Swimming posture display system
By using wearable devices with inertial measurement units and water pressure sensors in swimming, combined with inertial motion capture technology, the problems of swimming posture capture accuracy and complexity are solved, and low-cost and easy-to-deploy multi-user posture capture is achieved, which is suitable for mass fitness and training scenarios.
Patent Information
- Application Number
- CN202511042414.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-28
- Publication Date
- 2025-09-16
AI Technical Summary
Existing optical motion capture systems have difficulty ensuring accuracy during swimming, and their deployment is complex and costly, making it difficult to capture the motion postures of multiple users simultaneously.
Wearable devices using inertial measurement units and water pressure sensors, combined with inertial motion capture technology, dynamically display swimming postures through angular velocity, acceleration and water pressure data, achieve the acquisition of joint posture information and dynamic coupling with the pool coordinate system, and use the wall correction mechanism to eliminate the accumulated error of horizontal position.
It achieves millimeter-level joint angle restoration and centimeter-level position trajectory tracking of swimming postures. The system is simple, fast, and low-cost to deploy. It is suitable for multi-user posture capture and is applicable to mass fitness and training scenarios.
Smart Images
Figure CN120643884A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of sports assistance technology, and more particularly to a swimming posture display system. Background Art
[0002] As people increasingly prioritize health, they are increasingly enthusiastic about participating in sports. However, incorrect posture during exercise can lead to poor results and potentially even injuries. Therefore, a solution is needed to record and display users' exercise posture, making it easier for users to correct their own posture or for coaches to provide oversight and guidance.
[0003] At present, the common solution is to capture motion postures through an optical motion capture system. The inventors found that this solution is applicable to most sports, but not very applicable to swimming. The main reason is that the water in the swimming pool will refract, scatter, and attenuate the light, and splashes and bubbles will inevitably be generated during swimming, resulting in imaging position offset. The accuracy of using conventional optical motion capture systems cannot be guaranteed. If additional designs such as customized lens design and image optimization processing are performed for underwater environments, the complexity and cost of the system will be greatly increased. In addition to the difficulty in meeting the accuracy requirements, the solution of capturing swimming postures through an optical motion capture system also has problems such as high deployment complexity, long deployment time, and difficulty in capturing multiple users' motion postures at the same time. Summary of the Invention
[0004] The purpose of the present disclosure is to provide a swimming posture display system to solve at least one of the problems existing in the prior art.
[0005] To achieve the above objectives, the present disclosure adopts the following technical solutions: The present disclosure provides a swimming posture display system, comprising a terminal device and a plurality of wearable devices for being worn on the joints of a user, wherein the wearable devices are provided with an inertial measurement unit and a water pressure sensor; The terminal device is configured as follows: receiving the angular velocity data and acceleration data collected by the inertial measurement unit and the water pressure data collected by the water pressure sensor; Determine the pool depth position of the corresponding joint at each acquisition moment based on the water pressure data; Determine the attitude angle of the joint corresponding to each acquisition moment according to the angular velocity data and the acceleration data, wherein the attitude angle includes the roll angle, the pitch angle and the yaw angle; Determine the moment when the user touches the wall based on the joint acceleration data and the pool depth direction position; determine the pool length direction position and pool width direction position of the corresponding joint at the current collection moment based on the previous user touch-the-wall moment, the pool length acceleration data, the pool width acceleration data, and the yaw angle at the current collection moment, and obtain the pool length direction position of each joint at each collection moment; The user's swimming posture is dynamically displayed based on the posture angle of each joint at each collection moment, the length position, width position and depth position of the pool.
[0006] Optionally, the wearable device includes a first wearable device for wearing on the user's head, a second wearable device for wearing on the user's waist, and a third wearable device for wearing on the user's limbs.
[0007] Optionally, determining the swimming pool depth position of the joint corresponding to each acquisition moment according to the water pressure data includes: The water pressure in-situ value at the time of entering the water is determined according to the change of the water pressure data collected by the water pressure sensor of the first wearable device, and the depth direction position of the joint corresponding to each collection time is determined according to the difference between the water pressure data and the water pressure in-situ value.
