System for evaluating symmetry of swimming based on light and small IMU (Inertial Measurement Unit)

Abstract

The invention discloses a system for evaluating swimming symmetry based on a light and small IMU (Inertial Measurement Unit). The system comprises a wearable inertial measurement unit, a data relay server and an upper computer, the wearable inertial measurement unit transmits initial data composed of a time sequence, acceleration and angular velocity information to the data receiving module; the data receiving module transmits the initial data to the data analysis module; the data analysis module obtains swimming data from the initial data and transmits the swimming data to the data processing module; the data processing module processes the swimming data through a data processing algorithm to obtain a swimming symmetry evaluation result, and transmits the swimming symmetry evaluation result to the data display module; and the data display module receives and displays the swimming symmetry evaluation result. The problem that swimming monitoring equipment is complex in arrangement is solved, swimming strokes are analyzed through swimming data, the swimming symmetry evaluation result is obtained, the method can be applied to physical fitness analysis, manual intervention is not needed in the whole process, and the method can be directly used for reference of coaches and trainers.

Description

Technical Field

[0001] The present invention belongs to the technical field of motion measurement, and in particular relates to a system for evaluating swimming motion symmetry based on a lightweight IMU. Background Art

[0002] In recent years, the use of wearable devices has increased rapidly, mainly due to their multiple advantages such as ease of use, low cost, ability to obtain different vital signs, and usability in a variety of fields, thus meeting the needs of a large number of different users. Furthermore, with the widespread application of wearable devices, their measurement accuracy and precision have gradually improved, making them play a key role in many important fields.

[0003] Wearable devices can be used in the healthcare field for disease symptom detection, tracking, and monitoring of heart-related diseases, and in the field of sports monitoring for sports such as mountaineering, running, and swimming. On the other hand, wearable devices are often combined with artificial intelligence technology, especially in sports applications, where wearable devices are used to monitor training effects and athlete performance. The current trend in sports applications is to monitor athletes' physiological parameters during training or competition to maximize performance and provide reference for coaches.

[0004] In swimming, video analysis is the mainstream method of evaluation technology. However, this evaluation method is challenging, time-consuming and labor-intensive, and may affect normal training courses and competitions. In the past few decades, alternative methods of video-based evaluation technology have gained attention. MEMS, or micro-electromechanical system, is a multidisciplinary frontier research field developed on the basis of microelectronics technology. It is equipped with an "inertial measurement unit (IMU)" composed of sensor modules such as accelerometers, gyroscopes and magnetometers. This device can provide a comprehensive set of kinematic variables, such as acceleration, angular velocity, speed, rotation, direction, etc., and more kinematic and dynamic variables can be extracted from the above data. Summary of the invention

[0005] The technical problem solved by the present application is: to overcome the deficiencies of the prior art, to provide a method for evaluating the symmetry of swimming motion based on a lightweight IMU, to realize a method for automatically evaluating the symmetry of swimming motion using a wearable inertial measurement unit, to solve the problem that swimming monitoring equipment affects the performance of athletes, to realize the analysis of swimming postures through swimming data, and to obtain the swimming motion symmetry evaluation results by the inertial wearable device, without the need for human intervention throughout the process, and to provide direct reference for coaches and trainers.

[0006] The technical solutions provided by this application are as follows:

[0007] A system for evaluating the symmetry of swimming motion based on a lightweight IMU, comprising a wearable inertial measurement unit, a data transfer server and a host computer; wherein the host computer comprises a data receiving module, a data parsing module, a data processing module and a data display module;

[0008] Wearable inertial measurement unit, which is installed on the athlete to monitor swimming movements, is used to measure the acceleration and angular velocity information of the human shoulder, hip or chest, and send the initial data consisting of time series, acceleration and angular velocity information to the data transfer server;

[0009] A data transfer server, used for receiving the initial data and transmitting the initial data to the data receiving module;

[0010] A data receiving module, receiving initial data and transmitting the initial data to the data parsing module;

[0011] A data analysis module receives initial data, obtains swimming data by analyzing the received initial data through an algorithm, and transmits the swimming data to the data processing module;

[0012] A data processing module receives swimming data, processes the received swimming data through a data processing algorithm to obtain a swimming motion symmetry evaluation result, and transmits the swimming motion symmetry evaluation result to the data display module;

[0013] A data display module receives and displays the swimming motion symmetry evaluation result.

