Football training method and system based on unmanned aerial vehicle cluster

Abstract

The invention discloses a football training method and system based on an unmanned aerial vehicle cluster, and relates to the crossing field of football training and unmanned aerial vehicle application. The method comprises the following steps: carrying out video acquisition and analysis on a high-level match, obtaining movement tracks of players and footballs, and establishing a track database; selecting a training mode; the motion trail of each player and football in the match corresponds to the track of the unmanned aerial vehicle and is input to the unmanned aerial vehicle cluster control module; the football unmanned aerial vehicle and the player unmanned aerial vehicle perform combined flight in the football field according to the planned track, the player displacement and football movement process in the match is reproduced, the trainee follows the designated player unmanned aerial vehicle for training, and the movement data of the trainee is collected in real time; the training performance of the trainee is scored. According to the invention, the trainee can carry out football training in an environment close to an actual combat, clear performance feedback is provided for the training of the trainee, and the pertinence and efficiency of the training are improved at the same time.

Description

Technical Field

[0001] The present invention relates to the intersection of football training and unmanned aerial vehicle application, and in particular to a football training method and system based on unmanned aerial vehicle clusters. Background Art

[0002] Training football players is a high-cost, high-investment, and long-term process. Excellent players not only need to have good physical fitness, but also excellent football literacy and teamwork awareness, and the awareness of running in the game is a key factor. At present, in order to train football players at a high level, it is often necessary to spend a huge amount of money to invite strong teams to compete. Under the leadership of strong teams, high-intensity, high-tech confrontations can truly train players' physical fitness, game awareness, and running ability. However, the cost of this kind of competition training is too high and cannot be replicated. The lack of high-level competition training has always been a shortcoming and pain point in the training of football talents.

[0003] The rapid development of robots, drones and drone swarm technologies has gradually led to their widespread application in various fields. However, there is no mature solution or system for applying drone swarms to football training. The flight speed and response capabilities of drones have surpassed normal humans, so human movements and activities can be completely simulated and replicated by drones. Therefore, it is completely feasible to use drone swarms to simulate the movement process of a football team.

[0004] The invention patent with publication number RU2645505C1 and publication date of February 21, 2018 proposes a method for training and testing athletes, including: on the court, the software and hardware control a given number of radio-controlled drones, which have a gyro positioning system on the size of the court, and each drone has a skirt of a given size made of elastic material at the bottom, which is inflated by the airflow of the drone to form a three-dimensional model simulating a static or dynamic partner, competitor or training equipment. However, the above technical solution has the following shortcomings: 1. The UAV system in this technical solution uses a gyro positioning system for static or dynamic position control, but does not involve automated reproduction and simulation of the movement trajectory of real high-level competitions, and lacks accurate reproduction of the actual running positions, passing and confrontation strategies of athletes; 2. This technical solution cannot monitor and provide feedback on the trainee’s exercise status in real time; 3. This technical solution is only applicable to the simulation of simple physical obstacles or opponents. It cannot flexibly reproduce the complex tactical coordination and dynamic changes in actual games, and cannot provide a variety of training modes and training evaluations based on running, tactical coordination, confrontation and receiving training. Summary of the invention

[0005] In order to overcome the defects and shortcomings of the above-mentioned prior art, the present invention provides a football training method and system based on drone clusters. The present invention can simulate and reproduce high-level competition scenes through drone clusters, combined with real-time collection of trainees' position, speed, acceleration and heart rate data, to achieve personalized and scientific football training, provide clear performance feedback for trainees' training, and ensure the efficiency and safety of training.

[0006] In order to solve the above problems existing in the prior art, the present invention is implemented through the following technical solutions.

[0007] A football training method based on drone clusters includes the following steps: S1. Use the football match information collection module to collect and analyze the video of the high-level football match process, obtain the running trajectory of each player and the movement trajectory of the football during the game, and establish a trajectory database; S2. Select a training mode, the training mode including: Position training mode: the player drone runs according to the running path of one of the games in the trajectory database, and the trainees follow the movement of the player drone with the designated number; Tactical coordination training mode: select the tactical coordination running tracks in the game, and the player drones run according to the tactical coordination running tracks. The trainees follow the player drones with designated numbers to complete tactical running positions and coordinate with other player drones in terms of timing and position; Confrontation training mode: The players’ tackling trajectories in the game are screened out, and the trainees and the players’ drones engage in a fixed-point confrontation around the football, and the trainees cut off the passing line or complete the steal; Ball receiving training mode: the football passing trajectories in the game are screened out, and the football drones run according to different passing trajectories. The trainees need to follow the player drones with designated numbers to complete the ball receiving. The number of the player drones is 2-22 and is an integer multiple of 2. The player drones carry player human models and the football drones carry football models to simulate the game scene. The player human models are marked with different numbers. S3, according to the training mode selected in S2, the motion trajectory of the player corresponding to one of the games in the trajectory database is matched with the track of each player's drone, and converted into the track of each player's drone. At the same time, the motion trajectory of the football in the same game is converted into the track of the football drone, and the planned track is input into the drone cluster control module; S4, the football drone and several player drones, under the control of the drone cluster control module, respectively fly in combination in the actual football field according to the trajectory planned in S3, reproducing the running process of each player in the football game and the movement process of the football. The trainees follow the player drones with designated numbers for training, and the trainees' positions, movement directions, movement speeds, accelerations and heart rates at every moment are collected and stored in real time; S5. Score the trainees’ training performance, which includes the trainees’ physical fitness score and the positioning awareness score, tactical awareness score, confrontation score or receiving score in the corresponding training mode.

