A simulation training device for guiding team ball sports

By collecting individual dynamic data and analyzing panoramic 3D modules, accurate 3D training reports are generated, solving the problem of difficulty in quantifying team collaboration in group ball sports training. This enables data-driven tactical optimization and immersive guidance, thereby improving training effectiveness.

CN121623265BActive Publication Date: 2026-11-10XINJIANG NORMAL UNIVERSITY
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202512029095.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-12-30
Publication Date
2026-11-10
Estimated Expiration
2045-12-30

AI Technical Summary

Technical Problem

Existing technologies struggle to accurately quantify athletes' instantaneous dynamics and teamwork in team ball sports training, lack real-time guidance capabilities, and cannot solve the problems of traditional reliance on coach experience through digital simulation of training scenarios.

Method used

The individual dynamic acquisition module acquires speed dynamics and frequency, and combines them with the panoramic 3D acquisition module to analyze motion coordination information, generating accurate 3D training reports. These reports are then visualized through AR glasses or a virtual projection sand table, providing data-driven performance evaluation and tactical optimization guidance.

Benefits of technology

It enables precise assessment of individual athletes and teams, identifies positioning deficiencies and trajectory coordination time differences, provides immediate and intuitive training guidance, and improves training efficiency and quality.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121623265B_ABST
    Figure CN121623265B_ABST
Patent Text Reader

Abstract

This invention relates to the field of sports simulation training, specifically a simulation training device for guiding team ball sports. This invention precisely acquires speed dynamics and speed change frequency to generate accurate individual three-dimensional training reports, providing objective performance assessments for each athlete. The individual three-dimensional training reports are then jointly analyzed with movement coordination information. This not only identifies surface problems such as positioning defects and trajectory coordination time differences, but also deeply analyzes the essence of the problems, thus providing precise directions for tactical optimization. Finally, the analysis results of the tactical optimization direction are visualized through terminal output components such as AR glasses or virtual projection sandboxes in a three-dimensional overlay manner, realizing a structured training analysis report and immersive scene reproduction.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of sports simulation training, specifically to a simulation training device for guiding team ball sports. Background Technology

[0002] In team ball sports training, guidance mainly relies on the coach's on-site observation and personal experience. The coach observes the athletes' positioning, speed, and teamwork with the naked eye, and provides verbal guidance or tactical adjustments based on his own experience. The disadvantage of this model is that it is difficult to accurately quantify the instantaneous dynamics of each athlete, such as minute acceleration changes, high-frequency change of direction, and the precise spatiotemporal relationships between team members from a full-court perspective.

[0003] Currently, although there are solutions for motion analysis using motion capture or motion sensors, as disclosed in patent applications CN201510393885A and CN2020103819797, most focus on capturing individual ball movements, such as golf swings or table tennis ball trajectories, or using general-purpose motion sensors such as heart rate monitors. Therefore, there is a lack of a technical solution that can cover the analysis of collective collaboration parameters and immersive team simulation training scenarios. Similar fields only have basic team data acquisition devices (such as football heat map generators), but they lack real-time guidance functions and cannot effectively solve the problems of relying on coach experience and difficulty in quantifying team collaboration in traditional group training through digital simulation training scenarios. Therefore, a solution is proposed here. Summary of the Invention

[0004] This invention utilizes an individual dynamic acquisition module to precisely acquire speed dynamics and speed variation frequency, generating accurate individual 3D training reports. This provides each athlete with a data-driven, objective performance evaluation basis. The individual 3D training reports are then jointly analyzed with motion coordination information generated by a panoramic 3D acquisition module. This analysis not only identifies surface issues such as positioning defects and trajectory coordination time differences but also, by introducing individual dynamic information as a weighting factor, delves deeper into the essence of the problems, providing precise direction for tactical optimization. Finally, the analysis results of the tactical optimization direction are visualized through AR glasses or a virtual projection sand table, using a 3D overlay method. This achieves structured training analysis reports and immersive scene reproduction, thus proposing a simulation training device for guiding team ball sports.