[0014] The wearable inertial measurement unit includes at least one IMU sensor, which is used to measure acceleration, angular velocity information and magnetic intensity information, and obtain corresponding time series; acceleration includes X-axis acceleration, Y-axis acceleration and Z-axis acceleration, angular velocity information includes X-axis angular velocity, Y-axis angular velocity and Z-axis angular velocity, and magnetic intensity information includes X-axis magnetic intensity, Y-axis magnetic intensity and Z-axis magnetic intensity.

[0015] The IMU sensor has a built-in data storage module, a WIFI communication module and a repeatedly rechargeable polymer physical battery. The data storage module is used to store initial data; the WIFI communication module cooperates with the data transfer server to transmit the initial data to the data receiving module; the repeatedly rechargeable polymer physical battery is used to power the data storage module and the WIFI communication module.

[0016] After the IMU sensor is turned on, it is in a state of waiting to connect to WiFi. After connecting to WiFi, the host computer displays an online state. After the IMU sensor receives the "offline storage start command" issued by the host computer, the IMU sensor starts measuring and replies in real time "offline storage in progress"; after the IMU sensor sends an "offline storage end command" to the host computer through the WIFI communication module, the host computer displays an online state, and then the IMU sensor transmits the offline stored data to the host computer through the WIFI communication module; if the "offline storage end command" is not received but the data storage module is full at this time and offline storage cannot continue, the IMU sensor actively sends an "offline storage end reply frame" according to the preset protocol after sending the last frame of data, the IMU sensor has stopped storing, and the host computer displays an offline state.

[0017] The swimming data includes movement state and swimming posture; the swimming movement symmetry evaluation result includes rotation angle, rotation time, rotation speed and rotation symmetry index.

[0018] The data analysis module analyzes the received initial data to determine whether the frame header checksum of each frame of data is correct and the frame sequence number is continuous without frame loss. If it is found that the frame header checksum is incorrect and the frame sequence number is discontinuous, a "data incomplete" prompt will be given, and a strapdown inertial navigation update algorithm for handling this situation will be provided for the user's reference to determine whether it can be used directly; if the frame header checksum of each frame of data is correct and the frame sequence number is continuous without frame loss, strapdown inertial navigation is performed to obtain swimming data based on the attitude angle obtained by navigation.

[0019] In the strapdown inertial navigation, the initial attitude angle of the IMU sensor is obtained according to the acceleration and magnetic field strength information at the time of power-on, and then the initial attitude matrix from the carrier coordinate system to the navigation coordinate system is obtained; then, the quaternion is continuously updated according to the time series, acceleration and angular velocity information and magnetic field strength information measured by the wearable inertial measurement unit, and then the real-time attitude angle of the IMU sensor is obtained through quaternion calculation.

[0020] The data analysis module obtains swimming data by analyzing the received initial data through an analysis algorithm, including: taking a 50m swimming distance as the minimum unit for determining the swimming posture, referring to the IMU sensor data worn on the hip, if the Z-axis acceleration average is close to +1g (9.81m / s 2 ), it is judged as backstroke; if the mean Z-axis acceleration is close to -1g (-9.81m / s 2), and the amplitude change of Y-axis acceleration is greater than that of X-axis and Z-axis, it is judged as freestyle; when backstroke and freestyle swimming styles are excluded; the change of acceleration time series is spectrally analyzed to obtain the center frequency of X-axis acceleration; if the center frequency of X-axis acceleration is much larger than the action frequency, it is judged as butterfly stroke, otherwise it is breaststroke, and the swimming style is obtained through the above steps; then the roll angle extracted from the real-time attitude angle of the IMU sensor of freestyle and backstroke, and the pitch angle extracted from the real-time attitude angle of the IMU sensor of butterfly stroke and breaststroke are used as the motion state; the swimming style and the corresponding motion state are used as swimming data.