[0008] Optionally, determining the swimming pool depth position of the joint corresponding to each acquisition moment according to the water pressure data includes: The depth position of the joint corresponding to each collection moment is determined based on the water pressure data, and the depth change of the joint corresponding to each collection moment is determined based on the acceleration data in the depth direction of the swimming pool. The depth position of the joint corresponding to each collection moment and the depth change of the joint corresponding to each collection moment are fused by Kalman filtering to obtain the depth direction position of the joint corresponding to each collection moment at the swimming pool.
[0009] Optionally, determining the moment when the user touches the wall based on the joint acceleration data and the position in the depth direction of the swimming pool includes: The moment when the user touches the wall is determined based on the joint acceleration data, the pool depth position and the angular velocity data.
[0010] Optionally, determining the moment when the user touches the wall based on the acceleration data of the joint, the position in the depth direction of the swimming pool, and the angular velocity data includes: If the acceleration data of the joint at the collection moment is greater than a first preset threshold, the position in the depth direction of the swimming pool is less than a second preset threshold, and the angular velocity data is greater than a third preset threshold, the collection moment is determined as the user touching the wall moment.
[0011] Optionally, determining the lengthwise position of the joint corresponding to the current acquisition moment includes: The pool length position correction value at the current collection moment is determined based on the pool length position of the corresponding joint at the previous collection moment. The pool length position of the corresponding joint at the current collection moment is determined based on the pool length position correction value at the current collection moment, the previous user wall touch moment at the current collection moment, the pool length acceleration data, the pool width acceleration data and the yaw angle.
[0012] Optionally, determining the posture angle of the joint corresponding to each acquisition moment according to the angular velocity data and the acceleration data includes: determining the posture angle of the joint corresponding to each acquisition moment according to the angular velocity data and the acceleration data based on a quaternion method.
[0013] Optionally, the terminal device is further configured to: When the pitch angle of the joint corresponding to the current acquisition moment is greater than the fourth preset threshold, the quaternion data of the joint corresponding to the current acquisition moment and the quaternion data of the joint corresponding to the previous acquisition moment are interpolated to obtain transition state quaternion data, and the posture angle of the joint corresponding to the current acquisition moment is determined based on the transition state quaternion data.
[0014] Optionally, the terminal device is further configured to communicate with each wearable device based on a frequency division multiplexing communication mode.
[0015] The beneficial effects of the present disclosure are as follows: The present disclosure provides a swimming posture display system that uses an inertial motion capture system optimized for swimming to acquire user joint posture information during swimming and dynamically couple joint positions with a pool coordinate system. The system collects user joint posture information based on distributed wearable device deployment, determines the pool depth position of the joint based on water pressure data collected by distributed water pressure sensors, determines the horizontal position of the joint, including both the pool length and pool width positions, based on pool length acceleration data, pool width acceleration data, and yaw angle, and eliminates accumulated errors in the horizontal position based on a wall contact correction mechanism. For example, an adaptive coordinate system can be constructed by integrating acceleration kinematics with the wall contact correction mechanism, eliminating the need for pool coordinate system calibration. This system can achieve millimeter-level joint angle restoration and centimeter-level position trajectory tracking of the user's swimming movements, thereby accurately displaying swimming postures in three dimensions.
[0016] Furthermore, the deployment of the swimming posture display system provided by the present disclosure requires only the user wearing a wearable device, without requiring modifications to the swimming pool, such as pre-embedded beacons. The deployment method is simple and fast, and the wearable device's acquisition components only require a six-axis inertial measurement unit and a water pressure sensor. The user's smartphone, tablet, and other electronic devices can directly serve as terminal devices, resulting in a low system cost. Furthermore, through wireless communication connection settings, the wearable devices of multiple users in a swimming pool can communicate with their respective terminal devices to simultaneously capture the movement postures of multiple users. Alternatively, a single terminal device can communicate separately with the wearable devices of multiple users to simultaneously capture the movement postures of multiple users.
