Shooting training method based on training big data

Abstract

The invention discloses a training big data-based shooting training method, which comprises the following steps of: acquiring expression data and physiological data of a shooter and shooting environment data, and calculating an expression score of the shooter according to the expression data of the shooter; analyzing the influence degree among the performance data, the physiological data and the environmental data according to the performance score of the shooter in combination with the physiological data of the shooter and the environmental data of shooting; fusing the expression data, the physiological data and the environment data according to the influence degree among the expression data, the physiological data and the environment data to obtain training state data; inputting the training state data into the training strategy model to obtain an optimized training strategy of the shooter, and dynamically adjusting the current shooting training method according to the optimized training strategy.

Description

Technical Field

[0001] The present invention relates to the technical field of shooting training methods, and in particular to a shooting training method based on training big data. Background Art

[0002] With the growing demand for military, police and competitive shooting training, traditional shooting training methods face many technical bottlenecks and challenges. Traditional training mainly relies on fixed shooting ranges or actual combat simulation sites. Although these methods can provide basic shooting training, they are obviously insufficient in terms of realism, dynamic adaptability, environmental diversity and personalized training.

[0003] In traditional training scenarios, fixed targets or simple simulated targets lack dynamism and realism, and cannot reproduce the complex environments and target behaviors in real combat. Shooting training is often limited to fixed conditions (such as indoor shooting ranges or single weather environments), and it is difficult to simulate complex environmental factors (such as wind speed, light changes, noise interference, etc.). In traditional shooting ranges, dynamic targets are usually achieved through mechanical devices, with a single motion trajectory and difficult to adjust flexibly. Traditional training is difficult to effectively simulate high-pressure situations (such as enemy attacks, teamwork, sudden threats, etc.), resulting in shooters being unable to make quick decisions or remain calm in actual combat.

[0004] In recent years, technologies such as virtual reality (VR) and augmented reality (AR) have been introduced into shooting training, improving the flexibility of training through virtual scene simulation. However, these technologies also have limitations. For example, VR devices require wearing a headset, which may affect the shooter's natural movements and sense of reality; although AR devices can superimpose virtual targets, they still have technical limitations in dynamic simulation and interactivity in complex environments. Due to the limitations of equipment (such as headsets or sensors), the collected training data may not be accurate enough, and even errors may occur. These errors may be reflected in the shooter's motion capture, aiming trajectory, hit point and other key data. If these inaccurate data are used in subsequent training optimization or guidance strategies, it may cause the adjustment of training content to not meet actual needs, thereby misleading or negatively affecting the shooter's training effect.

[0005] The purpose of this invention is to design a shooting training method based on training big data in view of the above-mentioned problems in the prior art. Summary of the invention

[0006] In view of this, the purpose of the present invention is to propose a shooting training method based on training big data, which can solve the above-mentioned problems.

[0007] The present invention provides a shooting training method based on training big data, comprising:

[0008] obtaining performance data and physiological data of the shooter, shooting environment data, and calculating a performance score of the shooter based on the performance data of the shooter;

[0009] According to the shooter's performance score, combined with the shooter's physiological data and shooting environment data, the influence degree between the performance data, physiological data and environmental data is analyzed;

[0010] The performance data, the physiological data, and the environmental data are integrated according to the influence degree among the performance data, the physiological data, and the environmental data to obtain the training status data;

[0011] The training status data is input into the training strategy model to obtain the shooter's optimized training strategy, and the current shooting training method is dynamically adjusted according to the optimized training strategy.

[0012] Furthermore, the acquisition of the shooter's performance data and physiological data, and the shooting environment data includes:

[0013] Obtain the shooter's shooting accuracy, reaction time, and aiming deviation;

[0014] Get the shooter’s heart rate and breathing rate;

[0015] Get the light intensity, wind speed, and noise around the shooting.

[0016] Furthermore, the calculation formula for calculating the shooter's performance score based on the shooter's performance data is as follows:

[0017] S = w1·hit rate-w2·reaction time-w3·aiming deviation,

[0018] Among them, w1, w2, w3 are weight coefficients, and S is the performance score.

[0019] Further, the analyzing the influence degree between the performance data, the physiological data, and the environmental data according to the shooter's performance score combined with the shooter's physiological data and the shooting environment data includes:

[0020] If the performance score is satisfactory, the advanced training weight will be obtained;

[0021] If the performance score is unsatisfactory, determine whether the physiological data affects the performance data. If the physiological data does not affect the performance data, obtain the action training weight. Otherwise, determine whether the environmental data affects the physiological data.

[0022] If the environmental data affects the physiological data, the environmental training weight is obtained, otherwise the psychological training weight is obtained.