[0008] In S5, the physical fitness score is scored using the speed and acceleration compliance rate as the characterization dimensions, and is scored using a physical fitness scoring model, which is as follows: Formula 1 In Formula 1, the speed and acceleration reaching standard duration refers to the total duration of all time periods when the trainee's movement speed is greater than 80% of the actual movement speed of the player drone followed by the trainee and the acceleration is greater than 80% of the actual acceleration of the player drone followed by the trainee, and the total training process duration refers to the total duration that the trainee participates in the training process.

[0009] In S5, the positioning awareness score uses the positioning accuracy as the representation dimension and is scored through the positioning awareness scoring model, which is as follows: Formula 2 In Formula 2, the position accuracy duration refers to the total duration of all time periods when the distance between the trainee's position and the actual position of the player drone followed by the trainee is less than 10 meters, the total training process duration refers to the total time the trainee participates in the training process, and the football scramble duration refers to the total duration of all time periods when the distance between the trainee's position and the actual position of the football drone is less than 5 meters.

[0010] In S5, the tactical awareness score uses the reasonable running rate as the representation dimension and is scored through a tactical awareness scoring model, which is as follows: Formula 3 In Formula 3, the ideal position duration refers to the total duration of all periods when the positions of the football drone and the trainee are both in the opponent's half and the average distance between the trainee's position and all opponent player drones is greater than 5 meters. The total duration of the offensive process refers to the total duration of all periods when the positions of the football drone and the trainee are both in the opponent's half.

[0011] In S5, the confrontation score uses the confrontation positivity rate as the characterization dimension and is scored through the confrontation scoring model. The model is as follows: Formula 4 In Formula 4, the confrontation duration refers to the total duration of all periods when the positions of the football drone and the trainee are both in their own half and the minimum distance between the trainee and all opponent player drones is less than 3 meters. The total duration of the defense process refers to the total duration of all periods when the positions of the football drone and the trainee are both in their own half.

[0012] In S5, the ball receiving score uses the ball receiving position accuracy as the characterization dimension and is scored through the ball receiving scoring model, which is as follows: Formula 5 In Formula 5, the accurate duration of the ball receiving position refers to the total duration of all time periods when the movement direction of the football drone changes by more than 90 degrees or the acceleration of the football drone is greater than 20m / s², and the distance between the trainee's position and the actual position of the football drone is less than 2 meters. The total ball receiving duration refers to the total duration of all time periods when the movement direction of the football drone changes by more than 90 degrees or the acceleration of the football drone is greater than 20m / s², and the distance between the trainee's position and the actual position of the football drone is less than 10 meters.

[0013] The S1 specifically includes: First, we obtain panoramic video footage of a high-level football match in a public manner. Then, we use computers to perform intelligent trajectory analysis on the video, identify the football and each player through the video, and track their positions throughout the video. Then, with the center of the football field as the origin, and the length and width of the football field as the x-axis and y-axis respectively, the positions of the football and each player in the video are converted into relative position coordinates on the football field; Finally, according to the changes in the relative position coordinates of the football and each player at different moments, the movement direction and speed of the football at that moment are obtained by calculation; the collection of the relative position coordinates, movement direction and movement speed information of each player and the football at all moments in the entire football match constitutes a trajectory database of the football match; the high-level football matches at least include the FIFA World Cup, the European Football Championship, the UEFA Champions League, the English Premier League, the Spanish Football League and the German Football League; the disclosure method at least includes the Internet, television or live recording.

[0014] It also includes, when the trainee's heart rate is monitored to exceed the system's preset safety heart rate threshold, an alarm signal is issued, a vibration and sound reminder is given, and the drone cluster control module is instructed to reduce the operating speed or suspend training.

[0015] It also includes recording the historical training data and scores of all trainees, forming personalized training plans based on the historical training data and scores, automatically recommending training modes or adjusting the flight speed of the drone, and being able to display the training data and scores in real time on a two-dimensional interface.

[0016] A football training system based on drone clusters, comprising: A drone swarm used to simulate a game scene, which includes several player drones and a football drone; A drone cluster control module for controlling the drone cluster to fly along the planned trajectory in an actual football field; A football match information collection module for collecting and analyzing videos of high-level football matches, obtaining the plane relative position coordinates, movement direction and movement speed of each player on the field and the plane relative position coordinates, movement direction and movement speed of the football on the field during the match, and establishing a trajectory database; A training selection module for selecting a training mode; A trajectory planning module for converting the player motion trajectory in the trajectory database into the corresponding player drone trajectory, converting the football motion trajectory into the football drone trajectory, and inputting the planned trajectory into the drone cluster control module; A trainee wearable module for real-time acquisition and storage of the trainee's position, movement direction, movement speed, acceleration and heart rate at every moment; the trainee wearable module includes a GPS unit, a speed sensor, an acceleration sensor, a heart rate monitor and a wireless communication unit, the wireless communication unit is used to transmit the acquired data to the training evaluation module in real time; A training evaluation module for scoring trainee performance based on data collected by the trainee's wearable module.

[0017] It also includes an early warning module that sends out an alarm signal when it monitors that the trainee's heart rate exceeds the safe heart rate threshold preset by the system, vibrates and sound reminders through the trainee's wearable module, and notifies the drone cluster control module to reduce the operating intensity or suspend training.

[0018] It also includes a training management module for recording the historical training data and scores of all trainees, forming personalized training plans based on the historical training data and scores, automatically adjusting the flight speed of the drone, and being able to display the training data and scores in real time on a two-dimensional interface.