[0017] Based on the above advantages, the swimming posture display system provided by the present invention is suitable for public fitness scenarios and training scenarios, providing a strong basis for users to correct their swimming posture or coaches to provide supervision and guidance, and has broad market application prospects. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] The specific embodiments of the present disclosure will be further described in detail below with reference to the accompanying drawings.
[0019] Figure 1 A schematic diagram showing a swimming posture display system provided by an embodiment of the present disclosure.
[0020] Figure 2 Shown is a top view of the swimming pool.
[0021] Figure 3 Shown is a side view of the swimming pool. DETAILED DESCRIPTION
[0022] To more clearly illustrate the present disclosure, the present disclosure is further described below in conjunction with the embodiments and drawings. Similar components in the drawings are represented by the same reference numerals. Those skilled in the art should understand that the specific description below is illustrative rather than restrictive and should not be used to limit the scope of protection of the present disclosure.
[0023] In the description of the present disclosure, it should be noted that the orientation or positional relationship indicated by the terms "upper" and "lower" is based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing the present disclosure and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present disclosure. Unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be internal communication between two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present disclosure can be understood according to the specific circumstances.
[0024] It should also be noted that, in the description of the present disclosure, relational terms such as first and second, etc., are merely used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further restrictions, an element defined by the statement "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.
[0025] The embodiment of the present disclosure provides a swimming posture display system, comprising a terminal device and a plurality of wearable devices for being worn on the joints of a user, wherein the wearable devices are provided with an inertial measurement unit and a water pressure sensor; The terminal device is configured as follows: receiving the angular velocity data and acceleration data collected by the inertial measurement unit and the water pressure data collected by the water pressure sensor; Determine the pool depth position of the corresponding joint at each acquisition moment based on the water pressure data; Determine the attitude angle of the joint corresponding to each acquisition moment according to the angular velocity data and the acceleration data, wherein the attitude angle includes the roll angle, the pitch angle and the yaw angle; Determine the moment when the user touches the wall based on the joint acceleration data and the pool depth direction position; determine the pool length direction position and pool width direction position of the corresponding joint at the current collection moment based on the previous user touch-the-wall moment, the pool length acceleration data, the pool width acceleration data, and the yaw angle at the current collection moment, and obtain the pool length direction position of each joint at each collection moment; The user's swimming posture is dynamically displayed based on the posture angle of each joint at each collection moment, the length position, width position and depth position of the pool.
[0026] The swimming posture display system provided in this embodiment uses an inertial motion capture system optimized for swimming to acquire user joint posture information during swimming and dynamically couple joint positions with the pool coordinate system. The system collects user joint posture information based on distributed wearable device deployment. The joint's pool depth position is determined based on water pressure data collected by distributed water pressure sensors. The joint's horizontal position, including its pool length and pool width positions, is determined based on pool length acceleration data, pool width acceleration data, and yaw angle. Accumulated errors in horizontal position are eliminated using a wall correction mechanism. For example, an adaptive coordinate system can be constructed by integrating acceleration kinematics with the wall correction mechanism, eliminating the need for pool coordinate system calibration. This system can achieve millimeter-level joint angle restoration and centimeter-level position trajectory tracking for the user's swimming motion, enabling precise three-dimensional visualization of swimming postures.
[0027] In one possible implementation, the wearable device includes a first wearable device for wearing on the user's head, a second wearable device for wearing on the user's waist, and a third wearable device for wearing on the user's limbs. This allows for a comprehensive and dynamic display of the user's swimming postures, including the postures of the main parts of the body.
[0028] In a specific example, for example Figure 1 As shown, the swimming posture display system includes a first wearable device 101 for wearing on the user's head, a second wearable device 102 for wearing on the user's waist, four third wearable devices 103 for wearing on the user's limbs, and a terminal device 104. The four third wearable devices 103 are respectively used to be worn on the user's two wrists and two ankles.
[0029] Exemplarily, the first wearable device 101, the second wearable device 102 and the third wearable device 103 respectively have a shell made of silicone material, and the shells of the first wearable device 101, the second wearable device 102 and the third wearable device 103 are respectively detachably connected to a silicone fixing strap, and the user fixes the fixing strap at the corresponding joint part when wearing it.