[0021] The data processing module obtains a swimming motion symmetry evaluation result by subjecting the received swimming data to a data processing algorithm, including: identifying effective peak points and valley points of the swimming data through a filtering algorithm, and then calculating the stroke count, stroke frequency and stroke ratio based on the effective peak points and valley points.

[0022] The filtering algorithm includes two parts: coarse extraction and fine extraction. The swimming data is coarsely extracted to obtain multiple local extreme points. The multiple local extreme points are finely extracted to remove pseudo extreme points to obtain effective peak points and valley points. Then, the swimming motion symmetry evaluation result is obtained based on the effective peak points and valley points.

[0023] Compared with the prior art, the present invention has the following beneficial effects:

[0024] The present invention automatically extracts the symmetry evaluation results of swimming motion through a wearable inertial measurement unit. First, according to the time series, angular velocity information and acceleration information of the sensor, the motion state and swimming posture are obtained through a pattern recognition method. Then, the effective peak points and valley points are identified through a filtering algorithm designed by us, thereby obtaining symmetrical swimming data, and then outputting data related to the rotation parameters of the swimming motion. No human intervention is required throughout the process, and it can be directly provided for reference by coaches and trainers. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 is a block diagram of a system structure for automatically extracting swimming motion symmetry evaluation results based on an inertial sensor provided by an embodiment of the present invention;

[0026] Figure 2 is a schematic diagram of processing abnormal peaks in the algorithm provided in an embodiment of the present invention;

[0027] Figure 3 is a schematic diagram of processing abnormal troughs in the algorithm provided in an embodiment of the present invention;

[0028] Figure 4 It is a schematic diagram of processing abnormal swimming cycles in the algorithm provided in an embodiment of the present invention.

[0029] Figure 2-Figure 4In the equation, out1 and out2 refer to the peak point and valley point sequences respectively;

[0030] L refers to the length of the sequence, that is, the number of peak points or valley points

[0031] out1(n,1) represents the time of the nth peak point;

[0032] out2(m,2) represents the value of the mth valley point. DETAILED DESCRIPTION

[0033] In order to make the objectives, technical solutions and advantages of the present invention more clear, the embodiments disclosed in the present invention will be further described in detail below with reference to the accompanying drawings.

[0034] The present application embodiment discloses a method for evaluating swimming motion symmetry based on a lightweight IMU, referring to Figure 1 , Figure 2 , Figure 3 , Figure 4 .

[0035] Figure 1 is a structural block diagram of a system for automatically extracting swimming motion symmetry evaluation results based on an inertial sensor provided by an embodiment of the present invention. Figure 1 As shown, the system for automatically extracting the symmetry evaluation results of swimming motion based on inertial sensors includes: a wearable inertial measurement unit, a data transfer server and a host computer.

[0036] The host computer includes: a wearable inertial measurement unit, a data transfer server and a host computer; wherein the host computer includes a data receiving module, a data parsing module, a data processing module and a data display module; the wearable inertial measurement unit (IMU) is arranged on an athlete to monitor swimming movements, and is used to measure the acceleration and angular velocity information of the human shoulder, hip or chest position, and transmit the initial data composed of time series, acceleration and angular velocity information to the data receiving module through the data transfer server; the data receiving module transmits the initial data to the data parsing module; the data parsing module obtains swimming data from the received initial data through a parsing algorithm, and transmits the swimming data to the data processing module; the data processing module obtains a swimming movement symmetry evaluation result from the received swimming data through a data processing algorithm, and transmits the swimming movement symmetry evaluation result to the data display module; the data display module receives and displays the swimming movement symmetry evaluation result.