[0019] Compared with the prior art, the beneficial technical effects brought by the present invention are as follows: 1. The present invention obtains the movement and passing trajectories in high-level football matches, establishes a trajectory database, and then reproduces the player running trajectories and football movement trajectories in high-level football matches through a drone cluster, so that the training content can truly reproduce the game scene, thereby helping trainees to train more efficiently. The player drone carries a player human body model, and the football drone carries a football model, so that trainees can train in an environment close to actual combat, effectively improving the tactical awareness and adaptability of athletes in actual games. The present invention provides multiple modes of running training, tactical coordination training, confrontation training and receiving training, comprehensively covering the key skills in football games, and significantly improving the pertinence and practicality of training through high-precision trajectory matching and motion data analysis. At the same time, the present invention is suitable for players of different levels and positions, and while improving individual abilities, it enhances teamwork ability, helps to improve the overall tactical level, and provides a new method for football training.

[0020] 2. The present invention, by comparing the speed and acceleration of the trainee and the player drone they follow, the scoring model can accurately evaluate the trainee's physical fitness level during the training process, ensure that they have sufficient speed and reaction ability during the running and scrambling process, and set the speed and acceleration reaching the standard time, so that the scoring standard is more objective and quantitative, which can effectively evaluate the trainee's physical fitness and athletic ability, and help the trainee understand his physical condition during training, and provide direction for further improving physical fitness.

[0021] 3. In the present invention, the positioning awareness scoring model takes positioning accuracy as the core, and adopts the quantitative indicators of positioning accuracy time and football scramble time. The definition of positioning accuracy time and football scramble time enables the scoring model to accurately reflect the distance between the trainee and the target UAV, ensuring that during the positioning training, the trainee can follow the movement trajectory of the player UAV as accurately as possible and make contact with the football UAV in time. This not only improves the pertinence and scientific nature of the positioning training, but also effectively improves the trainees' tactical awareness and on-field reaction ability in actual games.

[0022] 4. The present invention, by designing a tactical awareness scoring model, quantitatively evaluates the rationality of the athlete's running position during the offensive process and the relative position relationship between the athlete and the opposing defensive player, emphasizes the ability of "space utilization" and "position selection", so that the coach can clearly see the performance of the trainees in specific tactics, and provide scientific data support for coaches and players. According to the scoring results, the offensive awareness is further adjusted and optimized to improve the actual tactical coordination effect in the game.

[0023] 5. The present invention, by designing a confrontation scoring model, quantitatively evaluates the confrontation positivity rate of athletes in the defense process and the relative position relationship between athletes and the opposing offensive players, etc., to help accurately evaluate the trainees' performance in dealing with the opposing offensive players on the court. If the trainees' confrontation score is low, it indicates that they are not sufficiently motivated in defense, and they may need to strengthen pressure defense and quick response training in training. In addition, the present invention can comprehensively evaluate the players' confrontation awareness such as marking and close contact in the defense process, to help trainees identify their own shortcomings and effectively improve their defensive performance.

[0024] 6. The present invention, by designing a ball receiving scoring model, quantitatively evaluates the position accuracy of athletes in the ball receiving process and the relative position relationship between the athletes and the landing point of the football, so as to clarify the accuracy of trainees in catching the ball in a dynamic and rapidly changing game environment. If the time of accurate ball receiving position is short, it indicates that the trainees may not be able to judge the passing line in time and accurately or fail to move to the ideal position quickly during the game, suggesting that more ball receiving training is needed to improve their reaction speed and position judgment ability, thereby improving the trainees' ball receiving success rate and game performance.

[0025] 7. The present invention adopts intelligent video analysis and coordinate conversion when collecting game information, which can automatically and accurately extract the relative position coordinates, movement direction, and movement speed information of players and footballs from high-level games, form a trajectory database, reduce manual intervention, and improve the accuracy and efficiency of data collection.

[0026] 8. The system of the present invention helps to ensure that trainees train within a safe range through dynamic heart rate monitoring. Once the heart rate is too high, the alarm system and drone cluster control can be used to reduce the intensity of training or suspend training, thereby improving the safety of training.

[0027] 9. The present invention, by recording the trainees' historical training data and scores, can formulate personalized training plans, automatically adjust the flight speed of the drone, and display the training data in real time, thereby enhancing the flexibility of training and the effect tracking function. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a schematic diagram of the process of the present invention; Figure 2 It is a schematic diagram of the system architecture of the present invention; Figure 3 It is a schematic diagram of the comparison between a real football match and a simulated training according to the present invention. DETAILED DESCRIPTION

[0029] The technical solution of the present invention will be clearly and completely described below in conjunction with specific embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] Example 1 As the most basic embodiment of the present invention, this embodiment discloses a football training method based on drone clusters, comprising the following steps: S1. Use the football match information collection module to collect and analyze the video of the high-level football match process, obtain the running trajectory of each player and the movement trajectory of the football during the game, and establish a trajectory database; S2. Select a training mode, the training mode including: Position training mode: the player drone runs according to the running path of one of the games in the trajectory database, and the trainees follow the movement of the player drone with the designated number; Tactical coordination training mode: select the tactical coordination running tracks in the game, and the player drones run according to the tactical coordination running tracks. The trainees follow the player drones with designated numbers to complete tactical running positions and coordinate with other player drones in terms of timing and position; Confrontation training mode: The players’ tackling trajectories in the game are screened out, and the trainees and the players’ drones engage in a fixed-point confrontation around the football, and the trainees cut off the passing line or complete the steal; Ball receiving training mode: the football passing trajectories in the game are screened out, and the football drones run according to different passing trajectories. The trainees need to follow the player drones with designated numbers to complete the ball receiving. There are 22 player drones, and the player drones carry player human models and the football drones carry football models to simulate the game scene. The player human models are marked with different numbers. S3, according to the training mode selected in S2, the motion trajectory of the player corresponding to one of the games in the trajectory database is matched with the track of each player's drone, and converted into the track of each player's drone. At the same time, the motion trajectory of the football in the same game is converted into the track of the football drone, and the planned track is input into the drone cluster control module; S4, the football drone and several player drones, under the control of the drone cluster control module, respectively fly in combination in the actual football field according to the trajectory planned in S3, reproducing the running process of each player in the football game and the movement process of the football. The trainees follow the player drones with designated numbers for training, and the trainees' positions, movement directions, movement speeds, accelerations and heart rates at every moment are collected and stored in real time; S5. Score the trainees’ training performance, which includes the trainees’ physical fitness score and the positioning awareness score, tactical awareness score, confrontation score or receiving score in the corresponding training mode.

