quantitatively assess the degree of synchronization between respiratory rhythm and head movement rhythm.

[0061] Based on the data change sequence corresponding to electromyography (EMG) data and the head movement feature sequence, an EMG-head movement phase coupling index sequence is constructed. Within a single window, the corresponding EMG-head movement phase coupling index is calculated using the following formula: PCI_EMG = 1 – |φ(EMG,H)| / π; this is used to quantitatively assess the synergistic relationship between muscle activity and head movement control.

[0062] Within any window, PCI_total is calculated based on the phase coupling index: PCI_total = w1·PCI_HR + w2·PCI_BR + w3·PCI_EMG; where w1, w2, and w3 are weighting parameters. As the sliding window advances along the time axis, the PCI_total values ​​output by each window are arranged in chronological order, thus forming a sequence of temporal coupling parameters.

[0063] As another implementation method, time-frequency domain analysis can also be used to calculate PCI. For example, wavelet packet transform can be used to perform time-frequency decomposition on the physiological data change sequence and the head motion feature sequence, extracting wavelet coefficients of characteristic frequency bands related to the motion frequency, and calculating the time-frequency coherence coefficient between the two sets of wavelet coefficients. This time-frequency coherence coefficient reflects the degree of phase synchronization between signals in a specific frequency band, and therefore can be used as a PCI value to generate a sequence of temporal coupling parameters. Compared to methods that only calculate time-domain correlation or frequency-domain energy, time-frequency coherence analysis can more accurately characterize the dynamic coupling relationship between physiological signals and motion signals, thereby improving the sensitivity and robustness of motion state assessment.

[0064] The following describes the coupling analysis process using heart rate (HR) data and a sampling frequency of 25Hz as an example. First, the heart rate data is bandpass filtered (e.g., a passband range of 0.5Hz to 5Hz) in the cloud to remove baseline drift and high-frequency noise. Then, the filtered heart rate data change sequence is coupled with the main components of the head motion feature sequence (e.g., head angular velocity change sequence) for time-frequency coupling analysis. The specific operation is as follows: Using the Daubechies-4 wavelet basis function, the heart rate data change sequence and the head angular velocity change sequence were decomposed into 3-level wavelet packet decompositions, resulting in 8 frequency sub-bands (covering the range of 0–12.5 Hz). The sub-band closest to the current motion frequency (determined by the spectral peak of the head angular velocity change sequence, typically 1.5 Hz–3 Hz) was selected, and the wavelet coefficients of that sub-band within each window were extracted.

[0065] Within each window, calculate the coherence coefficient between the heart rate wavelet coefficient and the head angular velocity wavelet coefficient:

[0066] in, and The autopower spectra of heart rate and head angular velocity signals are shown below. The cross-power spectrum and coherence coefficient of the two are given. The value range is [0, 1], which is used to represent the degree of linear correlation between the heart rate signal and the head angular velocity signal at a given frequency f.

[0067] Since the coherence coefficient reflects the degree of phase lock between heart rate fluctuations and head movements near a specific exercise frequency, and wavelet packet decomposition has confined the signal to the frequency band most relevant to the exercise frequency, the average coherence coefficient within this frequency band can be directly calculated as the phase coupling index (PCI) corresponding to that window. A PCI value closer to 1 indicates a high degree of phase in which heart rate fluctuations and head movements are in phase at the exercise frequency, reflecting good stability of the exercise state and good cardiopulmonary-exercise coordination. A decreasing PCI value indicates weakened physiological-exercise coupling, suggesting fatigue accumulation or decreased motor control ability. As the sliding window progresses along the time axis, the PCI value for each window is calculated sequentially, resulting in the temporal coupling parameter sequence {PCI_1, PCI_2, ..., PCI_N}.

[0068] By elevating coupling analysis from simple numerical comparison to a refined characterization of temporal phase synchronization and time-frequency phase coupling, the correlation between physiological data and head motion data can be quantitatively assessed at multiple levels. At the temporal level, phase difference analysis can directly reflect the synchronization relationship between physiological data and head motion data at each moment. At the time-frequency level, coherence analysis can focus on the degree of phase locking within specific motion frequency bands, effectively filtering out noise interference unrelated to motion. This coupling analysis can effectively distinguish between different causes of decreased motion state, such as physiological delay and motor control dysregulation, improving the sensitivity and specificity of the assessment.