[0037] Wearable inertial measurement units are placed on athletes for a long time to monitor swimming movements. The forward direction of swimming movement is the X-axis, the right direction of the horizontal direction is the Y-axis, and the skyward direction is the Z-axis. They are used to measure the angular velocity and acceleration information of the human shoulder, hip or chest, and ensure that the measured data is stored in the internal storage unit. The wearable inertial measurement unit includes several IMU sensors. The IMU sensor can be sensitive to time series, acceleration and angular velocity information. Specifically, the wearable inertial measurement unit is composed of 1 or 2 IMU sensors. When using 1 IMU for measurement, it can be placed on the shoulder, hip or chest. When using 2 IMUs for measurement, one can be placed on the hip and the other can be placed on the shoulder or chest. The IMU sensor adopts a low-power design and has a built-in 1GB storage space to meet the long-term underwater training needs of athletes. The IMU sensor collects, saves and uploads the time series, acceleration and angular velocity information that the sensor is sensitive to.

[0038] The IMU sensor has a built-in data storage module, a WIFI communication module, and a rechargeable polymer battery. The IMU sensor has a built-in magnetometer that can measure magnetic field strength information. The data storage module is used to store initial data; the WIFI communication module cooperates with the data transfer server to transmit the initial data to the data receiving module; the rechargeable polymer battery is used to power the data storage module and the WIFI communication module.

[0039] The system for automatically extracting swimming motion symmetry evaluation results based on inertial sensors also includes: a data transfer server, which connects the wearable inertial measurement unit and the host computer to it, and establishes a wireless link via a WIFI signal through a network configuration, and uploads the time series, acceleration, and angular velocity information stored in the wearable inertial measurement unit to a data receiving module in the host computer through the wireless link.

[0040] The data receiving module transmits the data to the data analysis module, which obtains the swimming data from the random, noisy original sensor signal, and then obtains the swimming rotation parameters through the parameter extraction algorithm and displays them by the host computer.

[0041] The offline data of the sensor is transmitted wirelessly, each sensor is independent of each other, and the data can be stored at the same time to ensure data integrity. The specific implementation method is as follows:

[0042] The online reporting information is sent to the data transfer router through wireless WIFI communication, and then a stable and reliable wireless data transmission link is established with it. After the IMU sensor is turned on, it is in the state of waiting to connect to WiFi. After connecting to WiFi, the host computer displays the online status. After the IMU sensor receives the "offline storage start command" from the host computer, it starts measuring and replies in real time "offline storage in progress"; after the IMU sensor sends the "offline storage end command" to the host computer, the host computer displays the online status, and then the IMU sensor transmits the offline stored data to the host computer via WIFI; if the "offline storage end command" is not received but the data storage module is full at this time and cannot continue offline storage, the IMU sensor will actively send an "offline storage end reply frame" according to the preset protocol after sending the last frame of data. The IMU sensor has stopped storing, and the host computer displays the offline status. The user must clear the storage unit before continuing to use it normally.

[0043] The issuance of all instructions and uploading of measurement data are transmitted via wireless WIFI, which greatly reduces the complexity of equipment connection, wearing, and information transmission of the swimming measurement system using traditional video monitoring methods, and greatly improves the convenience of system application.

[0044] The data analysis module analyzes the time series, magnetic intensity information, angular velocity information and acceleration information, and obtains the motion state and swimming posture through the pattern recognition method. The specific implementation method is as follows:

[0045] The collected data is analyzed to determine whether the frame header checksum of each frame of data is correct and the frame sequence number is continuous without frame loss. If the frame header checksum is incorrect or the frame sequence number is discontinuous, an "incomplete data" prompt will be given, and the strapdown inertial navigation update algorithm will be used to perform strapdown inertial navigation to obtain the attitude angle; if the frame header checksum of each frame of data is correct and the frame sequence number is continuous without frame loss, strapdown inertial navigation will be performed to obtain the swimming data based on the attitude angle obtained by navigation.