[0031] This embodiment provides a new football training method by combining drone swarm technology with high-level football game data analysis. Trainees can improve their positioning, tactical coordination, confrontation, and catching skills in a targeted manner by following the game scenes simulated by drones. At the same time, the trainees' motion data, such as position, speed, and heart rate, are collected and evaluated in real time to help comprehensively monitor the training effect and promote the overall improvement of the trainees' skills, physical fitness, and tactical awareness. It has extremely high application value and promotion potential.

[0032] Example 2 As a preferred embodiment of the present invention, this embodiment is a further detailed supplement and elaboration of the technical solution of the present invention based on the above-mentioned embodiment 1. This embodiment discloses a football training method based on drone clusters, comprising the following steps: S1. Use the football match information collection module to collect and analyze the video of the high-level football match process, obtain the running trajectory of each player and the movement trajectory of the football during the game, and establish a trajectory database; S2. Select a training mode, the training mode including: Position training mode: the player drone runs according to the running path of one of the games in the trajectory database, and the trainees follow the movement of the player drone with the designated number; Tactical coordination training mode: select the tactical coordination running tracks in the game, and the player drones run according to the tactical coordination running tracks. The trainees follow the player drones with designated numbers to complete tactical running positions and coordinate with other player drones in terms of timing and position; Confrontation training mode: The players’ tackling trajectories in the game are screened out, and the trainees and the players’ drones engage in a fixed-point confrontation around the football, and the trainees cut off the passing line or complete the steal; Ball receiving training mode: the football passing trajectories in the game are screened out, and the football drones run according to different passing trajectories. The trainees need to follow the player drones with designated numbers to complete the ball receiving. There are two player drones, and the player drones carry player human models and the football drones carry football models to simulate the game scene. The player human models are marked with different numbers. S3, according to the training mode selected in S2, the motion trajectory of the player corresponding to one of the games in the trajectory database is matched with the track of each player's drone, and converted into the track of each player's drone. At the same time, the motion trajectory of the football in the same game is converted into the track of the football drone, and the planned track is input into the drone cluster control module; S4, the football drone and several player drones, under the control of the drone cluster control module, respectively fly in combination in the actual football field according to the trajectory planned in S3, reproducing the running process of each player in the football game and the movement process of the football. The trainees follow the player drones with designated numbers for training, and the trainees' positions, movement directions, movement speeds, accelerations and heart rates at every moment are collected and stored in real time; S5. Score the trainees’ training performance. The score includes the trainees’ physical fitness score and the running awareness score, tactical awareness score, confrontation score or receiving score in the corresponding training mode. The physical fitness score uses speed and acceleration compliance rate as the representation dimension and is scored through the physical fitness scoring model. The model is as follows: Formula 1 In Formula 1, the speed and acceleration reaching standard duration refers to the total duration of all periods when the trainee's movement speed is greater than 80% of the actual movement speed of the player drone followed by the trainee and the acceleration is greater than 80% of the actual acceleration of the player drone followed by the trainee, and the total training process duration refers to the total duration of the trainee's participation in the training process; The positioning awareness score uses the positioning accuracy as the representation dimension and is scored through the positioning awareness scoring model. The model is as follows: Formula 2 In Formula 2, the position accuracy duration refers to the total duration of all periods when the distance between the trainee's position and the actual position of the player drone followed by the trainee is less than 10 meters; the total training process duration refers to the total duration of the trainee's participation in the training process; and the football scramble duration refers to the total duration of all periods when the distance between the trainee's position and the actual position of the football drone is less than 5 meters. The tactical awareness score uses the reasonable running rate as the representation dimension and is scored through the tactical awareness scoring model. The model is as follows: Formula 3 In Formula 3, the ideal position duration refers to the total duration of all periods when the positions of the football drone and the trainee are both in the opponent's half, and the average distance between the trainee's position and all opponent player drones is greater than 5 meters. The total duration of the offensive process refers to the total duration of all periods when the positions of the football drone and the trainee are both in the opponent's half; The adversarial score uses the adversarial positivity rate as the representation dimension and is scored through the adversarial scoring model. The model is as follows: Formula 4 In Formula 4, the confrontation duration refers to the total duration of all periods during which the minimum distance between the trainee and all opposing player drones is less than 3 meters when both the football drone and the trainee are located in their own half of the field; the total duration of the defense process refers to the total duration of all periods during which both the football drone and the trainee are located in their own half of the field; The catch score uses the catch position accuracy as the representation dimension and is scored through the catch score model. The model is as follows: Formula 5 In Formula 5, the accurate duration of the ball receiving position refers to the total duration of all time periods when the movement direction of the football drone changes by more than 90 degrees or the acceleration of the football drone is greater than 20m / s², and the distance between the trainee's position and the actual position of the football drone is less than 2 meters. The total ball receiving duration refers to the total duration of all time periods when the movement direction of the football drone changes by more than 90 degrees or the acceleration of the football drone is greater than 20m / s², and the distance between the trainee's position and the actual position of the football drone is less than 10 meters.