[0030] Exemplarily, the housings of the first wearable device 101, the second wearable device 102, and the third wearable device 103 are respectively provided with a six-axis (also called six degrees of freedom, 6DoF) inertial measurement unit (IMU) composed of a three-axis accelerometer and a three-axis gyroscope, and are also respectively provided with an industrial-grade water pressure sensor with an accuracy of, for example, ±0.1 kPa. In addition, the housings of the first wearable device 101, the second wearable device 102, and the third wearable device 103 are also respectively provided with a wireless communication module for enabling communication between the inertial measurement unit and the water pressure sensor and the terminal device 104.
[0031] For example, after receiving the angular velocity data and acceleration data collected by the inertial measurement unit in each wearable device and the water pressure data collected by the water pressure sensor in real time, the terminal device 104 can immediately start calculating the posture angle of each joint, the length direction position, the width direction position and the depth direction position of the pool at each collection moment, that is, calculate while receiving the data, or wait until all the data of the swimming process is received before starting to calculate the posture angle of each joint, the length direction position, the width direction position and the depth direction position of the pool at each collection moment, for example Figure 2 and Figure 3 As shown, the length direction of the swimming pool is the X direction, the width direction of the swimming pool is the Y direction, and the depth direction of the swimming pool is the Z direction.
[0032] For example, the terminal device 104 can perform 3D rendering on the calculated posture angles, length-direction positions, width-direction positions, and depth-direction positions of each joint at each collection moment to dynamically display the user's posture during swimming, or it can wait until the posture angles, length-direction positions, width-direction positions, and depth-direction positions of each joint at all collection moments are calculated before performing 3D rendering to dynamically display the user's posture during swimming.
[0033] In one possible implementation, the terminal device is further configured to communicate with each wearable device based on a frequency division multiplexing communication mode, thereby ensuring the real-time and reliability of communication between the multiple wearable devices and the terminal device.
[0034] Exemplarily, spatial synchronization in this embodiment utilizes frequency division multiple access (FDMA) technology to achieve data collaboration between multiple wearable devices and a terminal device. A dynamic logical ID allocation mechanism is employed to generate a configurable logical ID (1-255) based on the device MAC address. The frequency mapping range is 17.8kHz-22kHz, with a total of 220 frequency points in 0.2kHz increments. The frequency point (logical ID) of a wearable device is calculated as 17800 + 200 * (MAC-derived ID - 1), expressed in Hz. Furthermore, for at least some wearable devices, such as a first wearable device designed to be worn on the user's head and a second wearable device designed to be worn on the user's waist, interference adaptation using a frequency hopping spread spectrum (FHSS) mode triggered by real-time noise detection can be implemented. For example, when a signal strength RSSI (RSSI) is detected to be greater than -70dBm, Wi-Fi Direct mode communication with a transmission rate of 5MB / s is adopted, while when a signal strength RSSI (RSSI) is detected to be ≤-70dBm, Hotspot mode communication with a transmission rate of 1.2MB / s is adopted.
[0035] In a possible implementation, determining the posture angle of the joint corresponding to each acquisition moment according to the angular velocity data and the acceleration data includes: determining the posture angle of the joint corresponding to each acquisition moment according to the angular velocity data and the acceleration data based on a quaternion method.
[0036] Continuing with the previous example, the three-axis accelerometer of the six-axis inertial measurement unit can collect the linear acceleration value of the corresponding joint in three-dimensional space, and the three-axis gyroscope can collect the angular velocity value of the corresponding joint in three-dimensional space. By combining the angular velocity data and the acceleration data, the posture of the joint in three-dimensional space can be solved. For example, the angular velocity data and acceleration data can be mapped to the posture angle of the joint based on the standard human body model in the BVH format. Among them, BVH (BiovisionHierarchy) is a file format for representing motion capture data, which is used to describe the hierarchical relationship of the human body's skeletal structure and movement. The BVH file contains the bone hierarchy, the rotation angle of the joint, and the time information of the action frame. The joints are defined as nodes in the file and are connected through the parent-child relationship of the hierarchy.