[0005] The objective of this invention can be achieved through the following technical solution: a simulation training device for group ball sports, comprising a terminal sensing component, a terminal output component, and a cloud analysis device. The terminal sensing component is worn on the body of the athlete to sense the athlete's movement status. The cloud analysis device analyzes the sensory results of the athlete's movement status and transmits the analysis results back to the terminal output component. The terminal output component displays the analysis results through an output device.

[0006] It also includes an individual dynamic acquisition module, a panoramic 3D acquisition module, a single-unit dynamic analysis module, an ensemble analysis module, and an overlay output module;

[0007] The individual dynamic acquisition module collects individual dynamic information through the terminal sensing component and transmits it to the individual dynamic analysis module via wireless transmission.

[0008] The individual dynamic analysis module performs motion analysis based on individual dynamic information and generates a three-dimensional training report of the individual's movements.

[0009] The panoramic 3D acquisition module collects collective information of the motion scene to obtain motion coordination information;

[0010] The set analysis module acquires motion coordination information and three-dimensional training reports of individual movements, performs coordination degree analysis based on motion coordination information, and corrects the coordination degree analysis results through the three-dimensional training reports of individual movements to obtain an overall coordination assessment.

[0011] The overlay output module acquires a three-dimensional training report of individual actions and an overall coordination assessment, and outputs early warning reminders based on the three-dimensional training report and the overall coordination assessment.

[0012] In a preferred embodiment of the present invention, the individual dynamic information acquired by the individual dynamic acquisition module includes velocity dynamics and velocity variation frequency.

[0013] The method for the individual dynamic acquisition module to obtain the frequency of speed change is as follows: the acceleration and velocity fed back by the terminal sensing component are accumulated and recorded, and the real-time speed change of the individual is obtained through the motion model and recorded as speed dynamics;

[0014] The method for the individual dynamic acquisition module to obtain the velocity variation frequency is as follows: filter the acceleration fed back by the terminal sensing component, select the effective acceleration, divide the effective acceleration into linear acceleration and anti-linear acceleration, and respectively count the linear acceleration variation frequency and the anti-linear acceleration variation frequency as the velocity variation frequency.

[0015] As a preferred embodiment of the present invention, the method for filtering acceleration by the individual dynamic acquisition module is as follows: obtain a preset trigger threshold through the database, and then compare the obtained acceleration with the trigger threshold in real time. If the acceleration is greater than the set trigger threshold, it is recorded as effective acceleration; if the acceleration is not greater than the set trigger threshold, it is recorded as invalid acceleration.

[0016] The method for classifying effective acceleration by the individual dynamic acquisition module is as follows: obtain the direction of each acceleration, and compare the acceleration with the direction of the previous group of adjacent accelerations. If the angle between the directions is greater than a set threshold, the acceleration is recorded as an opposite acceleration; if the angle between the directions is less than the set threshold, the acceleration is recorded as linear acceleration.

[0017] In a preferred embodiment of the present invention, the step of the individual dynamic acquisition module to statistically analyze the linear acceleration variation frequency and the anti-linear acceleration variation frequency is as follows:

[0018] The individual dynamic acquisition module counts the time elapsed from the first group of accelerations to the last group of accelerations and records it as the total time. It also records the number of occurrences of counter-acceleration and linear acceleration as t, and calculates the counter-acceleration frequency and linear velocity frequency using formulas.

[0019] As a preferred embodiment of the present invention, the three-dimensional training report generated by the single-unit dynamic analysis module includes: peak velocity, average velocity, frequency of opposite velocity variation, and frequency of linear velocity variation.

[0020] Peak and average speeds are extracted based on speed dynamics.

[0021] In a preferred embodiment of the present invention, the motion coordination information generated by the panoramic three-dimensional acquisition module includes the real-time position of the personnel and the movement trajectory of the personnel.

[0022] The method for the panoramic 3D acquisition module to obtain real-time positions is as follows: the coordinate data of each athlete in the field is captured by the terminal sensing component, and then combined with the preset digital model of the field, the real-time position of each athlete is mapped into three-dimensional spatial coordinates, and a dynamic position distribution map is generated according to the time series.