[0046] Furthermore, the attitude angle of each IMU sensor (the carrier coordinate system relative to the navigation coordinate system) is calculated from the acceleration information and the magnetic field strength information, and then the initial attitude matrix from the carrier coordinate system to the navigation coordinate system is obtained to solve the problem of human body wearing and installation. After that, the new attitude angle is calculated by the continuously updated quaternion. The specific implementation method is as follows:

[0047] The strapdown inertial navigation system (SINS) selects the "east-north-sky" geographic coordinate system as the navigation coordinate system to implement the recursive update algorithm of the strapdown inertial navigation. The strapdown inertial navigation update algorithm is divided into three parts: attitude, velocity and position update, and the attitude update algorithm is the core.

[0048] The letters in the formula below refer to the following:

[0049] 1. Carrier coordinate system (b)

[0050] 2. Navigation coordinate system (n)

[0051] 3. Inertial coordinate system (i)

[0052] 4. Earth coordinate system (e)

[0053] The formula below explains:

[0054] Represents the projection of the angular velocity of the b system relative to the n system in the b system, similarly;

[0055] Represents the component of the angular velocity of the b system relative to the n system projected on the X axis in the b system, similarly;

[0056] From the angular velocity equation we get

[0057]

[0058] Since the MEMS sensor has low accuracy and cannot be sensitive to the angular velocity of the earth's rotation, it is ignored. At the same time, the speed of a person in a swimming scene is less than 2m / s, the radius of the earth R is 6371393m, and so For 10 -7 ~10 -6 Therefore, for MEMS sensors, the above formula can be equivalent to:

[0059]

[0060] and Q([q0,q1,q2,q3]) are the attitude matrix and attitude quaternion, and the initial values ​​are the initial attitude angles θ, γ obtained by the initial alignment. Calculated, obtained from the initial attitude matrix, and then calculated from the continuously updated quaternion. After that, the quaternion update equation is as follows:

[0061]

[0062] Calculated by Q The formula is as follows:

[0063]

[0064] According to the updated The attitude angle is calculated to obtain the three new attitude angles θ, γ Thus, the data of the three posture angles during the whole swimming process are obtained.

[0065] Furthermore, the flight time is determined as the starting point of the swimming exercise, and the pitch angle θ exceeding 90° after the start of the swimming exercise is determined as the end point of the swimming exercise.

[0066] Furthermore, according to the swimming stage data, the Fourier transform is performed to obtain the frequency value f corresponding to the maximum value in the power spectrum density. As the sampling interval, the MATLAB built-in function “islocalmax” is used to obtain the peak and valley values ​​of a section of the data, and then the pseudo peak and pseudo valley values ​​in the turning stage are removed.

[0067] Furthermore, taking a 50m swimming distance as the minimum unit for determining swimming style, referring to the data of the IMU sensor worn on the hip, if the average Z-axis acceleration is close to +1g (9.81m / s 2 ), it is judged as backstroke; if the mean Z-axis acceleration is close to -1g (-9.81m / s 2 ), and the amplitude change of the Y-axis acceleration is greater than that of the other two axes, it is judged as freestyle; when the above two swimming styles are excluded, if the central frequency of the X-axis acceleration is much greater than the action frequency, it is judged as butterfly stroke, otherwise it is breaststroke; then the roll angle γ of freestyle and backstroke, and the pitch angle θ of butterfly stroke and breaststroke are extracted.

[0068] The data processing module uses an algorithm to identify effective peak points and valley points of the swimming data, thereby obtaining symmetrical swimming data, and then obtaining swimming parameters such as rotation angle, rotation time, rotation speed, rotation symmetry index, etc. The specific implementation method is as follows:

[0069] The effective peak points and valley points include: multiple roll angles or pitch angles in the time series corresponding to the swimming styles, such as: multiple roll angles in the time series corresponding to freestyle, multiple roll angles in the time series corresponding to backstroke, multiple pitch angles in the time series corresponding to butterfly stroke, or multiple pitch angles in the time series corresponding to breaststroke.