[0037] For example, based on the quaternion method, the process of determining the posture angle of the joint corresponding to each acquisition moment according to the angular velocity data and acceleration data is as follows: Solve the quaternion q=[q0,q1,q2,q3] T : (1) Solve the attitude angle through the attitude transformation matrix: (2) in, 、 、 is the three-axis angular velocity collected by the three-axis gyroscope, is the time difference of the collection time, is the roll angle of the corresponding joint, is the pitch angle of the corresponding joint, is the yaw angle and roll angle of the corresponding joint is the angle of rotation around the length of the pool (X direction), pitch angle The yaw angle is the angle of rotation around the width of the pool (Y direction). The angle of rotation around the depth direction (Z direction) of the pool.
[0038] Based on the above formulas (1) and (2), if the quaternion at the previous acquisition moment is obtained, the quaternion at the current acquisition moment can be updated by combining the angular velocity data at the current acquisition moment through the iterative method of formula (1), thereby gradually updating the quaternion at each acquisition moment. That is to say, at the initial moment, an initial value of the quaternion is given (for example, defining the zero moment or giving q0=1, q1=0, q2=0, q3=0 at the initial moment), and the angular velocity continuously measured by the three-axis gyroscope is used to update the quaternion according to formula (1). The attitude angle can be calculated according to formula (2) through the quaternion, that is, the continuous update of the joint attitude is achieved by the continuous update of the quaternion. However, the attitude angle obtained in this way will have a large error, because the angular velocity measured by the gyroscope will have a deviation, resulting in a deviation in the final angle. Therefore, it is necessary to use the angle information obtained from the three-axis acceleration data collected by the three-axis accelerometer to correct the deviation to eliminate the deviation, specifically including: First, the three-axis acceleration data (i.e., gravity vector) collected by the three-axis accelerometer is obtained and normalized to obtain a normalized value of the gravity vector obtained based on the three-axis accelerometer; Then, the gravity component in the attitude transformation matrix obtained based on the three-axis gyroscope is obtained; Then, the normalized value of the gravity vector obtained based on the three-axis accelerometer is cross-multiplied with the gravity component in the attitude transformation matrix obtained based on the three-axis gyroscope to obtain the attitude error vector; Then, the attitude error vector is integrated to obtain the error integration result; Then, the error integration result is input into the PID (Proportional Integral Derivative) control unit and added to the three-axis angular velocity collected by the three-axis gyroscope in the current attitude update to obtain a corrected angular velocity data. The corrected angular velocity data is used to update the quaternion to obtain the accurate attitude angle.
[0039] In a possible implementation, the terminal device is further configured to: When the pitch angle of the joint corresponding to the current acquisition moment is greater than a fourth preset threshold, the quaternion data corresponding to the joint at the current acquisition moment is interpolated with the quaternion data corresponding to the joint at the previous acquisition moment to obtain transitional quaternion data. The attitude angle of the joint corresponding to the current acquisition moment is determined based on the transitional quaternion data. This avoids singularities when solving the attitude angle using the quaternion method, preventing gimbal lock and computational failure.
[0040] Continuing with the previous example, during swimming motion capture, when the joint pitch angle approaches ±90°, a phenomenon similar to a universal joint deadlock may occur, resulting in calculation failure. This implementation solves this problem by using a singularity avoidance method based on quaternion space interpolation conversion. For example, if the fourth preset threshold is set to 85°, the critical state detection mechanism established is: Calculate the pitch angle of the joint corresponding to the current acquisition moment Post-judgment | Is |>85° true (i.e., has it entered the ±5° critical zone)? If so, use the spherical linear interpolation algorithm (Slerp) to interpolate the current quaternion with the quaternion at the previous acquisition time, take the interpolation midpoint as the transition state quaternion (the interpolation coefficient is, for example, 0.5), and calculate the posture angle of the corresponding joint at the current acquisition time based on the transition state quaternion.