[0069] Step 302: Based on the comparison results of the parameters in the temporal coupling parameter sequence with the motion state stability decision threshold, and / or the parameter change trend of the temporal coupling parameter sequence, generate the motion state stability index and / or motion risk trend index of the target object.

[0070] After obtaining the temporal coupling parameter sequence, each parameter in the sequence is compared with a preset motion state stability decision threshold to determine whether the motion state of each window is stable; and / or, by analyzing the changing trends of each parameter in the temporal coupling parameter sequence over time, it can be predicted whether the motion state will develop into a disadvantageous state. Based on the comparison results and / or parameter changing trends, motion state stability indicators and / or motion risk trend indicators of the target object are generated to obtain the motion state evaluation results of the target object.

[0071] Among them, the motion stability index indicates the degree of physiological-motor coordination of the target object during the motion process, such as motion stability score and / or motion control efficiency index, and the motion risk trend index indicates the abnormal trend of the target object's motion state, such as fatigue development trend and physiological-motor mismatch warning.

[0072] As a specific example, each PCI value in the temporal coupling parameter sequence is compared with a preset motion stability decision threshold (e.g., 0.7). If the PCI values ​​for multiple consecutive windows (e.g., 3 windows) are all below 0.7, it is determined that motion stability has decreased. Simultaneously, a linear trend fitting is performed on the PCI values ​​over a period of time in the temporal coupling parameter sequence. If the slope obtained from the fitting is continuously negative, it indicates that the PCI values ​​are showing a continuous downward trend, and a motion risk trend warning is generated (e.g., fatigue accumulation, suggesting a reduction in intensity). Finally, motion stability indicators (e.g., comprehensive score) and motion risk trend indicators (e.g., normal / attention / warning level) are output.

[0073] In the above implementation scheme, by coupling the change sequence of physiological data with the head motion feature sequence, a temporal coupling parameter sequence is obtained. Based on the comparison results of the parameters in the temporal coupling parameter sequence with the motion state stability decision threshold, and / or the change trend of the parameters over time, a motion state assessment result is generated. This can analyze the motion state from the perspective of physiological and motor coordination, thereby improving the accuracy and reliability of motion state assessment.

[0074] In one embodiment of this application, the cloud can provide motion status reminders based on the physiological state of the target object during its movement. For example... Figure 4 As shown, the method also includes the following steps: Step 401: In response to at least one of the physiological data satisfying its corresponding warning trigger condition, generate motion status reminder information, wherein the motion status reminder information is used to guide the target object to adjust the exercise intensity and / or prompt exercise risk in the current exercise state, and the warning trigger condition is a personalized threshold condition determined based on the target object's physiological parameters, or a default threshold condition determined based on population statistics.

[0075] The physiological data received by the cloud includes heart rate data, and selectively includes respiratory data and / or electromyography (EMG) data. For each received physiological data point, the cloud monitors it in real-time or periodically to ensure it meets its corresponding warning trigger conditions. For example, it monitors whether heart rate data exceeds the heart rate warning value, whether respiratory data exceeds the respiratory rate warning value, and whether EMG data exceeds the EMG activity warning range. Each physiological data point corresponds to a warning trigger condition, and the warning trigger conditions for each physiological data point are independent of each other.

[0076] The warning trigger conditions can be determined based on the target object's physiological parameters, serving as a personalized threshold to measure whether its physiological data deviates from its normal range. Alternatively, these trigger conditions can be determined based on population statistics, serving as a default threshold to measure whether the target object's physiological data deviates from the statistical mean. If at least one of the physiological data points meets its corresponding warning trigger condition, it indicates an abnormality in the physiological data, and a motion status alert should be generated.