[0023] Further, the determining whether the physiological data is related to the performance-affecting data includes:

[0024] A multivariate linear regression model was constructed based on physiological data and performance scores. The model formula is as follows:

[0025] Y 表现 =β0+β1·heart rate+β2·respiratory rate+∈

[0026] Among them, Y 表现 is the performance score, β1 and β2 are regression coefficients, and ∈ is the error term;

[0027] If any regression coefficient of β1 or β2 is greater than the significance threshold, the physiological data affects the performance data; otherwise, the physiological data does not affect the performance data.

[0028] Further, the determining whether the environmental data affects the physiological data includes:

[0029] A random forest regression model was constructed based on environmental data and physiological data. The model formula is as follows:

[0030] Y 心率 , Y 呼吸频率 =f(light intensity, wind speed, noise),

[0031] Among them, Y 心率 is the target heart rate, Y 呼吸频率 is the target respiratory rate, f is the random forest model;

[0032] The environmental data is used as the input of the random forest regression model, and the physiological data is used as the output of the random forest regression model. The random forest regression model is trained to obtain the feature importance parameters of the random forest regression model.

[0033] If any of the feature importance parameters of the random forest regression model is greater than the feature importance threshold, the environmental data affects the physiological data, otherwise the environmental data does not affect the physiological data.

[0034] Further, the fusing of the performance data, the physiological data, and the environmental data according to the degree of influence between the performance data, the physiological data, and the environmental data to obtain the training state data includes:

[0035] The performance data, physiological data, and environmental data are fused according to the obtained different types of weights to obtain the training status data.

[0036] Furthermore, the training strategy model is obtained through the following steps:

[0037] The performance data, physiological data, and environmental data are used as state variables, and the adjustment of simulation parameters of the training environment, training rhythm adjustment, and training content adjustment are used as action variables to build a training strategy model;

[0038] Construct a reward function based on shooting hit rate, shooting reaction time, heart rate, and aiming deviation to optimize the training strategy model.

[0039] Furthermore, the calculation formula for constructing the reward function based on the shooting hit rate, shooting reaction time, heart rate, and aiming trajectory is as follows:

[0040] R = α1 hit rate - α2 reaction time - α3 heart rate deviation + α4 aiming deviation,

[0041] Among them, α1, α2, α3, and α4 are weight parameters.

[0042] Furthermore, the step of inputting the training status data into the training strategy model to obtain the shooter's optimized training strategy includes:

[0043] The training status data is input into the training strategy model to obtain environmental simulation parameters, training rhythm parameters, and training content parameters. The environmental simulation parameters include: light intensity, wind speed, and noise. The training rhythm parameters include: number of shootings, shooting intervals, and shooting duration. The training content parameters include: static shooting, dynamic shooting, and static and dynamic combined shooting.

[0044] Beneficial effects of the present invention:

[0045] First, performance data is the core indicator of the shooter's shooting ability. Shooting hit rate reflects the shooter's shooting accuracy. Reaction time reflects the shooter's reaction speed. Aim deviation reflects the shooter's precision and stability. Physiological data reflects the shooter's physical state during shooting, including heart rate and breathing rate. Environmental data reflects the external conditions of the shooting site, including light intensity, wind speed and noise. These data can be used to analyze the impact of the external environment on the shooter's performance and physiological state.

[0046] Second, the performance score is used to provide feedback on the direction that the shooter needs to improve. If the performance score is qualified, the shooter will obtain advanced training weights, and the training content will be adjusted to make the training more focused on improving difficult skills. If the performance score is unsatisfactory, the factors affecting performance (physiological state and environmental conditions) will be further analyzed to develop training weights with different focuses for the shooter (such as action training or environmental adaptation training).

[0047] Third, the contribution of different types of data to the training strategy is obviously different. If these data are simply mixed and the weights are ignored, the model may overfit some data and ignore other more critical information. The training weights reflect the areas where the current training needs to be strengthened or adjusted. The training weights obtained based on performance scores, physiological data, and environmental data can determine the specific training direction.

[0048] Fourth, the fused data (training status data) is the result of performance data, physiological data, and environmental data after weight adjustment, which can comprehensively reflect the shooter's current training status. The training strategy model is based on this comprehensive status to analyze the shooter's current ability level, adaptability, and weak links, and formulate an optimized training strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0050] Figure 1 It is a flow chart of the method of this embodiment. DETAILED DESCRIPTION

[0051] To facilitate understanding by those skilled in the art, the structure of the present invention is now further described in detail with reference to the embodiments and the accompanying drawings. It should be understood that the steps mentioned in the present embodiment, unless otherwise specified in their order, can be adjusted in order according to actual needs, and can even be executed simultaneously or partially simultaneously.