[0041] In a possible implementation, determining the pool depth position of the joint corresponding to each acquisition moment according to the water pressure data includes: The water pressure in-situ value at the time of entering the water is determined according to the change of the water pressure data collected by the water pressure sensor of the first wearable device, and the depth direction position of the joint corresponding to each collection time is determined according to the difference between the water pressure data and the water pressure in-situ value.
[0042] Therefore, the depth position of each joint in the swimming pool can be obtained based on the water pressure sensor corresponding to each joint.
[0043] Continuing with the above example, the water pressure sensor in the first wearable device worn on the user's head will detect a sudden change in water pressure when the user enters the water, and the water pressure data at the time of the sudden change can be automatically recorded as the water pressure in-situ value P surface , represents the water surface reference, which serves as the origin or reference point of the depth direction (Z direction) of the swimming pool. In the subsequent collection moments, according to the water pressure data P collected by each water pressure sensor water(t) and water pressure in situ value P surface The difference between the two determines the pool depth position d of the corresponding joint at each acquisition moment z(t) , the calculation formula is, for example: d z(t) = k *(P water(t) -P surface), where k is the conversion coefficient, for example, k = 0.99 cm / mbar.
[0044] Based on the above, the reference point in the depth direction (Z direction) of the swimming pool is Z0=0.99cm / mbar*P surface In addition, the reference point in the length direction (X direction) of the swimming pool can be set on the pool wall at the starting end (or starting end) of the swimming pool, and the reference point in the width direction (Y direction) of the swimming pool can be set on the center line of the lane.
[0045] In a possible implementation, determining the pool depth position of the joint corresponding to each acquisition moment according to the water pressure data includes: The depth position of the joint corresponding to each collection moment is determined based on the water pressure data, and the depth change of the joint corresponding to each collection moment is determined based on the acceleration data in the depth direction of the swimming pool. The depth position of the joint corresponding to each collection moment and the depth change of the joint corresponding to each collection moment are fused by Kalman filtering to obtain the depth direction position of the joint corresponding to each collection moment at the swimming pool.
[0046] In this way, the accuracy of the acquired position of the joint in the depth direction of the swimming pool can be further improved.
[0047] Continuing with the above example, this implementation can obtain the relative height change of the corresponding joint by integrating the acceleration data in the depth direction of the swimming pool, and use this data and the joint depth position d obtained from the water pressure data of the corresponding joint to calculate the relative height change of the corresponding joint. z(t) Perform Kalman filter fusion to obtain the pool depth position of the corresponding joint, which can be expressed in the form of:
[0048] Among them, z(t) is the pool depth position of the corresponding joint at the current acquisition moment obtained by Kalman filter fusion, a z The acceleration data of the joint in the direction of the pool depth is g, which represents the acceleration due to gravity. For example, the acceleration data with a sampling frequency of 100 Hz can be integrated to ensure short-term accuracy. The joint position d in the direction of the pool depth can be obtained by using the water pressure data collected by the water pressure sensor at a set time interval (e.g., 5 seconds). z(t) To correct drift.
[0049] In one possible implementation, determining the moment when the user touches the wall based on the joint acceleration data and the position in the depth direction of the swimming pool includes: The moment when the user touches the wall is determined based on the joint acceleration data, the pool depth position and the angular velocity data.
[0050] In this way, the moment when the user touches the wall and flips during swimming can be determined more accurately to avoid misjudgment.
[0051] In one possible implementation, determining the moment when the user touches the wall based on the joint acceleration data, the pool depth position, and the angular velocity data includes: If the acceleration data of the joint at the collection moment is greater than a first preset threshold, the position in the depth direction of the swimming pool is less than a second preset threshold, and the angular velocity data is greater than a third preset threshold, the collection moment is determined as the user touching the wall moment.
[0052] Continuing with the above example, the first preset threshold is, for example, 5g, the second preset threshold is, for example, 0.3m, and the third preset threshold is, for example, 500° / s. That is, at the collection moment when the acceleration data is greater than 5g, the position z(t) in the depth direction of the swimming pool is less than 0.3m, and the angular velocity data is greater than 500° / s, a wall touch event is determined to have occurred, and the collection moment is determined to be the moment when the user touched the wall.