[0077] For the cloud-based system, when it detects at least one physiological data point that meets the corresponding warning trigger condition, it generates an exercise status alert. This alert is primarily generated based on the physiological data that meets the warning trigger condition and is used to guide the target individual to reduce intensity and / or indicate potential exercise risks. For example, when heart rate data exceeds a personalized heart rate warning value, the cloud can generate an exercise status alert message such as "Heart rate is too high, it is recommended to reduce speed."

[0078] Step 402: Send the motion status reminder information to the smart wearable device or user terminal via the communication connection.

[0079] After generating the exercise status reminder information, the cloud sends the reminder information to the smart wearable device or user terminal via a communication connection. The smart wearable device then broadcasts the reminder via voice, or the user terminal displays it through an application interface, in order to provide the exercise status reminder information to the target object, thereby guiding the target object to reduce the intensity and / or alerting the target object to the existence of exercise risks.

[0080] As a preferred method, the cloud can also monitor whether the target's heart rate transitions from one heart rate zone to another. For example, when the heart rate is detected to transition from zone 1 to zone 2, the cloud can generate a zone transition reminder, helping the target to understand changes in their exercise intensity in a timely manner so as to proactively adjust their exercise pace and avoid exercise risks caused by excessive intensity. Similarly, if respiratory data or electromyography data show a zone shift, the cloud can also generate a reminder for the target.

[0081] The above-described implementation scheme of this application, when the physiological data of the target object meets the warning triggering conditions, generates exercise status reminder information to guide the target object to adjust the exercise intensity and / or indicate exercise risks, and feeds the exercise status reminder information back to the smart wearable device or user terminal, which can realize real-time warning and timely intervention of exercise risks, and assist the target object in scientific exercise management.

[0082] In another embodiment of this application, the cloud can interact with the target object during or after its movement to provide the target object with demand information. Specifically, the method further includes the following process: In response to receiving a phased query request initiated during the movement of the target object, based on the phased analysis of the received physiological data and head movement data, phased movement information is generated and sent to the smart wearable device or user terminal via a communication connection; and after sending the phased movement information, in response to a first interaction request from the smart wearable device or user terminal, a first interactive feedback information is generated and sent via a communication connection. and / or In response to the end of the exercise or receiving a motion status query request initiated by the target object after the end of the exercise, a motion performance report for the exercise is generated based on the motion status evaluation results of the target object during the exercise and sent to the smart wearable device or user terminal. After sending the motion performance report, in response to a second interaction request from the smart wearable device or user terminal, a second interaction feedback information is generated and sent via the communication connection.

[0083] During the target's movement, if a phased query request is received from the target through a smart wearable device or user terminal, the system responds to the request by performing a phased analysis on the received physiological and head movement data of the target, generating phased movement information, and sending it to the smart wearable device or user terminal. This allows the target to understand the movement performance of the completed phase and the current movement status in a timely manner, thereby assisting the target in making reasonable adjustments to subsequent movements.

[0084] After sending phased motion information, if a first interaction request is received from a smart wearable device or user terminal, first interaction feedback information adapted to the first interaction request is generated. The received first interaction request may or may not be related to the phased motion information fed back in the previous stage; this embodiment uses the example of the first interaction request being related to the previous stage of motion for explanation. In this case, the cloud can generate the first interaction feedback information based on the coupling analysis results of the previous stage (such as the changing trend of time-series coupling parameters, motion state stability indicators, etc.). The first interaction feedback information is provided to the smart wearable device or user terminal via a communication connection, enabling the target object to conduct multi-round conversational interactions with the cloud during the motion process, converting motion analysis into intelligent question-and-answer sessions that can be explored in depth as needed, thereby improving the user experience.

[0085] For example, after the cloud sends a phased exercise message stating "Your exercise status was stable in the previous stage," the target user can further ask, "Why do I feel more tired than usual?" Upon receiving this first interaction request, the cloud generates first interaction feedback information based on the coupling analysis of the previous stage, such as "The synchronization between heart rate and head movement has decreased; it is recommended to appropriately reduce the intensity," and sends this information to the target user's headphones for voice broadcast.

[0086] After the exercise concludes, or upon receiving a request from the target user to query their exercise status, the cloud generates an exercise performance report based on the exercise status assessment results generated at each stage during the exercise and sends it to the smart wearable device or user terminal. The exercise status assessment results are intermediate assessment information generated and fed back in real time during the exercise, while the exercise performance report is a comprehensive summary report generated after the exercise based on the overall exercise status assessment results.