[0070] For the effective peak points and valley points, the first type of pseudo extreme point in each lap of swimming data is first removed, that is, there is no valley point between two peak points or there is no peak point between two valley points. The algorithm flow is shown in Figure 2 and Figure 3 , and then remove the second type of pseudo extreme point in the turning stage. The judgment method is that if the difference between the peak point and the valley point in the same stroke cycle is more than 0.7 stroke cycles, the smaller time point is judged as the second type of pseudo extreme point. The algorithm flow chart is shown in Figure 4 , thus symmetrical swimming data can be obtained.

[0071] Figure 2 , Figure 3 and Figure 4 In the equation, out1 and out2 refer to the peak point and valley point sequences respectively;

[0072] L refers to the length of the sequence, that is, the number of peak points or valley points

[0073] out1(n,1) represents the time of the nth peak point;

[0074] out2(m,2) represents the value of the mth valley point.

[0075] by Figure 2 For example, it includes:

[0076] S1: Take the minimum value of the number of peak points in the peak point sequence out1 and the number of valley points in the valley point sequence out2 to get the L value;

[0077] S2: Starting from i=1, determine whether the time of the i+1th peak point in the peak point sequence out1 is less than the time of the ith peak point in the valley point sequence out2. If the determination result is yes, proceed to step S3; if the determination result is no, proceed to step S4;

[0078] S3: Delete the i-th peak point data in the peak point sequence out1, update the L value, and then proceed to step S2;

[0079] S4: Set the value of i to +1; then determine the relative size of i+1 and L. If i+1 < L, proceed to step S2, otherwise end.

[0080] Furthermore, if the time difference between two adjacent peak points is more than 1.5 average peak periods, it is determined to be the rotation time. The rotation time is not taken into account when calculating the rotation speed, and only the data during the swimming movement is counted. The specific implementation method is as follows:

[0081] The data extracted for freestyle swimming is the roll angle, whose peak and valley points represent the maximum amplitude of the athlete's left and right rotation. The difference between the peak and valley points in the same cycle is the total amplitude difference within the cycle. Then, according to the time point when the rotation from left to right reaches the peak, the difference is obtained to obtain the time required for the left and right body rotation to reach the peak and the cycle rotation time. It is also necessary to remove the turning time period by checking the time interval of each cycle. The left and right angular velocity values ​​and rotation frequency are calculated from the total amplitude difference within the cycle and the cycle rotation time. Finally, the rotation angle ratio is obtained from the maximum amplitude of the left and right rotations of the dominant side (left side) and the non-dominant side (right side) in each rotation cycle, and then the rotation symmetry index is obtained. The same is true for backstroke.

[0082] Similar to the analysis of roll angle in freestyle, the data extracted for breaststroke is the pitch angle, whose peak and valley points represent the maximum amplitude of the athlete's up and down flipping. The difference between the peak and valley points in the same cycle is the total amplitude difference within the cycle. Then, according to the time point when the up and down flipping reaches the peak, the difference is obtained to obtain the time required for the up and down flipping to reach the peak and the cycle flipping time. It is also necessary to remove the flipping time period by checking the time interval of each cycle. Then, the up and down angular velocity values ​​and rotation frequency are calculated from the total amplitude difference within the cycle and the cycle flipping time. Finally, the phase angle is obtained by the difference in the maximum amplitude of the shoulder and hip in each rotation cycle. The same is true for butterfly stroke.

[0083] Finally, swimming parameters such as swimming style, rotation angle, rotation time, rotation speed, rotation symmetry index, etc. are displayed on the host computer. No human intervention is required throughout the process and it can be directly provided for reference by coaches and trainers.

[0084] The present invention solves the problem that traditional video monitoring methods have strict environmental requirements. According to the time series, angular velocity and acceleration information of the inertial wearable device, the motion state and swimming posture are obtained through pattern recognition, and then the effective peak and valley value points are identified through a filtering algorithm, thereby obtaining the swimming motion symmetry evaluation result.