[0053] In a possible implementation, determining the lengthwise position of the pool corresponding to the joint at the current acquisition moment includes: The pool length position correction value at the current collection moment is determined based on the pool length position of the corresponding joint at the previous collection moment. The pool length position of the corresponding joint at the current collection moment is determined based on the pool length position correction value at the current collection moment, the previous user wall touch moment at the current collection moment, the pool length acceleration data, the pool width acceleration data and the yaw angle.
[0054] In this way, the accuracy of the acquired position in the length direction of the swimming pool can be further guaranteed.
[0055] Continuing with the previous example, the formulas for determining the position along the length and width of the pool are:
[0056] Where x(t) is the position of the joint in the direction of the length of the pool at the current acquisition time, and y(t) is the position of the joint in the direction of the width of the pool at the current acquisition time; x reset The correction value for the pool length position at the current acquisition moment. For example, the setting method is: the correction value is when the distance between the pool length position of the corresponding joint at the previous acquisition moment and the side wall of the pool starting point is greater than 70% of the pool length L (the ratio value is adjustable). x reset The value is the pool length L. The correction value is when the distance between the pool length position of the corresponding joint and the side wall of the pool starting point at the last acquisition moment is less than 30% of the pool length L (the ratio value is adjustable). x reset The value is 0; t reset Indicates the last time a user touched the wall before the current collection time; and Represents the acceleration data a in the length direction of the swimming pool x The integral value of And the acceleration data a in the width direction of the swimming pool y The integral value of ; is the yaw angle at the current acquisition moment.
[0057] For example, terminal device 104 may utilize a lightweight 3D rendering engine accelerated by WebGL to reduce performance requirements, enabling smooth 50fps playback on a smartphone with standard performance. For example, the 3D rendering engine, or 3D reconstruction engine, may integrate BVH joint angle depth color bands and trajectory heatmaps for WebGL rendering, using four-dimensional data for overlay rendering. The four-dimensional data is shown in Table 1.
[0058] Table 1
[0059] The following is an example of the hardware parameters of the wearable device in the swimming posture display system provided by this embodiment: the shell is made of medical-grade silicone material and must meet the IP68 waterproof level and chlorine corrosion resistance design, such as the use of nano-coating packaging technology. The wireless communication module uses a 2.4GHz / 900MHz dual-band WIFI chip, and the antenna uses a dual-band flexible PCB antenna. The inertial measurement unit uses a six-axis IMU with a sampling rate of 1000Hz and a range of ±16g. The accuracy of the water pressure sensor is ±0.1kPa. The battery uses a 200mAh lithium polymer battery with a magnetic contact charging interface.
[0060] The following two specific scenarios are used for illustration.
[0061] Scenario 1: Training Athletes wear wearable devices and train in a standard 50-meter swimming pool. The 3D motion model generated by the terminal device can be displayed to the coach for supervision and guidance. On the other hand, the artificial intelligence model can be used to analyze the action information such as the paddling hand entry angle (marked as 153° for the elbow joint vs. the standard value of 165°) frame by frame.
[0062] Scenario 2: Public Fitness Swimming enthusiasts wearing wearable devices swim in a short pool of 25 meters in length. The 3D motion model obtained by the terminal device can be directly displayed to the user to correct the swimming posture. The 3D motion model of the user's swimming posture can be superimposed on the standard motion model for more intuitive viewing. The artificial intelligence model can also be used to analyze the motion information and display prompts such as "Head position is too high (+15cm), it is recommended to lower the height to reduce resistance" in conjunction with the freeze-frame picture.
[0063] Obviously, the above embodiments of the present disclosure are merely examples for clearly illustrating the present disclosure, and are not intended to limit the implementation methods of the present disclosure. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is impossible to enumerate all the implementation methods here. Any obvious changes or modifications derived from the technical solution of the present disclosure are still within the scope of protection of the present disclosure.