[0087] After sending the exercise performance report, if a second interaction request is received from a smart wearable device or user terminal, second interaction feedback information adapted to the second interaction request is generated. This embodiment uses an example where the second interaction request is related to the current exercise. In this case, the cloud can generate second interaction feedback information based on the coupling analysis during the exercise process, and provide the second interaction feedback information to the smart wearable device or user terminal via a communication connection. This allows the target user to engage in multi-round conversational interaction with the cloud after the exercise, transforming the exercise analysis into intelligent question-and-answer sessions that can be explored in depth as needed, thus improving the user experience.

[0088] The above implementation scheme, by providing feedback on the needs of the target object during and / or after exercise, and by providing interactive feedback based on interactive requests, can transform exercise analysis into intelligent question answering that can be explored in depth as needed. This allows the target object to obtain more targeted content without interrupting exercise or reviewing it after exercise, thereby improving the practicality of exercise status assessment results and user experience.

[0089] In yet another embodiment of this application, the method further includes Figure 5 The following steps are shown: Step 501: Before the target object moves, based on the target object's historical movement status assessment results and / or historical physiological data, combined with additional information, generate movement suggestion information for this movement. The additional information includes at least one of the following: the target object's current physiological data collected in real time, and personal status information indicating the target object's subjective feelings input through a smart wearable device or user terminal.

[0090] Before the target object begins movement, movement suggestions can be provided based on the target object's movement suggestion request, or the cloud can proactively push movement suggestions to the target object. When generating movement suggestions for this movement, the suggestions are based on the target object's historical movement status assessment results and / or historical physiological data, combined with additional information related to the target object's current state.

[0091] Historical exercise status assessment results reflect the physiological-motor coupling state and long-term trends of the target subject during historical exercise; historical physiological data reflects the historical levels and fluctuation characteristics of the target subject's physiological indicators. Additional information includes at least one of the following: real-time collected current physiological data of the target subject and personal status information indicating the target subject's subjective feelings. The target subject's current physiological data reflects their immediate physiological state before exercise, such as heart rate and respiration; the personal status information indicating the target subject's subjective feelings is input through smart wearable devices or user terminals, reflecting the target subject's self-assessment of their current state, such as fatigue perception, sleep duration and quality, etc. By combining historical exercise status assessment results and / or historical physiological data with additional information to generate exercise recommendations for the current exercise, a correlation can be established between historical exercise performance and current physical condition, making the exercise recommendations more targeted, scientific, and safe.

[0092] Step 502: Send the exercise suggestion information to the smart wearable device or user terminal via the communication connection. The exercise suggestion information includes at least one of the following: recommended exercise intensity, recommended exercise duration, exercise risk warning, and warm-up suggestions.

[0093] The generated exercise recommendations include at least one of the following: recommended exercise intensity, recommended exercise duration, exercise risk warnings, and warm-up suggestions, to help the target individual scientifically plan their exercise and prevent exercise risks. The generated exercise recommendations are sent to a smart wearable device or user terminal, which then provides the recommendations to the target individual.

[0094] The above implementation plan generates exercise suggestions based on the target subject's historical exercise status assessment results and / or historical physiological data, combined with additional information, before exercise. This can provide more targeted and safe planning for the exercise based on a comprehensive consideration of the target subject's long-term exercise performance and current real-time status, thus assisting the target subject in exercising scientifically.

[0095] In another embodiment of this application, the method further includes Figure 6 The following steps are shown: Step 601: Based on the motion status assessment results generated by the target object during multiple exercise sessions, and / or the long-term trend of the motion status assessment results, generate personalized exercise and health recommendations; wherein, the personalized exercise and health recommendations are generated autonomously by the cloud or based on the recommendation query command.

