An intelligent guiding system for physical training

Abstract

The application provides a physical training intelligent guidance system, comprising a collection module, a preprocessing module, a feature extraction module, a rhythm analysis module and a feedback module; the collection module is used for collecting training state data of a user, wherein the training state data comprises IMU data and breathing signal data; the preprocessing module is used for preprocessing the obtained training state data to obtain preprocessed training state data; the feature extraction module is used for performing feature extraction processing on the preprocessed training state data to obtain training state features of the user, wherein the training state features comprise force exertion features and breathing phase features of the user; the rhythm analysis module is used for performing rhythm analysis on the training state features of the user to obtain a user training rhythm analysis result; and the feedback module is used for adjusting the current user training rhythm according to the user training rhythm analysis result. The application helps to improve the effect of physical training of the user.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent guidance of physical training, and particularly relates to an intelligent guidance system for physical training. BACKGROUND

[0002] At present, for monitoring of a user in a physical training process (for example, strength training or rhythm-based load training), only the heart rate and training times of the user are simply recorded, and based on the simple data, the user cannot accurately know the current training situation and cannot adjust the training intensity (for example, load intensity or rhythm speed of load training). SUMMARY

[0003] In view of the above problems, the present application aims to provide an intelligent guidance system for physical training.

[0004] The object of the present application is achieved by adopting the following technical solutions:

[0005] The present application provides an intelligent guidance system for physical training, comprising a collection module, a preprocessing module, a feature extraction module, a rhythm analysis module and a feedback module, wherein,

[0006] The collection module is used for collecting training state data of a user and transmitting the training state data to the preprocessing module, wherein the training state data comprises IMU data and breathing signal data;

[0007] The preprocessing module is used for preprocessing the obtained training state data to obtain preprocessed training state data;

[0008] The feature extraction module is used for performing feature extraction processing on the preprocessed training state data to obtain training state features of the user, wherein the training state features comprise force features and breathing phase features of the user;

[0009] The rhythm analysis module is used for performing rhythm analysis on the training state features of the user to obtain a user training rhythm analysis result;

[0010] The feedback module is used for adjusting the current user training rhythm according to the user training rhythm analysis result.

[0011] Preferably, the collection module comprises an IMU sensor unit and a breathing belt unit, wherein,

[0012] The IMU sensor unit is used for obtaining IMU data of the user;

[0013] The breathing belt unit is used for obtaining breathing signal data of the user, wherein the breathing signal data.

[0014] Preferably, the preprocessing module comprises a filtering unit and an outlier detection unit; wherein,

[0015] The filtering unit is configured to perform band-pass filtering on the acquired IMU data and the respiratory signal data respectively, to eliminate high-frequency signal interference in the data.

[0016] The outlier detection unit is configured to perform outlier detection processing on the acquired IMU data and the respiratory signal data respectively, to correct outliers in the data.

[0017] Preferably, the feature extraction module comprises a force feature extraction unit and a respiratory phase feature extraction unit; wherein,

[0018] The force feature extraction unit is configured to perform vector modulo processing on the three-axis data of the preprocessed IMU data, and square the obtained modulus value to obtain continuous force feature data:

[0019]

[0020] wherein, represents the force feature corresponding to time t, represents the data modulus value obtained by fusing the three-axis vector of the IMU data;

[0021] The respiratory phase feature extraction unit is configured to perform Hilbert transform on the preprocessed respiratory signal data to obtain the instantaneous phase of the respiratory signal data corresponding to each time as the respiratory phase feature, wherein and extracts the corresponding respiratory period / frequency according to the obtained respiratory phase feature.