[0033] This embodiment, through specific scoring models, such as physical fitness score, positioning awareness score, tactical awareness score, confrontation score and receiving score, can comprehensively and accurately evaluate the performance of trainees in different training modes, help coaches to more carefully analyze the technical and tactical abilities of trainees, and provide personalized suggestions, which not only improves the effect of training, but also strengthens the real-time feedback and monitoring during the training process, providing football players with a more efficient and intelligent training method.

[0034] Example 3 As another preferred embodiment of the present invention, this embodiment is a further detailed supplement and elaboration of the technical solution of the present invention based on the above-mentioned embodiment 2. This embodiment discloses a football training method based on drone clusters, comprising the following steps: S1. Video collection and analysis of high-level football matches are performed through a football match information collection module to obtain the running trajectories of each player and the movement trajectory of the football during the match, and to establish a trajectory database; specifically, the following steps are performed: first, panoramic video recordings of the entire process of the FIFA World Cup, European Football Championship, UEFA Champions League, English Premier League, Spanish Football League and German Football League are obtained through the Internet, television or on-site recordings; second, a computer is used to perform intelligent trajectory analysis on the video, and the identity of the football and each player is identified through the video and their positions throughout the process are tracked in the video; Then, with the center of the football field as the origin, and the length and width of the football field as the x-axis and y-axis respectively, the positions of the football and each player in the video are converted into relative position coordinates on the football field; Finally, according to the changes in the relative position coordinates of the football and each player at different times, the movement direction and speed of the football at that moment are obtained by calculation; the collection of the relative position coordinates, movement direction and movement speed information of each player and the football at all times in the entire football game constitutes the trajectory database of the football game; S2. Select a training mode, the training mode including: Position training mode: the player drone runs according to the running path of one of the games in the trajectory database, and the trainees follow the movement of the player drone with the designated number; Tactical coordination training mode: select the tactical coordination running tracks in the game, and the player drones run according to the tactical coordination running tracks. The trainees follow the player drones with designated numbers to complete tactical running positions and coordinate with other player drones in terms of timing and position; Confrontation training mode: The players’ tackling trajectories in the game are screened out, and the trainees and the players’ drones engage in a fixed-point confrontation around the football, and the trainees cut off the passing line or complete the steal; Ball receiving training mode: the football passing trajectories in the game are screened out, and the football drones run according to different passing trajectories. The trainees need to follow the player drones with designated numbers to complete the ball receiving. There are 14 player drones, and the player drones carry player human models and the football drones carry football models to simulate the game scene. The player human models are marked with different numbers. S3, according to the training mode selected in S2, the motion trajectory of the player corresponding to one of the games in the trajectory database is matched with the track of each player's drone, and converted into the track of each player's drone. At the same time, the motion trajectory of the football in the same game is converted into the track of the football drone, and the planned track is input into the drone cluster control module; S4, the football drone and several player drones, under the control of the drone cluster control module, respectively fly in combination in the actual football field according to the trajectory planned in S3, reproducing the running process of each player in the football game and the movement process of the football. The trainees follow the player drones with designated numbers for training, and the trainees' positions, movement directions, movement speeds, accelerations and heart rates at every moment are collected and stored in real time; S5. Score the trainees’ training performance. The score includes the trainees’ physical fitness score and the running awareness score, tactical awareness score, confrontation score or receiving score in the corresponding training mode. The physical fitness score uses speed and acceleration compliance rate as the representation dimension and is scored through the physical fitness scoring model. The model is as follows: Formula 1 In Formula 1, the speed and acceleration reaching standard duration refers to the total duration of all periods when the trainee's movement speed is greater than 80% of the actual movement speed of the player drone followed by the trainee and the acceleration is greater than 80% of the actual acceleration of the player drone followed by the trainee, and the total training process duration refers to the total duration of the trainee's participation in the training process; The positioning awareness score uses the positioning accuracy as the representation dimension and is scored through the positioning awareness scoring model. The model is as follows: Formula 2 In Formula 2, the position accuracy duration refers to the total duration of all periods when the distance between the trainee's position and the actual position of the player drone followed by the trainee is less than 10 meters; the total training process duration refers to the total duration of the trainee's participation in the training process; and the football scramble duration refers to the total duration of all periods when the distance between the trainee's position and the actual position of the football drone is less than 5 meters. The tactical awareness score uses the reasonable running rate as the representation dimension and is scored through the tactical awareness scoring model. The model is as follows: Formula 3 In Formula 3, the ideal position duration refers to the total duration of all periods when the positions of the football drone and the trainee are both in the opponent's half, and the average distance between the trainee's position and all opponent player drones is greater than 5 meters. The total duration of the offensive process refers to the total duration of all periods when the positions of the football drone and the trainee are both in the opponent's half; The adversarial score uses the adversarial positivity rate as the representation dimension and is scored through the adversarial scoring model. The model is as follows: Formula 4 In Formula 4, the confrontation duration refers to the total duration of all periods during which the minimum distance between the trainee and all opposing player drones is less than 3 meters when both the football drone and the trainee are located in their own half of the field; the total duration of the defense process refers to the total duration of all periods during which both the football drone and the trainee are located in their own half of the field; The catch score uses the catch position accuracy as the representation dimension and is scored through the catch score model. The model is as follows: Formula 5 In Formula 5, the accurate duration of the ball receiving position refers to the total duration of all time periods when the movement direction of the football drone changes by more than 90 degrees or the acceleration of the football drone is greater than 20m / s², and the distance between the trainee's position and the actual position of the football drone is less than 2 meters. The total ball receiving duration refers to the total duration of all time periods when the movement direction of the football drone changes by more than 90 degrees or the acceleration of the football drone is greater than 20m / s², and the distance between the trainee's position and the actual position of the football drone is less than 10 meters.