[0023] The method for the panoramic 3D acquisition module to obtain the movement trajectory of personnel is as follows: based on the station position data within a continuous time period, the movement path of each person is reconstructed through a trajectory fitting algorithm, key movement nodes are marked, and the relative distance change trend between adjacent people is calculated.

[0024] In a preferred embodiment of the present invention, the set analysis module performs embedding analysis based on motion coordination information through a motion model, records the positional defects and trajectory coordination time differences between different personnel at each moment, and sums the positional defects and trajectory coordination time differences by assigning weights to obtain the coordination degree difference.

[0025] In a preferred embodiment of the present invention, the step of the set analysis module to obtain the overall coordination evaluation is as follows:

[0026] The frequency of each athlete's individual directional velocity change is spatiotemporally matched with the athlete's critical nodes of sudden stops and changes of direction in the movement trajectory to assess whether the individual's high-frequency change of direction actions are consistent with the needs of team tactical execution.

[0027] For moments with positioning defects, retrieve the speed dynamic data of the person moving at that position. If the individual is moving at high speed at that moment, it is determined that the positioning defect may be due to reasonable dynamic replacement; if the individual is moving at low speed or stationary, increase the negative evaluation weight of the positioning defect.

[0028] For trajectory coordination time difference, correlation analysis is performed with the peak speed and reaction speed of relevant athletes to identify whether the cause of coordination time difference is lag in consciousness or insufficient individual absolute speed ability.

[0029] Compared with the prior art, the beneficial effects of the present invention are:

[0030] 1. This invention uses an individual dynamic acquisition module to precisely acquire speed dynamics and speed variation frequency, and an individual dynamic analysis module to summarize the peak speed, average speed, and frequency of anisotropic and linear speed variation. This enables the generation of accurate individual three-dimensional training reports, overcoming the vague evaluations that rely on coach experience in the traditional approach, and providing each athlete with a data-driven and objective performance evaluation basis.

[0031] 2. This invention combines an individual's 3D training report with motion coordination information generated by a panoramic 3D acquisition module for joint analysis. It can not only identify surface problems such as positioning defects and trajectory coordination time differences, but also deeply analyze the essence of the problem by introducing individual dynamic information as a weighting factor. It can determine whether positioning defects are due to tactical misalignment or reasonable dynamic compensation, and can identify whether coordination time differences are due to lagging tactical awareness or insufficient individual speed ability, thus providing a precise direction for tactical optimization.

[0032] 3. In this invention, complex analysis results such as individual key data, personnel movement trajectory, coordination time difference warnings, and positioning defect areas are visualized in a three-dimensional overlay manner through terminal output components such as AR glasses or virtual projection sand table. This enables structured and immersive scene reproduction of training analysis reports, allowing athletes and coaches to understand training problems in real time and intuitively, effectively guiding them to adjust technical movements and tactical formations, and greatly improving the efficiency and quality of training. Attached Figure Description

[0033] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0034] Figure 1 This is a system block diagram of the present invention;

[0035] Figure 2 This is a system flowchart of the present invention. Detailed Implementation

[0036] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0037] Example 1: Please refer to Figure 1 - Figure 2 As shown, a collective ball sports guidance simulation training device includes a terminal sensing component, which is worn on the body of the athlete and obtains the athlete's precise movement state, including speed and acceleration during movement, through built-in inertial sensors, acceleration sensors, etc.

[0038] The terminal sensing component also includes a panoramic sensing device installed at the edge of the field, which acquires collective information of the athletes in the field, including their real-time location.

[0039] The terminal sensing component transmits the acquired motion status and collective information back to the cloud analysis device. The cloud analysis device performs comprehensive analysis through a preset program and provides an assessment of the individual's motion status and the coordination of the collective motion information based on the analysis results. This assessment is then fed back to the terminal output component via the network.

[0040] The terminal output component outputs the evaluation results through a display output device, such as a sports headset. All athletes wear sports headsets and communicate with the athletes through the coaching machine, enabling coaches to provide timely corrections and guidance, such as adjusting shooting angles and optimizing defensive formations. This solves the problems of difficult real-time communication between coaches and the difficulty in quantifying team collaboration in traditional group training. The coaching machine and the sports headsets worn by the athletes can conduct point-to-point or group communication, improving the flexibility of communication.