[0022] Preferably, the rhythm analysis module comprises a windowing unit and a rhythm analysis unit; wherein,

[0023] The windowing unit is configured to align the force features and the respiratory phase features according to the time scale, and perform windowing processing on the force features and the respiratory phase features respectively, which contain continuous N respiratory periods before the current time based on the respiratory period, to obtain training state features in the window time period

[0024] The rhythm analysis unit is configured to perform rhythm analysis on the training state features in the current window time period, including:

[0025] Calculate the rhythm consistency factor of the training state features, wherein the rhythm consistency factor calculation function used is:

[0026]

[0027] wherein, represents the rhythm consistency factor, wherein , The method comprises the following steps: representing a window time period, representing the i-th sampling point in the window time period; representing the force feature of the i-th sampling point in the window time period, representing the breathing phase feature of the i-th sampling point in the window time period;

[0028] The obtained rhythm consistency factor is used as the user training rhythm analysis result. The method comprises the following steps:

[0029] Preferably, the feedback module comprises a time recording unit, an initial recording unit and an intensity adjusting unit; wherein,

[0030] The time recording unit is used to obtain the total training time of the current user ;

[0031] The initial recording unit is used to record the rhythm consistency factor data obtained by the user in the warm-up stage ;

[0032] The intensity adjusting unit is used to adjust the training rhythm of the current user according to the user training rhythm analysis result.

[0033] Preferably, the intensity adjusting unit adjusts the training rhythm of the current user according to the user training rhythm analysis result, specifically comprising:

[0034] The training intensity factor is updated according to the current rhythm consistency factor, wherein the training intensity factor updating function used is:

[0035]

[0036] wherein, represents the updated training intensity factor, represents the training intensity factor at the current time, represents the initial training intensity factor, which is initialized and set by the user, represents an adjustment control factor, wherein , represents a time control factor, wherein when , when , , and respectively represent the set adjustment time constant and the training cycle time constant, wherein , represents the rhythm consistency factor at the current time, represents the median of the rhythm consistency factor obtained by the user in the warm-up stage, a standard deviation of the rhythm consistency factor obtained by the user in the warm-up stage, represents a micro constant, , and respectively represent a set threshold value, wherein , , and respectively represent a set boundary coefficient, wherein . ; represents a threshold clipping function, so that the function , and satisfies The maximum and minimum values of and ;

[0037] According to the obtained updated training intensity factor, the load control factor or the rhythm control factor is adjusted, wherein

[0038]

[0039]

[0040] wherein, represents the adjusted rhythm control factor, represents the current rhythm control factor, represents the adjusted load control factor, represents the current load control factor;

[0041] The rhythm control factor is used to control the speed of the beat in the rhythm training, and the greater the rhythm control factor, the faster the beat rhythm of the training; the load control factor is used to control the resistance / load size of the load training, wherein the greater the load control factor, the greater the load of the load training.

[0042] According to the rhythm control factor or the load control factor, the current rhythm training or load training intensity is adjusted.

[0043] The beneficial effects of the present application are: the present application proposes a physical training intelligent guidance system, first, by obtaining the action data and breathing data of the user in the physical training process, according to the obtained action data and breathing data, the user training state feature is extracted, and further according to the user training state feature, the current training rhythm condition of the user is analyzed, and the training intensity of the user is adjusted according to the training rhythm analysis result.

[0044] The present application is based on the consistency of the user training action and the breathing rhythm as the basis, can reflect the real rhythm condition of the user current in physical training (especially the repeated strength / load training, or the repeated action training according to the rhythm / beat, etc.), thereby intelligently adjusting the training intensity based on the rhythm state of the user, can effectively improve the effect of the user's physical training. BRIEF DESCRIPTION OF DRAWINGS

[0045] The present application is further described with the help of the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present application, and other drawings can also be obtained by the ordinary skilled in the art without creative labor according to the following drawings.

[0046] Figure 1 The framework structure diagram of a physical training intelligent guidance system shown in the embodiments of the present application is shown in the accompanying drawings.

[0047] Figure 2 For Figure 1 The framework structure diagram of each functional module in the embodiments. DETAILED DESCRIPTION

[0048] The present application is further described in combination with the following application scenarios.