[0035] In this embodiment, by using panoramic video recordings of high-level football matches to perform intelligent trajectory analysis, the movement trajectory, speed and direction information of the football and the players are accurately extracted, and a real and accurate trajectory database is constructed. The database can truly reproduce high-level game scenes and provide accurate data support for drone simulation games. At the same time, the authority and reference value of the trajectory database are improved by obtaining data from top international competitions in a public manner, which provides a high-reliability foundation for scientific and standardized football training and greatly improves the training effect and practical application value.

[0036] Example 4 As another preferred embodiment of the present invention, this embodiment is a further detailed supplement and elaboration of the technical solution of the present invention based on the above-mentioned embodiment 3. This embodiment discloses a football training method based on drone clusters, comprising the following steps: S1. Video collection and analysis of high-level football matches are performed through a football match information collection module to obtain the running trajectories of each player and the movement trajectory of the football during the match, and to establish a trajectory database; specifically, the following steps are performed: first, panoramic video recordings of the entire process of the FIFA World Cup, European Football Championship, UEFA Champions League, English Premier League, Spanish Football League and German Football League are obtained through the Internet, television or on-site recordings; second, a computer is used to perform intelligent trajectory analysis on the video, and the identity of the football and each player is identified through the video and their positions throughout the process are tracked in the video; Then, with the center of the football field as the origin, and the length and width of the football field as the x-axis and y-axis respectively, the positions of the football and each player in the video are converted into relative position coordinates on the football field; Finally, according to the changes in the relative position coordinates of the football and each player at different times, the movement direction and speed of the football at that moment are obtained by calculation; the collection of the relative position coordinates, movement direction and movement speed information of each player and the football at all times in the entire football game constitutes the trajectory database of the football game; S2. Select a training mode, the training mode including: Position training mode: the player drone runs according to the running path of one of the games in the trajectory database, and the trainees follow the movement of the player drone with the designated number; Tactical coordination training mode: select the tactical coordination running tracks in the game, and the player drones run according to the tactical coordination running tracks. The trainees follow the player drones with designated numbers to complete tactical running positions and coordinate with other player drones in terms of timing and position; Confrontation training mode: The players’ tackling trajectories in the game are screened out, and the trainees and the players’ drones engage in a fixed-point confrontation around the football, and the trainees cut off the passing line or complete the steal; Ball receiving training mode: the football passing trajectories in the game are screened out, and the football drones run according to different passing trajectories. The trainees need to follow the player drones with designated numbers to complete the ball receiving. There are 22 player drones, and the player drones carry player human models and the football drones carry football models to simulate the game scene. The player human models are marked with different numbers. S3, according to the training mode selected in S2, the motion trajectory of the player corresponding to one of the games in the trajectory database is matched with the track of each player's drone, and converted into the track of each player's drone. At the same time, the motion trajectory of the football in the same game is converted into the track of the football drone, and the planned track is input into the drone cluster control module; S4, the football drone and several player drones are controlled by the drone cluster control module to fly in combination in the actual football field according to the trajectory planned in S3, reproducing the running process of each player in the football game and the movement process of the football. The trainees follow the player drones with designated numbers for training, and the trainees' positions, movement directions, movement speeds, accelerations and heart rates at each moment are collected and stored in real time; refer to Figure 3 The color image can clearly show how the trajectories of the two players and the ball in a real football game are converted into the trajectories of drone clusters and used for training. For example, the black and white image cannot show the one-to-one correspondence between the real football game and the drones in training; S5. Score the trainees’ training performance. The score includes the trainees’ physical fitness score and the running awareness score, tactical awareness score, confrontation score or receiving score in the corresponding training mode. The physical fitness score uses speed and acceleration compliance rate as the representation dimension and is scored through the physical fitness scoring model. The model is as follows: Formula 1 In Formula 1, the speed and acceleration reaching standard duration refers to the total duration of all periods when the trainee's movement speed is greater than 80% of the actual movement speed of the player drone followed by the trainee and the acceleration is greater than 80% of the actual acceleration of the player drone followed by the trainee, and the total training process duration refers to the total duration of the trainee's participation in the training process; The positioning awareness score uses the positioning accuracy as the representation dimension and is scored through the positioning awareness scoring model. The model is as follows: Formula 2 In Formula 2, the position accuracy duration refers to the total duration of all periods when the distance between the trainee's position and the actual position of the player drone followed by the trainee is less than 10 meters; the total training process duration refers to the total duration of the trainee's participation in the training process; and the football scramble duration refers to the total duration of all periods when the distance between the trainee's position and the actual position of the football drone is less than 5 meters. The tactical awareness score uses the reasonable running rate as the representation dimension and is scored through the tactical awareness scoring model. The model is as follows: Formula 3 In Formula 3, the ideal position duration refers to the total duration of all periods when the positions of the football drone and the trainee are both in the opponent's half, and the average distance between the trainee's position and all opponent player drones is greater than 5 meters. The total duration of the offensive process refers to the total duration of all periods when the positions of the football drone and the trainee are both in the opponent's half; The adversarial score uses the adversarial positivity rate as the representation dimension and is scored through the adversarial scoring model. The model is as follows: Formula 4 In Formula 4, the confrontation duration refers to the total duration of all periods during which the minimum distance between the trainee and all opposing player drones is less than 3 meters when both the football drone and the trainee are located in their own half of the field; the total duration of the defense process refers to the total duration of all periods during which both the football drone and the trainee are located in their own half of the field; The catch score uses the catch position accuracy as the representation dimension and is scored through the catch score model. The model is as follows: Formula 5 In Formula 5, the accurate duration of the ball receiving position refers to the total duration of all time periods when the movement direction of the football drone changes by more than 90 degrees or the acceleration of the football drone is greater than 20m / s², and the distance between the trainee's position and the actual position of the football drone is less than 2 meters. The total ball receiving duration refers to the total duration of all time periods when the movement direction of the football drone changes by more than 90 degrees or the acceleration of the football drone is greater than 20m / s², and the distance between the trainee's position and the actual position of the football drone is less than 10 meters.