[0041] Example 2: Please refer to Figure 1 - Figure 2 As shown, a collective ball sports guidance simulation training device also includes an individual dynamic acquisition module, a panoramic 3D acquisition module, a single dynamic analysis module, a group analysis module, and an overlay output module. The individual dynamic acquisition module and the panoramic 3D acquisition module are connected to the terminal sensing component, while the single dynamic analysis module, the group analysis module, and the overlay output module are deployed on a cloud analysis device, and the overlay output module is connected to the terminal output component.

[0042] The individual dynamic acquisition module obtains individual dynamic information through the terminal sensing component. This individual dynamic information includes speed dynamics and speed change frequency. Specifically, the method by which the individual dynamic acquisition module obtains speed dynamics is as follows:

[0043] The individual dynamic acquisition module accumulates and records the acceleration and velocity fed back by the terminal sensing components, and obtains the real-time changes in the individual's velocity through a motion model, which is recorded as velocity dynamics;

[0044] The method for the individual dynamic acquisition module to obtain the speed change frequency is as follows: the individual dynamic acquisition module records the acceleration fed back by the terminal sensing component, obtains the preset trigger threshold through the database, and then compares the obtained acceleration with the trigger threshold in real time. If the acceleration is greater than the set trigger threshold, it is recorded as effective acceleration; if the acceleration is not greater than the set trigger threshold, it is recorded as invalid acceleration.

[0045] The individual dynamic acquisition module will effectively accelerate the statistical process and obtain the direction of each acceleration. The individual dynamic acquisition module will compare the acceleration with the direction of the previous group of adjacent accelerations. If the angle between the directions is greater than the set threshold, the acceleration will be recorded as an opposite acceleration. If the angle between the directions is less than the set threshold, the acceleration will be recorded as a linear acceleration.

[0046] The individual dynamic acquisition module counts the time elapsed from the first set of accelerations to the last set of accelerations, recording it as the total time T. The number of occurrences of counter-current acceleration and linear acceleration are recorded as Ay and At, respectively. The counter-current velocity variation frequency Fy is calculated using a formula. The frequency of linear velocity variation Ft is calculated using the formula. .

[0047] The individual dynamic analysis module acquires the collected individual dynamic information through the individual dynamic acquisition module, and analyzes it based on velocity dynamics to obtain the peak velocity and average velocity. The individual dynamic analysis module then combines the anisotropic velocity variation frequency and the linear velocity variation frequency in the velocity variation frequency to form a three-dimensional training report of the individual's actions.

[0048] The panoramic 3D acquisition module obtains collective information through the terminal sensing component and generates motion coordination information, which includes the real-time position and movement trajectory of the personnel. The method by which the panoramic 3D acquisition module obtains the real-time position is as follows:

[0049] The terminal sensing component captures the coordinate data of each athlete in the field in real time through multi-view camera units and positioning tags. The panoramic 3D acquisition module obtains the coordinate data and combines it with the preset digital model of the field to map the real-time position of each athlete into three-dimensional spatial coordinates and generate a dynamic position distribution map according to the time series.

[0050] The panoramic 3D acquisition module obtains the movement trajectory of people by: reconstructing the movement path of each person based on the station position data within a continuous time period, marking key movement nodes, and calculating the relative distance change trend between adjacent people.

[0051] The ensemble analysis module performs embedding analysis on the motion model based on the motion coordination information. It analyzes the positioning defects by measuring the real-time positions of personnel at different times, records the positioning defects at each time, and performs embedding analysis on the motion model by measuring the personnel movement trajectory to obtain the trajectory coordination time difference between different personnel. Finally, it sums the positioning defects and trajectory coordination time difference by assigning weights to obtain the coordination degree difference.