[0049] Referring to Figure 1 , which shows a physical training intelligent guidance system, comprising a collection module, a preprocessing module, a feature extraction module, a rhythm analysis module and a feedback module; wherein,

[0050] The collection module is used for collecting the training state data of the user, and transmitting the training state data to the preprocessing module, wherein the training state data includes IMU data and breathing signal data;

[0051] The preprocessing module is used for preprocessing according to the obtained training state data, to obtain the preprocessed training state data;

[0052] The feature extraction module is used for feature extraction processing according to the preprocessed training state data, to obtain the training state features of the user, wherein the training state features include the force features and the breathing phase features of the user;

[0053] The rhythm analysis module is used for rhythm analysis according to the training state features of the user, to obtain the user training rhythm analysis result;

[0054] The feedback module is used for adjusting the current user training rhythm according to the user training rhythm analysis result.

[0055] The above embodiment of the present application provides a physical training intelligent guidance system, which first acquires motion data and breathing data of a user in a physical training process, extracts user training state features according to the obtained motion data and breathing data, analyzes a current training rhythm condition of the user according to the user training state features, and adjusts training intensity of the user according to an analysis result of the training rhythm.

[0056] Generally, when a user performs physical training (especially repeated strength / load training or repeated motion training according to rhythm / beat), the user usually cooperates force and breathing rhythm with each other, for example, exhales when lifting dumbbells and inhales when returning, and the whole process is cooperated with each other. If breathing is disordered or the timing of force is inconsistent with breathing rhythm, the training effect is reduced or the risk is increased. Therefore, the present application is based on the consistency of user training motion and breathing rhythm, can reflect the real rhythm condition of the user in physical training, and intelligently adjusts training intensity based on the rhythm condition of the user, thereby effectively improving the effect of physical training of the user.

[0057] In one scene, the physical training intelligent guidance system can be connected with existing load adjustment equipment or beat adjustment equipment, thereby controlling and adjusting the equipment, and adjusting the training intensity of the user in real time.

[0058] Preferably, referring to Figure 2 The acquisition module includes an IMU sensor unit and a breathing belt unit; wherein

[0059] The IMU sensor unit is used to acquire IMU data of the user;

[0060] The breathing belt unit is used to acquire breathing signal data of the user, wherein the breathing signal data.

[0061] The user IMU data (for example, three-axis inertia data) can be acquired through the wearable intelligent device, and the breathing signal data of the user can be acquired through the breathing belt device, which facilitates subsequent extraction of motion and breathing feature data of the user, and meets the requirements of real-time monitoring and intelligent adjustment of the physical training process of the user.

[0062] Preferably, the preprocessing module includes a filtering unit and an outlier detection unit; wherein

[0063] The filtering unit is used to perform band-pass filtering processing on the acquired IMU data and breathing signal data respectively, and eliminate high-frequency signal interference in the data;

[0064] The outlier detection unit is configured to perform outlier detection processing on the acquired IMU data and the respiratory signal data respectively, and correct outliers in the data.

[0065] In one scenario, the manner of correcting outliers can be implemented by using median filtering or mean filtering.

[0066] In view of the noise interference that may exist in the acquired IMU data and the respiratory signal data, the pre-processing module is configured to first complete data filtering processing, eliminate high-frequency noise interference, and correct errors, thereby improving data quality.

[0067] Preferably, the feature extraction module includes a force feature extraction unit and a respiratory phase feature extraction unit; wherein,

[0068] The force feature extraction unit is configured to perform vector modulo processing on the three-axis data of the pre-processed IMU data, and square the obtained modulus value to obtain continuous force feature data:

[0069]

[0070] Among them, represents the force feature corresponding to the t moment, represents the data modulus value obtained by fusing the three-axis vector of the IMU data;

[0071] The respiratory phase feature extraction unit is configured to perform Hilbert transform on the pre-processed respiratory signal data to obtain the instantaneous phase of the respiratory signal data corresponding to each moment As the respiratory phase feature, among them , and extracts the corresponding respiratory period / frequency according to the obtained respiratory phase feature.

[0072] In the above embodiment, for the extraction of the user's action force feature, the modulus of the three-axis data of the IMU data is used to feedback the user's force state, and for the user's breathing rhythm state, the instantaneous phase obtained by performing Hilbert transform on the respiratory signal data is used as the user's breathing state feedback. Based on the above two state feature extraction methods, the action state feature and the breathing state feature of the user during the physical training process can be digitized and presented to the user in the form of an envelope line, thereby providing data support for further analysis of the consistency of the user's force and breathing rhythm.