[0037] It also includes that when the trainee's heart rate is monitored to exceed the safe heart rate threshold preset by the system, an alarm signal is issued, the trainee's wearable module vibrates and sounds a reminder, and the drone cluster control module is instructed to reduce the operating speed or suspend training; It also includes recording the historical training data and scores of all trainees, forming personalized training plans based on the historical training data and scores, automatically recommending training modes or adjusting the flight speed of the drone, and being able to display the training data and scores in real time on a two-dimensional interface.

[0038] This embodiment introduces heart rate monitoring and personalized training recommendation mechanisms to make the training process safer, smarter and more personalized. By monitoring the heart rate of the trainees and setting safety thresholds, timely warnings can be issued and corresponding measures can be taken to ensure the physical health of the trainees. Combined with historical training data and scores, training plans are tailored for each trainee, and training intensity and mode are automatically adjusted to improve training effects, helping them to continuously optimize training strategies, thereby more efficiently improving football skills and physical fitness levels.

[0039] Example 5 This embodiment provides a football training system based on drone clusters, including: A football training system based on drone clusters, comprising: A drone swarm used to simulate a game scene, which includes several player drones and a football drone; A drone cluster control module for controlling the drone cluster to fly along the planned trajectory in an actual football field; A football match information collection module for collecting and analyzing videos of high-level football matches, obtaining the plane relative position coordinates, movement direction and movement speed of each player on the field and the plane relative position coordinates, movement direction and movement speed of the football on the field during the match, and establishing a trajectory database; A training selection module for selecting a training mode; A trajectory planning module for converting the player motion trajectory in the trajectory database into the corresponding player drone trajectory, converting the football motion trajectory into the football drone trajectory, and inputting the planned trajectory into the drone cluster control module; A trainee wearable module for real-time acquisition and storage of the trainee's position, movement direction, movement speed, acceleration and heart rate at every moment; the trainee wearable module includes a GPS unit, a speed sensor, an acceleration sensor, a heart rate monitor and a wireless communication unit, the wireless communication unit is used to transmit the acquired data to the training evaluation module in real time; A training evaluation module for scoring trainee performance based on data collected by the trainee's wearable module.

[0040] It also includes an early warning module that sends out an alarm signal when it monitors that the trainee's heart rate exceeds the safe heart rate threshold preset by the system, vibrates and sound reminders through the trainee's wearable module, and notifies the drone cluster control module to reduce the operating intensity or suspend training.

[0041] It also includes a training management module for recording the historical training data and scores of all trainees, forming personalized training plans based on the historical training data and scores, automatically adjusting the flight speed of the drone, and being able to display the training data and scores in real time on a two-dimensional interface.

Claims

1. A football training method based on drone clusters, characterized in that: The following steps are involved: S1. Use the football match information collection module to collect and analyze the video of the high-level football match process, obtain the running trajectory of each player and the movement trajectory of the football during the game, and establish a trajectory database; S2. Select a training mode, the training mode including: Position training mode: the player drone runs according to the running path of one of the games in the trajectory database, and the trainees follow the movement of the player drone with the designated number; Tactical coordination training mode: select the tactical coordination running tracks in the game, and the player drones run according to the tactical coordination running tracks. The trainees follow the player drones with designated numbers to complete tactical running positions and coordinate with other player drones in terms of timing and position; Confrontation training mode: The players’ tackling trajectories in the game are screened out, and the trainees and the players’ drones engage in a fixed-point confrontation around the football, and the trainees cut off the passing line or complete the steal; Ball-catching training mode: select the football passing trajectories in the game, and the football drones will run along different passing trajectories. The trainees need to follow the designated numbered player drones to catch the ball; The number of the player drones is 2-22 and is an integer multiple of 2. The player drones carry player human models, and the football drones carry football models for simulating game scenes. The player human models are marked with different numbers. S3, according to the training mode selected in S2, the motion trajectory of the player corresponding to one of the games in the trajectory database is matched with the track of each player's drone, and converted into the track of each player's drone. At the same time, the motion trajectory of the football in the same game is converted into the track of the football drone, and the planned track is input into the drone cluster control module; S4, the football drone and several player drones, under the control of the drone cluster control module, respectively fly in combination in the actual football field according to the trajectory planned in S3, reproducing the running process of each player in the football game and the movement process of the football. The trainees follow the player drones with designated numbers for training, and the trainees' positions, movement directions, movement speeds, accelerations and heart rates at every moment are collected and stored in real time; S5. Score the trainees’ training performance, which includes the trainees’ physical fitness score and the positioning awareness score, tactical awareness score, confrontation score or receiving score in the corresponding training mode.

2. The football training method based on drone cluster according to claim 1, characterized in that: In S5, the physical fitness score is scored using the speed and acceleration compliance rate as the characterization dimensions, and is scored using a physical fitness scoring model, which is as follows: Formula 1 In Formula 1, the speed and acceleration reaching standard duration refers to the total duration of all time periods when the trainee's movement speed is greater than 80% of the actual movement speed of the player drone followed by the trainee and the acceleration is greater than 80% of the actual acceleration of the player drone followed by the trainee, and the total training process duration refers to the total duration that the trainee participates in the training process.