[0052] The ensemble analysis module simultaneously receives the 3D training report output by the individual dynamic analysis module and the motion coordination information generated by the panoramic 3D acquisition module, and performs joint analysis through the following steps:

[0053] Each athlete's individual 3D training report, including peak speed, average speed, frequency of directional speed variation, and frequency of linear speed variation, is coupled with their real-time role and position in the motion coordination information. Specifically, the individual's frequency of directional speed variation is spatiotemporally matched with the athlete's sudden stops and changes of direction key nodes in the movement trajectory to assess whether the individual's high-frequency change of direction actions are consistent with the needs of team tactical execution. At the same time, the individual's frequency of linear speed variation is compared with the team's overall formation movement during the advancement or retreat phases to analyze the contribution efficiency of the individual's acceleration ability to the team's overall rhythm.

[0054] Based on the difference in coordination degree, which includes positioning defects and trajectory coordination time difference, individual dynamic information is further introduced as a weighting factor. For moments with positioning defects, the speed dynamic data of the person moving at that position is retrieved. If the individual is in a high-speed movement state at that moment, it is determined that the positioning defect may be due to reasonable dynamic replacement or tactical maneuver. If the individual is in a low-speed or stationary state, the negative evaluation weight of the positioning defect is increased.

[0055] For trajectory coordination time difference, it is correlated with the peak speed and reaction speed of the relevant athletes to identify whether the cause of coordination time difference is due to tactical awareness lag or insufficient individual absolute speed ability, and to complete the performance consistency assessment.

[0056] The overlay output module is connected to the terminal output component. It receives the joint analysis results output by the ensemble analysis module and processes and outputs them as follows:

[0057] S1: In the panoramic 3D scene presented by the terminal output component (such as AR glasses, stereoscopic projection device), the key data of each athlete's individual 3D training report are superimposed on the model in real time. At the same time, the movement trajectory of the athlete is visualized using light flow or trajectory lines of different colors, and based on the analysis results of the trajectory coordination time difference, the coordination path with significant delay is highlighted and flashed as a warning.

[0058] S2: The station defect time identified by the set analysis module is backtracked and reproduced in the 3D scene, and the defect area is circled with virtual markers.

[0059] S3: Automatically generates structured training analysis reports that integrate individual three-dimensional training reports of all athletes, team coordination information, and team synergy effectiveness index and tactical execution consistency assessment conclusions derived from joint analysis steps.

[0060] Thresholds, preset values, preset ranges, etc. are set for result comparison and analysis to determine whether they are good or bad. The value of these thresholds is determined by a combination of large-scale model analysis of sample data and human experience. They can also be adjusted appropriately based on seasonal or common-sense influences.

[0061] Furthermore, the settings for weighting ratios, influence factors, etc., are based on the magnitude of each parameter's influence on the results. The specific values ​​are allocated to ultimately reflect the impact on the results. The settings for input and storage are also determined by a combination of large-scale model analysis of sample data and human experience. Appropriate adjustments can also be made based on seasonal or rational influence conditions.

[0062] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A simulation training device for guiding group ball sports, characterized in that, The system includes a terminal sensing component, a terminal output component, and a cloud analysis device. The terminal sensing component is worn on the athlete's body to sense the athlete's movement status. The cloud analysis device analyzes the sensed movement status and transmits the results back to the terminal output component, which then displays the analysis results via an output device. It also includes an individual dynamic acquisition module, a panoramic 3D acquisition module, a single-unit dynamic analysis module, an ensemble analysis module, and an overlay output module; The individual dynamic acquisition module collects individual dynamic information through the terminal sensing component and transmits it to the individual dynamic analysis module via wireless transmission. The individual dynamic analysis module performs motion analysis based on individual dynamic information and generates a three-dimensional training report of the individual's movements. The panoramic 3D acquisition module collects collective information of the motion scene to obtain motion coordination information; The set analysis module acquires motion coordination information and three-dimensional training reports of individual movements, performs coordination degree analysis based on motion coordination information, and corrects the coordination degree analysis results through the three-dimensional training reports of individual movements to obtain an overall coordination assessment. The overlay output module acquires a three-dimensional training report of individual actions and an overall coordination assessment, and outputs early warning reminders based on the three-dimensional training report and the overall coordination assessment; The steps for the individual dynamic acquisition module to statistically analyze the frequency of linear acceleration variation and the frequency of anti-linear acceleration variation are as follows: The individual dynamic acquisition module counts the time elapsed from the first group of accelerations to the last group of accelerations, recording it as the total time T. The number of occurrences of counter-current acceleration and linear acceleration are recorded as Ay and At, respectively. The counter-current velocity variation frequency Fy is calculated using a formula. The frequency of linear velocity variation Ft is calculated using the formula. ; The set analysis module performs embedded analysis based on motion coordination information through a motion model, records the positional defects and trajectory coordination time differences between different personnel at each moment, and sums the positional defects and trajectory coordination time differences by assigning weights to obtain the coordination degree difference. The steps for obtaining the overall coordination evaluation by the set analysis module are as follows: The frequency of each athlete's individual directional velocity change is spatiotemporally matched with the athlete's critical nodes of sudden stops and changes of direction in the movement trajectory to assess whether the individual's high-frequency change of direction actions are consistent with the needs of team tactical execution. For moments with positioning defects, retrieve the speed dynamic data of the person moving at that position. If the individual is moving at high speed at that moment, it is determined that the positioning defect may be due to reasonable dynamic replacement; if the individual is moving at low speed or stationary, increase the negative evaluation weight of the positioning defect. For trajectory coordination time difference, correlation analysis is performed with the peak speed and reaction speed of relevant athletes to identify whether the cause of coordination time difference is lag in consciousness or insufficient individual absolute speed ability.