[0073] Preferably, the rhythm analysis module includes a windowing unit and a rhythm analysis unit; wherein,

[0074] The windowing unit is configured to align the acquired force features and breathing phase features in time scale, and perform windowing processing on the force features and the breathing phase features respectively based on the breathing cycle, to obtain training state features in a window time period

[0075] The rhythm analysis unit is configured to perform rhythm analysis on the training state features in the current window time period, including:

[0076] The rhythm consistency factor of the training state features is calculated, wherein the rhythm consistency factor calculation function adopted is:

[0077]

[0078] wherein, The rhythm consistency factor is represented by, wherein , The window time period is represented by The i-th sampling point in the window time period is represented by The force feature of the i-th sampling point in the window time period is represented by The breathing phase feature of the i-th sampling point in the window time period is represented by

[0079] The obtained rhythm consistency factor is taken as the user training rhythm analysis result.

[0080] The rhythm consistency factor proposed in the above embodiment can analyze the consistency of the user's force feature data and breathing phase feature data, so as to reflect whether the force and the breathing phase are consistent each time when the user performs the repeated action. If the force is always concentrated in the same breathing phase (for example, force in the early and middle exhalation), the vector is "pointing to consistency", and the rhythm consistency factor is close to 1. If the force is dispersed to various phases, the rhythm consistency factor tends to 0. Thus, the current training state of the user is reflected.

[0081] In one scenario, during the warm-up stage of the user's physical training (for example, the BAECX is calculated within 1-2 minutes after warm-up, the most stable 30 s segment is selected as the reference, and the median and standard deviation of the reference BAECX are further obtained as the subsequent judgment basis), the rhythm consistency factor of the training stage is first obtained based on the collected user data, and is taken as the standard to feedback the user's action habit, which lays the foundation for the comparison between the real-time rhythm consistency factor and the standard rhythm consistency factor to feedback the user's training state.

[0082] In one scenario, the rhythm consistency factor proposed above is used to feedback the user's action habit ​​This provides real-time feedback on the user's concentration of effort during breathing phases during physical training. When the rhythm consistency factor is close to 1, it indicates that the effort is highly concentrated in a fixed breathing phase range (e.g., repeatedly exerting force in the early to mid-exhalation phase), indicating that the user's movements are stable and within a good rhythm. When the rhythm consistency factor is close to 0, it indicates that the user's effort is scattered across various breathing phases, without a stable "breath-effort binding point," indicating that the user's movements are distorted or that a good rhythm has not been found.

[0083] Preferably, the feedback module includes a time recording unit, an initial recording unit, and an intensity adjustment unit; wherein,

[0084] The time recording unit is used to obtain the total training time of the current user. ;

[0085] The initial recording unit is used to record the rhythm consistency factor data obtained by the user during the warm-up phase. ;

[0086] The intensity adjustment unit is used to adjust the current user training rhythm based on the user training rhythm analysis results, specifically including:

[0087] The training intensity factor is updated based on the current rhythm consistency factor, where the training intensity factor update function is:

[0088]

[0089] in, This represents the updated training intensity factor. This represents the training intensity factor at the current moment. This represents the initial training intensity factor, which is set by the user. Represents the adjustment control factor, where , Represents the time control factor, where when hour, ,when hour, , and These represent the set adjustment time and training cycle time, respectively. , This represents the rhythm consistency factor at the current moment. This represents the median of the pacing consistency factor obtained by users during the warm-up phase. This represents the standard deviation of the rhythm consistency factor obtained by users during the warm-up phase. Represents a small constant. , and respectively represent a set threshold value, wherein , , and respectively represent a set boundary coefficient, wherein . ; represents a threshold clipping function, such that the function , and satisfies the maximum and minimum values of and ;

[0090] According to the obtained updated training intensity factor, the load control factor or the rhythm control factor is adjusted, wherein

[0091]

[0092]

[0093] wherein, represents the adjusted rhythm control factor, represents the current rhythm control factor, represents the adjusted load control factor, represents the current load control factor;

[0094] The rhythm control factor is used to control the speed of the beat in the rhythm training, and the greater the rhythm control factor, the faster the beat rhythm of the training; the load control factor is used to control the resistance / load size of the load training, wherein the greater the load control factor, the greater the load of the load training.