[0096] The cloud platform can provide personalized exercise and health recommendations to target individuals based on their suggested query commands, or it can generate personalized exercise and health recommendations autonomously. When generating personalized exercise and health recommendations, it considers the exercise status assessment results generated during multiple exercise sessions and / or the long-term trends of these assessment results. The generated personalized exercise and health recommendations include at least one of the following: training plan adjustment suggestions, exercise ability development trends, health risk assessments, daily routine suggestions, and recovery time suggestions. Furthermore, the cloud platform can regularly update the personalized exercise and health recommendations to ensure their timeliness and relevance.

[0097] Step 602: Send personalized sports and health recommendations to smart wearable devices or user terminals via communication connection.

[0098] After generating personalized exercise and health recommendations, these recommendations are sent to smart wearable devices or user terminals, which then provide them to the target individuals. By offering personalized exercise and health recommendations, individuals can be assisted in managing their exercise scientifically over the long term.

[0099] The above implementation plan, by generating personalized sports and health recommendations for the target group, can provide targeted health guidance based on a comprehensive consideration of the long-term changing trends of the target group's exercise status, helping the target group to optimize exercise habits and reduce the risk of sports injuries.

[0100] It is understood that the various analysis and processing operations performed in the cloud in this embodiment, such as generating exercise status reminder information, responding to interaction requests, generating exercise suggestion information and personalized exercise and health suggestions, can all be completed through the data analysis engine deployed on it.

[0101] This application provides a motion state assessment system, such as... Figure 7 As shown, it includes a cloud 71, a smart wearable device 72, and a user terminal 73. The smart wearable device 72 establishes a communication connection with the cloud 71. The communication connection can be a direct connection or an indirect connection established through the user terminal 73. The smart wearable device 72 collects physiological data and head movement data during the movement of the target object, and provides the collected physiological data and head movement data directly or through the user terminal 73 to the cloud 71. The head movement data includes at least head movement posture data and absolute orientation data. The cloud 71 receives the physiological data and head motion data provided by the smart wearable device 72, and performs the following operations: fusing the head motion posture data with the absolute orientation data to generate a head motion feature sequence of the target object; performing coupled analysis on the change sequence of the physiological data and the head motion feature sequence to generate a motion state evaluation result of the target object; and sending the motion state evaluation result to the smart wearable device 72 or the user terminal 73 via the communication connection.

[0102] Optionally, when receiving data provided by the smart wearable device 72, the cloud 71 is further configured to: Based on the direct connection with the smart wearable device 72, the system receives the physiological data and head movement data collected and provided by the smart wearable device 72; or, Based on the indirect connection between the user terminal 73 and the smart wearable device 72, the user terminal 72 receives the physiological data and head movement data collected by the smart wearable device 72 and forwarded by the user terminal 73. The smart wearable device 72 interacts with the cloud 71 through the application on the user terminal 73.

[0103] Optionally, the head motion feature sequence includes a head orientation drift sequence, and the cloud 71 is further used for: Based on the head orientation drift sequence, the orientation dwell mode and / or orientation change mode of the target object's head during the movement process are determined. The orientation dwell mode indicates that the head orientation remains within a stable directional range, and the orientation change mode indicates that the head orientation switches from one stable direction to another stable direction. When performing coupling analysis on the change sequence of the physiological data and the head motion feature sequence, the cloud 71 performs coupling analysis in the direction dwell mode and / or in the direction change mode.

[0104] Optionally, the head motion feature sequence further includes at least one of a head angular velocity change sequence and a head posture stability sequence. When generating the head motion feature sequence of the target object, the cloud platform 71 is further configured to: The head motion posture data and the absolute orientation data are processed by a sliding window to extract multiple feature values ​​representing the head motion state in each window. The multiple feature values ​​include: head orientation drift value, and at least one of head angular velocity change value and head posture stability value. The feature values ​​extracted from each window are arranged in chronological order to form the head motion feature sequence.