[0085] The wearable inertial measurement unit of the present invention adopts a streamlined design and is highly integrated to be light and small. One to two can be worn on the shoulders, hips or chest of the human body as needed, without affecting the movement of swimmers. At the same time, it adopts a professional-grade waterproof design to have IPX8 high waterproof performance, a low-power design and a built-in 1GB storage space to meet the different needs of athletes for long-term underwater training.

[0086] The present invention automatically extracts the symmetry evaluation results of swimming motion through a wearable inertial measurement unit. Firstly, the motion state and swimming posture are obtained through a pattern recognition method according to the time series, angular velocity information and acceleration information of the sensor. Then, the effective peak points and valley points are identified through the filtering algorithm designed by us, thereby obtaining symmetrical swimming data, and then outputting the symmetry evaluation results of swimming motion. No human intervention is required in the whole process, and it can be directly provided for reference by coaches and trainers.

[0087] The contents not described in detail in this application specification belong to the common knowledge of those skilled in the art.

[0088] The present application is described in detail above in conjunction with specific implementation methods and exemplary examples, but these descriptions cannot be understood as limiting the present application. Those skilled in the art understand that, without departing from the spirit and scope of the present application, a variety of equivalent replacements, modifications or improvements can be made to the technical solution of the present application and its implementation methods, all of which fall within the scope of the present application. The scope of protection of the present application shall be subject to the attached claims.

Claims

1. A system for evaluating swimming motion symmetry based on a lightweight IMU, characterized by: It includes a wearable inertial measurement unit, a data transfer server and a host computer; wherein the host computer includes a data receiving module, a data parsing module, a data processing module and a data display module; Wearable inertial measurement unit, which is installed on the athlete to monitor swimming movements, is used to measure the acceleration and angular velocity information of the human shoulder, hip or chest, and send the initial data consisting of time series, acceleration and angular velocity information to the data transfer server; A data transfer server, used for receiving the initial data and transmitting the initial data to the data receiving module; A data receiving module, receiving initial data and transmitting the initial data to the data parsing module; A data analysis module receives initial data, obtains swimming data by analyzing the received initial data through an algorithm, and transmits the swimming data to the data processing module; A data processing module receives swimming data, processes the received swimming data through a data processing algorithm to obtain a swimming motion symmetry evaluation result, and transmits the swimming motion symmetry evaluation result to the data display module; A data display module receives and displays the swimming motion symmetry evaluation result.

2. A system for evaluating swimming motion symmetry based on a lightweight IMU according to claim 1, characterized in that: The wearable inertial measurement unit includes at least one IMU sensor, which is used to measure acceleration, angular velocity information and magnetic intensity information, and obtain corresponding time series; acceleration includes X-axis acceleration, Y-axis acceleration and Z-axis acceleration, angular velocity information includes X-axis angular velocity, Y-axis angular velocity and Z-axis angular velocity, and magnetic intensity information includes X-axis magnetic intensity, Y-axis magnetic intensity and Z-axis magnetic intensity.

3. A system for evaluating swimming motion symmetry based on a lightweight IMU according to claim 1, characterized in that: The IMU sensor has a built-in data storage module, a WIFI communication module and a repeatedly rechargeable polymer physical battery. The data storage module is used to store initial data; the WIFI communication module cooperates with the data transfer server to transmit the initial data to the data receiving module; the repeatedly rechargeable polymer physical battery is used to power the data storage module and the WIFI communication module.