[0095] According to the rhythm control factor or the load control factor, the current rhythm training or load training intensity is adjusted.

[0096] The above-mentioned embodiments of the present application further adjust the training intensity of the user based on the obtained rhythm consistency factor BAECX, wherein the proposed training intensity factor can be adjusted based on the change of the current rhythm consistency factor BAECX, and combined with the current training time length of the user (for example, the cumulative training time of the current action group, or the cumulative training time of the current continuous action) to adjust the training intensity of the user. Thus, the training intensity is reasonably adjusted according to the action rhythm of the user in different training stages, and the training effect is improved.

[0097] In a scene, in the initial training stage (usually smaller in training), the user has not yet entered a stable training rhythm, so the current breathing is usually free, and has not yet formed stable synchronization with the action; therefore, in this stage, if the rhythm consistency factor is low, appropriately increasing the load intensity or rhythm can make the user more easily enter the state and make the breathing and action consistent, so that the user's action and breathing rhythm are adapted, and the training effect is improved. When a period of continuous training is passed, it is assumed that the user has entered the training state at this time, and if the rhythm consistency factor is reduced at this time, it indicates that the user's action is deformed or the rhythm is unstable (for example, the action cannot keep up or the breathing is disturbed, etc.). Therefore, at this time, the action intensity is appropriately reduced according to the training intensity factor, which can appropriately reduce the training pressure of the user, so as to help the user adjust the current training state and re-enter a good training rhythm, and further improve the training effect.

[0098] It should be noted that each functional unit / module in each embodiment of the present application can be integrated in one processing unit / module, or each unit / module can be physically present alone, or two or more units / modules can be integrated in one unit / module. The integrated unit / module can be realized in the form of hardware or software functional unit / module.

[0099] Through the above description of the embodiments, those skilled in the art can clearly understand that the embodiments described herein can be realized in hardware, software, firmware, middleware, code or any appropriate combination thereof. For hardware implementation, the processor can be implemented in one or more of the following units: application specific integrated circuit (ASIC), digital signal processor (DSP), digital signal processing device (DSPD), programmable logic device (PLD), field programmable gate array (FPGA), processor, controller, microcontroller, microprocessor, other electronic units designed to implement the functions described herein, or a combination thereof. For software implementation, part or all of the processes of the embodiments can be instructed by a computer program to relevant hardware. When implemented, the above program can be stored in a computer readable medium or transmitted as one or more instructions or codes on a computer readable medium. The computer readable medium includes computer storage medium and communication medium, wherein the communication medium includes any medium that facilitates the transmission of a computer program from one place to another. The storage medium can be any available medium that can be accessed by a computer. The computer readable medium can include, but is not limited to, RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer.

[0100] It should be pointed out finally that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit the protection scope of the present application. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should analyze that the technical solutions of the present application can be modified or equivalently replaced without departing from the essence and scope of the technical solutions of the present application.

Claims

1. A physical training intelligent guidance system, characterized in that, The method comprises a collection module, a preprocessing module, a feature extraction module, a rhythm analysis module, and a feedback module. The collection module is configured to collect training state data of a user, wherein the training state data comprises IMU data and breathing signal data. The preprocessing module is configured to preprocess the acquired training state data to obtain preprocessed training state data. The feature extraction module is configured to perform feature extraction processing on the preprocessed training state data to obtain training state features of the user, wherein the training state features comprise force exertion features and breathing phase features of the user. The rhythm analysis module is configured to perform rhythm analysis on the training state features of the user to obtain a user training rhythm analysis result. The feedback module is configured to adjust a current user training rhythm based on the user training rhythm analysis result. The rhythm analysis module comprises a windowing unit and a rhythm analysis unit. The windowing unit is configured to align the force exertion features and the breathing phase features according to a time scale, and perform windowing processing on the force exertion features and the breathing phase features that each comprise continuous N breathing cycles before the current time based on a breathing cycle to obtain training state features in a window time period. The rhythm analysis unit is configured to perform rhythm analysis on the training state features in the current window time period, including: calculating a rhythm consistency factor of the training state features, wherein a rhythm consistency factor calculation function is used as follows: wherein, represents a rhythm consistency factor, wherein , in represents a window time period, represents the i-th sampling point in the window time period; represents a force feature of the i-th sampling point in the window time period, represents a respiratory phase feature of the i-th sampling point in the window time period; The resulting tempo consistency factor As a user trains tempo analysis results.