[0105] Optionally, when generating the motion state evaluation result of the target object, the cloud 71 is further configured to: In response to a preset triggering condition, in the direction dwell mode and / or in the direction change mode, the change sequence of the physiological data and the head motion feature sequence are coupled and analyzed to obtain a temporal coupling parameter sequence, which includes a set of parameters arranged in time sequence that characterize the coupling relationship between the physiological data and the head motion data; Based on the comparison results between the parameters in the temporal coupling parameter sequence and the motion state stability decision threshold, and / or the parameter change trend of the temporal coupling parameter sequence, the motion state stability index and / or motion risk trend index of the target object are generated. The preset triggering conditions include at least one of the following: receiving an evaluation instruction sent by the smart wearable device or the user terminal, reaching the autonomous triggering time, or the received physiological data and / or head movement data meeting the autonomous triggering conditions.

[0106] Optionally, the physiological data includes at least heart rate data, and selectively includes respiratory data and / or electromyography data; the cloud platform 71 is also used for: In response to at least one of the physiological data satisfying its corresponding warning trigger condition, a motion status reminder is generated. The motion status reminder is used to guide the target object to adjust the exercise intensity and / or indicate exercise risks in the current exercise state. The motion status reminder information is sent to the smart wearable device 72 or the user terminal 73 via the communication connection; The warning triggering condition is either a personalized threshold condition determined based on the physiological parameters of the target object, or a default threshold condition determined based on population statistics.

[0107] Optionally, the cloud 71 is also used for: In response to receiving a phased query request initiated during the movement of the target object, based on a phased analysis of the received physiological data and head movement data, phased movement information is generated and sent to the smart wearable device 72 or the user terminal 73 via the communication connection. After sending the phased movement information, in response to a first interaction request from the smart wearable device 72 or the user terminal 73, a first interaction feedback information is generated and sent via the communication connection. and / or In response to the end of the current exercise or receiving a motion status query request initiated by the target object after the end of the current exercise, a motion performance report for the current exercise is generated based on the motion status evaluation result of the target object during the current exercise and sent to the smart wearable device 72 or the user terminal 73. After sending the motion performance report, in response to a second interaction request from the smart wearable device 72 or the user terminal 73, a second interaction feedback information is generated and sent through the communication connection.

[0108] Optionally, the cloud 71 is also used for: Before the target object moves, exercise suggestion information is generated based on the target object's historical exercise status assessment results and / or historical physiological data, combined with additional information. The additional information includes at least one of the following: the target object's current physiological data collected in real time, and personal status information indicating the target object's subjective feelings input through the smart wearable device or the user terminal. The exercise suggestion information is sent to the smart wearable device 72 or the user terminal 73 via the communication connection. The exercise suggestion information includes at least one of the following: recommended exercise intensity, recommended exercise duration, exercise risk warning, and warm-up suggestions.

[0109] Optionally, the cloud 71 is also used for: Based on the exercise status assessment results generated by the target object during multiple exercise sessions, and / or the long-term trend of the exercise status assessment results, personalized exercise and health recommendations are generated. The personalized sports and health recommendations are sent to the smart wearable device 72 or the user terminal 73 via the communication connection; The personalized sports and health recommendations are generated autonomously by the cloud platform 71 or based on recommendation query commands.

[0110] Optionally, the smart wearable device 72 is an open-back headphone, including but not limited to bone conduction headphones, air conduction headphones, ear-hook headphones, and clip-on headphones. These headphones typically have at least one of ear-hook, neckband, or clip-on structures, ensuring a secure fit and synchronized movement with the user's head. This guarantees accurate acquisition of head movement data and avoids measurement deviations caused by device loosening. Simultaneously, its open-back design does not occupy ear canal space, facilitating the integration of multiple physiological sensing components for multimodal data acquisition. It also allows users to maintain awareness of ambient sounds, improving safety during outdoor activities and comfort during extended wear. Furthermore, open-back headphones can collect physiological data from the head area. Compared to data collected from within the ear canal, the signal quality is less affected by factors such as sweat and ear canal deformation, contributing to more accurate physiological data and improving the accuracy of motion analysis results.

[0111] In this embodiment, the earphone serves as the sole data acquisition device, enabling synchronous acquisition and natural time alignment of physiological data and head movement data. This eliminates timestamp alignment errors between multiple devices and provides high-quality input data for coupled analysis.