2. The collective ball sports guidance simulation training device according to claim 1, characterized in that, The individual dynamic information acquired by the individual dynamic acquisition module includes velocity dynamics and velocity change frequency; The method for the individual dynamic acquisition module to obtain the frequency of speed change is as follows: the acceleration and velocity fed back by the terminal sensing component are accumulated and recorded, and the real-time speed change of the individual is obtained through the motion model and recorded as speed dynamics; The method for the individual dynamic acquisition module to obtain the velocity variation frequency is as follows: filter the acceleration fed back by the terminal sensing component, select the effective acceleration, divide the effective acceleration into linear acceleration and anti-linear acceleration, and respectively count the linear acceleration variation frequency and the anti-linear acceleration variation frequency as the velocity variation frequency.

3. The collective ball sports guidance simulation training device according to claim 2, characterized in that, The method for filtering acceleration by the individual dynamic acquisition module is as follows: obtain a preset trigger threshold from the database, and then compare the obtained acceleration with the trigger threshold in real time. If the acceleration is greater than the set trigger threshold, it is recorded as effective acceleration; if the acceleration is not greater than the set trigger threshold, it is recorded as invalid acceleration. The method for classifying effective acceleration by the individual dynamic acquisition module is as follows: obtain the direction of each acceleration, and compare the acceleration with the direction of the previous group of adjacent accelerations. If the angle between the directions is greater than a set threshold, the acceleration is recorded as an opposite acceleration; if the angle between the directions is less than the set threshold, the acceleration is recorded as linear acceleration.

4. The collective ball sports guidance simulation training device according to claim 1, characterized in that, The three-dimensional training report generated by the single-unit dynamic analysis module includes: peak velocity, average velocity, frequency of opposite velocity variation, and frequency of linear velocity variation. Peak and average speeds are extracted based on speed dynamics.

5. The collective ball sports guidance simulation training device according to claim 1, characterized in that, The motion coordination information generated by the panoramic 3D acquisition module includes the real-time position and movement trajectory of the personnel. The method for the panoramic 3D acquisition module to obtain real-time positions is as follows: the coordinate data of each athlete in the field is captured by the terminal sensing component, and then combined with the preset digital model of the field, the real-time position of each athlete is mapped into three-dimensional spatial coordinates, and a dynamic position distribution map is generated according to the time series. The method for the panoramic 3D acquisition module to obtain the movement trajectory of personnel is as follows: based on the station position data within a continuous time period, the movement path of each person is reconstructed through a trajectory fitting algorithm, key movement nodes are marked, and the relative distance change trend between adjacent people is calculated.

Citation Information

Patent Citations

  • Ball game guidance method, intelligent wearable device and server

    CN106354988A

  • Football training method

    CN109091837A