3. The football training method based on drone cluster according to claim 2, characterized in that: In S5, the positioning awareness score uses the positioning accuracy as the representation dimension and is scored through the positioning awareness scoring model, which is as follows: Formula 2 In Formula 2, the position accuracy duration refers to the total duration of all time periods when the distance between the trainee's position and the actual position of the player drone followed by the trainee is less than 10 meters, the total training process duration refers to the total time the trainee participates in the training process, and the football scramble duration refers to the total duration of all time periods when the distance between the trainee's position and the actual position of the football drone is less than 5 meters.

4. The football training method based on drone cluster according to claim 3 is characterized in that: In S5, the tactical awareness score uses the reasonable running rate as the representation dimension and is scored through a tactical awareness scoring model, which is as follows: Formula 3 In Formula 3, the ideal position duration refers to the total duration of all periods when the positions of the football drone and the trainee are both in the opponent's half and the average distance between the trainee's position and all opponent player drones is greater than 5 meters. The total duration of the offensive process refers to the total duration of all periods when the positions of the football drone and the trainee are both in the opponent's half.

5. The football training method based on drone cluster according to claim 4 is characterized in that: In S5, the confrontation score uses the confrontation positivity rate as the characterization dimension and is scored through the confrontation scoring model. The model is as follows: Formula 4 In Formula 4, the confrontation duration refers to the total duration of all periods when the positions of the football drone and the trainee are both in their own half and the minimum distance between the trainee and all opponent player drones is less than 3 meters. The total duration of the defense process refers to the total duration of all periods when the positions of the football drone and the trainee are both in their own half.

6. The football training method based on drone cluster according to claim 5, characterized in that: In S5, the ball receiving score uses the ball receiving position accuracy as the characterization dimension and is scored through the ball receiving score model, which is as follows: Formula 5 In Formula 5, the accurate duration of the ball receiving position refers to the total duration of all time periods when the movement direction of the football drone changes by more than 90 degrees or the acceleration of the football drone is greater than 20m / s², and the distance between the trainee's position and the actual position of the football drone is less than 2 meters. The total ball receiving duration refers to the total duration of all time periods when the movement direction of the football drone changes by more than 90 degrees or the acceleration of the football drone is greater than 20m / s², and the distance between the trainee's position and the actual position of the football drone is less than 10 meters.

7. A football training method based on drone clusters according to any one of claims 1 to 6, characterized in that: The S1 specifically includes: First, we obtain panoramic video footage of a high-level football match in a public manner. Then, we use computers to perform intelligent trajectory analysis on the video, identify the football and each player through the video, and track their positions throughout the video. Then, with the center of the football field as the origin and the length and width of the football field as the x-axis and y-axis respectively, the positions of the football and each player in the video are converted into relative position coordinates on the football field; Finally, according to the changes in the relative position coordinates of the football and each player at different moments, the movement direction and speed of the football at that moment are obtained by calculation; the collection of the relative position coordinates, movement direction and movement speed information of each player and the football at all moments in the entire football match constitutes a trajectory database of the football match; the high-level football matches at least include the FIFA World Cup, the European Football Championship, the UEFA Champions League, the English Premier League, the Spanish Football League and the German Football League; the disclosure method at least includes the Internet, television or live recording.

8. The football training method based on drone cluster according to claim 7 is characterized in that: It also includes, when the trainee's heart rate is monitored to exceed the system's preset safety heart rate threshold, an alarm signal is issued, a vibration and sound reminder is given, and the drone cluster control module is instructed to reduce the operating speed or suspend training.

9. The football training method based on drone cluster according to claim 8, characterized in that: It also includes recording the historical training data and scores of all trainees, forming personalized training plans based on the historical training data and scores, automatically recommending training modes or adjusting the flight speed of the drone, and being able to display the training data and scores in real time on a two-dimensional interface.

10. A football training system based on drone clusters, characterized in that: include: A drone swarm used to simulate a game scene, which includes several player drones and a football drone; A drone cluster control module for controlling the drone cluster to fly along the planned trajectory in an actual football field; A football match information collection module for collecting and analyzing videos of high-level football matches, obtaining the plane relative position coordinates, movement direction and movement speed of each player on the field and the plane relative position coordinates, movement direction and movement speed of the football on the field during the match, and establishing a trajectory database; A training selection module for selecting a training mode; A trajectory planning module for converting the player motion trajectory in the trajectory database into the corresponding player drone trajectory, converting the football motion trajectory into the football drone trajectory, and inputting the planned trajectory into the drone cluster control module; A trainee wearable module for real-time acquisition and storage of the trainee's position, movement direction, movement speed, acceleration and heart rate at every moment; The trainee wearable module includes a GPS unit, a speed sensor, an acceleration sensor, a heart rate monitor and a wireless communication unit, and the wireless communication unit is used to transmit the collected data to the training evaluation module in real time; a training evaluation module for scoring the trainee's performance based on data collected by the trainee's wearable module; It also includes an early warning module for sending out an alarm signal when the trainee's heart rate exceeds the safe heart rate threshold preset by the system, vibrating and sounding a reminder through the trainee's wearable module, and notifying the drone cluster control module to reduce the operating intensity or suspend training; It also includes a training management module for recording the historical training data and scores of all trainees, forming personalized training plans based on the historical training data and scores, automatically adjusting the flight speed of the drone, and being able to display the training data and scores in real time on a two-dimensional interface.

Images