Claims
1. A swimming posture display system, characterized in that: It includes a terminal device and a plurality of wearable devices for wearing on the joints of a user, wherein the wearable devices are provided with an inertial measurement unit and a water pressure sensor; The terminal device is configured as follows: receiving the angular velocity data and acceleration data collected by the inertial measurement unit and the water pressure data collected by the water pressure sensor; Determine the pool depth position of the corresponding joint at each acquisition moment based on the water pressure data; Determine the attitude angle of the joint corresponding to each acquisition moment according to the angular velocity data and the acceleration data, wherein the attitude angle includes the roll angle, the pitch angle and the yaw angle; Determine the moment when the user touches the wall based on the joint acceleration data and the pool depth direction position; determine the pool length direction position and pool width direction position of the corresponding joint at the current collection moment based on the previous user touch-the-wall moment, the pool length acceleration data, the pool width acceleration data, and the yaw angle at the current collection moment, and obtain the pool length direction position of each joint at each collection moment; The user's swimming posture is dynamically displayed based on the posture angle of each joint at each collection moment, the length position, width position and depth position of the pool.
2. The swimming posture display system according to claim 1, characterized in that: The wearable device includes a first wearable device for wearing on the user's head, a second wearable device for wearing on the user's waist, and a third wearable device for wearing on the user's limbs.
3. The swimming posture display system according to claim 2, characterized in that: Determining the pool depth position of the joint corresponding to each acquisition moment according to the water pressure data includes: The water pressure in-situ value at the time of entering the water is determined according to the change of the water pressure data collected by the water pressure sensor of the first wearable device, and the depth direction position of the joint corresponding to each collection time is determined according to the difference between the water pressure data and the water pressure in-situ value.
4. The swimming posture display system according to any one of claims 1 to 3, characterized in that: Determining the pool depth position of the joint corresponding to each acquisition moment according to the water pressure data includes: The depth position of the joint corresponding to each collection moment is determined based on the water pressure data, and the depth change of the joint corresponding to each collection moment is determined based on the acceleration data in the depth direction of the swimming pool. The depth position of the joint corresponding to each collection moment and the depth change of the joint corresponding to each collection moment are fused by Kalman filtering to obtain the depth direction position of the joint corresponding to each collection moment at the swimming pool.
5. The swimming posture display system according to claim 1, characterized in that: Determining the moment a user touches the wall based on joint acceleration data and pool depth position includes: The moment when the user touches the wall is determined based on the joint acceleration data, the pool depth position and the angular velocity data.
6. The swimming posture display system according to claim 5, characterized in that: Determining the moment when the user touches the wall based on the joint acceleration data, the pool depth position and the angular velocity data includes: If the acceleration data of the joint at the collection moment is greater than a first preset threshold, the position in the depth direction of the swimming pool is less than a second preset threshold, and the angular velocity data is greater than a third preset threshold, the collection moment is determined as the user touching the wall moment.
7. The swimming posture display system according to claim 1, characterized in that: Determining the lengthwise position of the joint at the current acquisition moment includes: The pool length position correction value at the current collection moment is determined based on the pool length position of the corresponding joint at the previous collection moment. The pool length position of the corresponding joint at the current collection moment is determined based on the pool length position correction value at the current collection moment, the previous user wall touch moment at the current collection moment, the pool length acceleration data, the pool width acceleration data and the yaw angle.
8. The swimming posture display system according to claim 1, characterized in that: Determining the posture angle of the joint corresponding to each acquisition moment according to the angular velocity data and the acceleration data includes: determining the posture angle of the joint corresponding to each acquisition moment according to the angular velocity data and the acceleration data based on a quaternion method.
9. The swimming posture display system according to claim 8, characterized in that: The terminal device is further configured as: When the pitch angle of the joint corresponding to the current acquisition moment is greater than the fourth preset threshold, the quaternion data of the joint corresponding to the current acquisition moment and the quaternion data of the joint corresponding to the previous acquisition moment are interpolated to obtain transition state quaternion data, and the posture angle of the joint corresponding to the current acquisition moment is determined based on the transition state quaternion data.
10. The swimming posture display system according to claim 1, characterized in that: The terminal device is further configured to communicate with each wearable device based on a frequency division multiplexing communication method.