[0112] In one optional implementation, the open-back headphones incorporate a low-power MCU (Microcontroller Unit) storing a lightweight evaluation model trained on historical data. When network connectivity is available (e.g., a normal Bluetooth connection between the headphones and the user's device, and a stable network connection on the user's device), the headphones upload collected physiological and head movement data to the cloud in real time. The cloud executes the aforementioned evaluation algorithm and provides the evaluation results to the user via voice feedback. When the user enters an area without network access (e.g., running in mountainous terrain), the headphones detect an interruption in their connection to the cloud and automatically switch to local evaluation mode: the headphone MCU uses the lightweight evaluation model to perform lightweight processing on the local data, quickly outputting a simplified stability score (e.g., "Good / Caution / Danger"), which is then broadcast through the headphone speaker. Once the network is restored, the headphones upload the local evaluation results and a portion of the cached raw data to the cloud. The cloud uses this data to incrementally learn and continuously optimize the model's accuracy using a personalized motion evaluation model.

[0113] By executing complex algorithms in the cloud when the network is normal to achieve high-precision evaluation, and providing real-time feedback through lightweight local evaluation by smart wearable devices when the network is interrupted, the system ensures both the depth of evaluation and takes into account the actual situation of unstable network in sports scenarios, thereby improving practicality and user experience.

[0114] As the system implementation is basically similar to the method implementation, it is described in a relatively simple way. For relevant details, please refer to the description of the method implementation.

[0115] This application also provides an electronic device, including: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the various processes of the above-described motion state evaluation method embodiments and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0116] For example, Figure 8 A schematic diagram of the physical structure of an electronic device is shown. (For example...) Figure 8As shown, the electronic device may include a processor 810, a communications interface 820, a memory 830, and a communication bus 840. The processor 810, communications interface 820, and memory 830 communicate with each other via the communication bus 840. The processor 810 can call logical instructions stored in the memory 830. The processor 810 is used to execute various processes of the motion state evaluation method according to the embodiments of this application, which will not be described in detail here.

[0117] Furthermore, the logical instructions in the aforementioned memory 830 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application.

[0118] This application also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the various processes of the above-described motion state evaluation method embodiments and achieves the same technical effect. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0119] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0120] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0121] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A motion state assessment method, applied in the cloud, characterized in that, The cloud establishes a communication connection with the smart wearable device, wherein the communication connection is a direct connection between the smart wearable device and the cloud or an indirect connection established through a user terminal, and the method includes: The device receives physiological data and head movement data collected and provided by the smart wearable device during the movement of the target object. The head movement data includes at least head movement posture data and absolute orientation data. The head motion posture data is fused with the absolute orientation data to generate a head motion feature sequence of the target object; The change sequence of the physiological data is coupled with the head motion feature sequence for analysis to generate an assessment result of the motion state of the target object. The motion state assessment results are sent to the smart wearable device or the user terminal via the communication connection.

2. The method according to claim 1, characterized in that, The receipt of physiological data and head movement data of the target object during its movement process, collected and provided by the smart wearable device, includes: Based on a direct connection with the smart wearable device, the system receives the physiological data and head movement data collected and provided by the smart wearable device; or, Based on the indirect connection between the user terminal and the smart wearable device, the user terminal receives the physiological data and head movement data collected by the smart wearable device and forwarded by the user terminal. The smart wearable device interacts with the cloud through an application on the user terminal.

3. The method according to claim 1, characterized in that, The head motion feature sequence includes a head orientation drift sequence, and the method further includes: Based on the head orientation drift sequence, the orientation dwell mode and / or orientation change mode of the target object's head during the movement process are determined. The orientation dwell mode indicates that the head orientation remains within a stable directional range, and the orientation change mode indicates that the head orientation switches from one stable direction to another stable direction. The coupling analysis of the physiological data change sequence and the head motion feature sequence includes: performing coupling analysis in the direction dwell mode, and / or performing coupling analysis in the direction change mode.

4. The method according to claim 3, characterized in that, The head motion feature sequence further includes at least one of a head angular velocity change sequence and a head posture stability sequence. The step of fusing the head motion posture data with the absolute orientation data to generate the head motion feature sequence of the target object includes: The head motion posture data and the absolute orientation data are processed by a sliding window to extract multiple feature values ​​representing the head motion state in each window. The multiple feature values ​​include: head orientation drift value, and at least one of head angular velocity change value and head posture stability value. The feature values ​​extracted from each window are arranged in chronological order to form the head motion feature sequence.