4. A system for evaluating swimming motion symmetry based on a lightweight IMU according to claim 1, characterized in that: After the IMU sensor is turned on, it is in a state of waiting to connect to WiFi. After connecting to WiFi, the host computer displays an online state. After the IMU sensor receives an "offline storage start command" from the host computer, the IMU sensor starts measuring and replies in real time with "offline storage in progress". After the IMU sensor sends an "offline storage end command" to the host computer through the WIFI communication module, the host computer displays an online state, and then the IMU sensor transmits the offline stored data to the host computer through the WIFI communication module. If the "offline storage end command" is not received but the data storage module is full at this time and offline storage cannot continue, the IMU sensor actively sends an "offline storage end reply frame" according to the preset protocol after sending the last frame of data, and the IMU sensor has stopped storing, and the host computer displays an offline state.

5. A system for evaluating swimming motion symmetry based on a lightweight IMU according to claim 1, characterized in that: The swimming data includes movement state and swimming posture; the swimming movement symmetry evaluation result includes rotation angle, rotation time, rotation speed and rotation symmetry index.

6. A system for evaluating swimming motion symmetry based on a lightweight IMU according to claim 1, characterized in that: The data analysis module analyzes the received initial data to determine whether the frame header checksum of each frame of data is correct, the frame sequence number is continuous and there is no frame loss; If the frame header checksum is incorrect or the frame sequence number is discontinuous, the "data is incomplete" prompt will be given, and the strapdown inertial navigation update algorithm will be used to perform strapdown inertial navigation to obtain the attitude angle; if the frame header checksum of each frame of data is correct and the frame sequence number is continuous without frame loss, the strapdown inertial navigation will be performed to obtain the attitude angle; The swimming data is obtained according to the attitude angle obtained by strapdown inertial navigation.

7. A system for evaluating swimming motion symmetry based on a lightweight IMU according to claim 6, characterized in that: In the strapdown inertial navigation, the initial attitude angle of the IMU sensor is obtained according to the acceleration and magnetic field strength information at the time of power-on, and then the initial attitude matrix from the carrier coordinate system to the navigation coordinate system is obtained; then the quaternion is continuously updated according to the time series, acceleration and angular velocity information and magnetic field strength information measured by the wearable inertial measurement unit, and then the real-time attitude angle of the IMU sensor is obtained through quaternion calculation.

8. A system for evaluating swimming motion symmetry based on a lightweight IMU according to claim 1, characterized in that: The data analysis module obtains swimming data from the received initial data through an analysis algorithm, including: taking a 50m swimming distance as the minimum unit for determining the swimming posture, referring to the IMU sensor data worn on the hip, if the mean value of the Z-axis acceleration is close to +1g, it is judged as backstroke; if the mean value of the Z-axis acceleration is close to -1g, and the amplitude change of the Y-axis acceleration is greater than that of the X-axis and Z-axis, it is judged as freestyle; when the backstroke and freestyle swimming postures are excluded; the change of the acceleration year time series is subjected to spectrum analysis to obtain the center frequency of the X-axis acceleration; if the center frequency of the X-axis acceleration is much greater than the action frequency, it is judged as butterfly stroke, otherwise it is breaststroke, and the swimming posture is obtained through the above steps; then the roll angle extracted from the real-time attitude angle of the IMU sensor of freestyle and backstroke, and the pitch angle extracted from the real-time attitude angle of the IMU sensor of butterfly stroke and breaststroke are used as the motion state; the swimming posture and the corresponding motion state are used as the swimming data.

9. A system for evaluating swimming motion symmetry based on a lightweight IMU according to claim 5, characterized in that: The data processing module obtains a swimming motion symmetry evaluation result by subjecting the received swimming data to a data processing algorithm, including: identifying effective peak points and valley points of the swimming data through a filtering algorithm, and then calculating the stroke count, stroke frequency and stroke ratio based on the effective peak points and valley points.

10. A system for evaluating swimming motion symmetry based on a lightweight IMU according to claim 9, characterized in that: The filtering algorithm includes two parts: coarse extraction and fine extraction. The swimming data is coarsely extracted to obtain multiple local extreme points. The multiple local extreme points are finely extracted to remove pseudo extreme points to obtain effective peak points and valley points. Then, the swimming motion symmetry evaluation result is obtained based on the effective peak points and valley points.

Images