2. The intelligent guiding system for physical training according to claim 1, wherein, The collection module comprises an IMU sensor unit and a breathing belt unit. The IMU sensor unit is configured to acquire IMU data of the user. The breathing belt unit is configured to acquire breathing signal data of the user.

3. The intelligent guidance system for physical training according to claim 2, wherein, The preprocessing module comprises a filtering unit and an outlier detection unit. The filtering unit is configured to perform band-pass filtering processing on the acquired IMU data and breathing signal data respectively to eliminate high-frequency signal interference in the data. The outlier detection unit is configured to perform outlier detection processing on the acquired IMU data and breathing signal data respectively to correct outliers in the data.

4. The intelligent guidance system for physical training according to claim 3, wherein, The feature extraction module comprises a force exertion feature extraction unit and a breathing phase feature extraction unit. The force exertion feature extraction unit is configured to perform vector modulo processing on three-axis data of the preprocessed IMU data, and square the obtained modulus to obtain continuous force exertion feature data as follows: The feedback module comprises a time recording unit, an initial recording unit, and an intensity adjustment unit. wherein, represents the force characteristic corresponding to the t time, represents a data modulus obtained according to the fusion of the three-axis vectors of the IMU data; The respiration phase feature extraction unit is configured to perform Hilbert transform on the preprocessed respiration signal data to obtain an instantaneous phase of the respiration signal data at each time point As the respiration phase feature, wherein And according to the obtained respiration phase feature, a corresponding respiration period / frequency is extracted.

5. The intelligent guidance system for physical training according to claim 1, wherein, The intensity adjustment unit is configured to adjust a current user training rhythm based on the user training rhythm analysis result. The time recording unit is used to obtain the total training time of the current user ; The initial recording unit is used to record the pace consistency factor data obtained by the user in the warm-up phase ; The intensity adjustment unit adjusts the current user training rhythm based on the user training rhythm analysis result, specifically including:

6. The intelligent guidance system for physical training according to claim 5, wherein, updating a training intensity factor based on the current rhythm consistency factor, wherein a training intensity factor update function is used as follows: adjusting a load control factor or a rhythm control factor based on the obtained updated training intensity factor, wherein wherein, represents an updated training intensity factor, represents a training intensity factor at a current time, represents an initial training intensity factor, which is initialized by a user, represents an adjustment control factor, wherein , represents a time control factor, wherein when , when , , and respectively represent a set adjustment time constant and a training period time constant, wherein , represents a pace consistency factor at a current time, represents a median of pace consistency factors obtained by a user in a warm-up phase, represents a standard deviation of pace consistency factors obtained by a user in a warm-up phase, represents a small constant, , and respectively represent a set threshold value, wherein , , and respectively represent a set boundary coefficient, wherein . ; represents a threshold clipping function, such that the function , and satisfies the maximum and minimum values of and ; The rhythm control factor is used to control the speed of the rhythm training, and the greater the rhythm control factor, the faster the rhythm of the training. The load control factor is used to control the resistance / load size of the load training, and the greater the load control factor, the greater the load of the load training. wherein, represents the adjusted pace control factor, represents the current pace control factor, represents the adjusted load control factor, represents the current load control factor; ​ The current tempo training or load training intensity is adjusted according to the tempo control factor or the load control factor. The current tempo training or load training intensity is adjusted according to the tempo control factor or the load control factor.

Images