5. The method according to claim 4, characterized in that, The coupling analysis of the change sequence of the physiological data and the head motion feature sequence to generate the motion state assessment result of the target object includes: In response to a preset triggering condition, in the direction dwell mode and / or in the direction change mode, the change sequence of the physiological data and the head motion feature sequence are coupled and analyzed to obtain a temporal coupling parameter sequence, which includes a set of parameters arranged in time sequence that characterize the coupling relationship between the physiological data and the head motion data; Based on the comparison results between the parameters in the temporal coupling parameter sequence and the motion state stability decision threshold, and / or the parameter change trend of the temporal coupling parameter sequence, the motion state stability index and / or motion risk trend index of the target object are generated. The preset triggering conditions include at least one of the following: receiving an evaluation instruction sent by the smart wearable device or the user terminal, reaching the autonomous triggering time, or the received physiological data and / or head movement data meeting the autonomous triggering conditions.

6. The method according to claim 1, characterized in that, The physiological data includes at least heart rate data, and optionally includes respiratory data and / or electromyographic data; the method further includes: In response to at least one of the physiological data satisfying its corresponding warning trigger condition, a motion status reminder is generated. The motion status reminder is used to guide the target object to adjust the exercise intensity and / or indicate exercise risks in the current exercise state. The motion status reminder information is sent to the smart wearable device or the user terminal via the communication connection; The warning triggering condition is either a personalized threshold condition determined based on the physiological parameters of the target object, or a default threshold condition determined based on population statistics.

7. The method according to claim 1, characterized in that, The method further includes: In response to receiving a phased query request initiated during the movement of the target object, based on a phased analysis of the received physiological data and head movement data, phased movement information is generated and sent to the smart wearable device or the user terminal via the communication connection; and after sending the phased movement information, in response to a first interaction request from the smart wearable device or the user terminal, a first interaction feedback information is generated and sent via the communication connection. and / or In response to the end of the current exercise or receiving a motion status query request initiated by the target object after the end of the current exercise, a motion performance report for the current exercise is generated based on the motion status evaluation result of the target object during the current exercise and sent to the smart wearable device or the user terminal. After sending the motion performance report, in response to a second interaction request from the smart wearable device or the user terminal, a second interaction feedback message is generated and sent via the communication connection.

8. The method according to claim 1, characterized in that, The method further includes: Before the target object moves, exercise suggestion information is generated based on the target object's historical exercise status assessment results and / or historical physiological data, combined with additional information. The additional information includes at least one of the following: the target object's current physiological data collected in real time, and personal status information indicating the target object's subjective feelings input through the smart wearable device or the user terminal. The exercise suggestion information is sent to the smart wearable device or the user terminal via the communication connection. The exercise suggestion information includes at least one of the following: recommended exercise intensity, recommended exercise duration, exercise risk warning, and warm-up suggestions.

9. The method according to claim 1, characterized in that, The method further includes: Based on the exercise status assessment results generated by the target object during multiple exercise sessions, and / or the long-term trend of the exercise status assessment results, personalized exercise and health recommendations are generated. The personalized sports and health recommendations are sent to the smart wearable device or the user terminal via the communication connection. The personalized sports and health recommendations are generated autonomously by the cloud or based on recommendation query commands.

10. A motion state assessment system, characterized in that, It includes a cloud, a smart wearable device, and a user terminal. The smart wearable device establishes a communication connection with the cloud, which can be a direct connection or an indirect connection established through the user terminal. The smart wearable device collects physiological data and head movement data during the movement of the target object, and provides the collected physiological data and head movement data directly or through the user terminal to the cloud. The head movement data includes at least head movement posture data and absolute orientation data. The cloud receives the physiological data and head movement data provided by the smart wearable device and executes the motion state assessment method as described in any one of claims 1 to 9.

11. The system according to claim 10, characterized in that, The smart wearable device is an open-back headphone.

Images