[0052] like Figure 1 As shown, an embodiment of the present invention provides a shooting training method based on training big data, including:

[0053] S1 obtains the performance data and physiological data of the shooter, and the shooting environment data, and calculates the performance score of the shooter according to the performance data of the shooter;

[0054] S101 obtains the shooter's shooting accuracy, reaction time, and aiming deviation;

[0055] S102 obtains the shooter's heart rate and breathing rate;

[0056] S103 obtains the light intensity, wind speed, and noise around the shooting area.

[0057] The calculation formula for calculating the shooter's performance score based on the shooter's performance data is as follows:

[0058] S = w1·hit rate-w2·reaction time-w3·aiming deviation,

[0059] Among them, w1, w2, w3 are weight coefficients, and S is the performance score.

[0060] In this step, the performance data directly reflects the core indicators of the shooter's shooting ability. The shooting hit rate is the ratio of the number of times the target is hit to the total number of shots, which reflects the shooter's shooting accuracy. The reaction time is the time interval from the target appearing to the shooter firing, which reflects the shooter's reaction speed. The aiming deviation is the deviation distance between the shooting point and the target point, which reflects the shooter's accuracy and stability.

[0061] Physiological data reflects the shooter's physical state during shooting, including heart rate and breathing rate. Heart rate is the number of heart beats per unit time (usually expressed in bpm), which reflects the shooter's tension and physical load. Breathing rate is the number of breaths per unit time, which reflects the shooter's breathing control and physical state. These physiological data can be recorded using a wristband.

[0062] Environmental data reflects the external conditions of the shooting range, including light intensity, wind speed and noise. These data can be used to analyze the impact of the external environment on the shooter's performance and physiological state. Light intensity is the light level of the shooting range (usually expressed in lux), reflecting the visual conditions of shooting. Wind speed is the wind speed of the shooting range (usually expressed in m / s), reflecting the air flow during shooting. Noise is the noise level of the shooting range (usually expressed in dB), reflecting the sound interference during shooting.

[0063] Since the hit rate, reaction time, and aiming deviation have different dimensions, adding them together directly will cause the indicator with a larger dimension to have too much influence on the score. Therefore, it is necessary to normalize the reaction time and aiming deviation. The weight setting needs to be based on the actual scenario and task requirements: the higher the performance score, the higher the shooter's level.

[0064] S2 analyzes the influence between the performance data, physiological data, and environmental data based on the shooter's performance score combined with the shooter's physiological data and shooting environment data;

[0065] If S201 performs well, it will obtain advanced training weights;

[0066] In this step, the training weight reflects the aspects that need to be strengthened or adjusted in the current training. The training weights obtained by judging the performance score, physiological data and environmental data can determine the specific training direction. If the performance score is qualified, the advanced training weights are obtained, and the training content is adjusted to make the training more focused on improving difficult skills. For example, adding more complex dynamic targets or more stringent time limits can improve the accuracy and quick response ability of the shooter. The advanced training weights distinguish between more skilled shooters and beginners, and focus on advanced skills rather than repeated improvement of basic abilities. If the performance score is unsatisfactory, further analyze the factors affecting performance (physiological state and environmental conditions), so as to formulate training weights with different focuses for the shooter (such as action training or environmental adaptation training). If the physiological data affects the performance score, it is necessary to coordinate the relationship between the action and the physiological state. It is recommended to obtain training weights with action control and breathing rhythm as the core (such as shooting stability training to control breathing). If the environmental data affects the physiological state, the training conditions are adjusted to help the shooter adapt to the environment, such as reducing light or noise interference, and obtaining training weights that are more suitable for environmental adaptation.

[0067] S202: if the performance score is unsatisfactory, determining whether the physiological data affects the performance data; if the physiological data does not affect the performance data, obtaining the action training weight; otherwise, determining whether the environmental data affects the physiological data;

[0068] A multivariate linear regression model was constructed based on physiological data and performance scores. The model formula is as follows:

[0069] Y 表现 =β0+β1·heart rate+β2·respiratory rate+∈

[0070] Among them, Y 表现 is the performance score, β1 and β2 are regression coefficients, and ∈ is the error term;

[0071] If any regression coefficient of β1 or β2 is greater than the significance threshold, the physiological data affects the performance data; otherwise, the physiological data does not affect the performance data.

[0072] In this step, there is usually a relatively simple linear relationship between physiological data (such as heart rate, respiratory rate) and performance scores. For example, a higher heart rate may indicate a high stress level, which may reduce performance scores. Multiple linear regression can directly quantify this linear relationship (represented by the regression coefficient β). Multiple linear regression provides a significance test result of the regression coefficient, which can determine the independent impact of each physiological feature on the performance data. In this application, the significance threshold can be set to 1. For example, if β1=2.5, it means that for every 1 unit increase in heart rate, the performance score will increase by 2.5 units. If β1 is greater than the significance threshold, the physiological data affects the performance data.

[0073] S203: If the environmental data affects the physiological data, then the environmental training weight is obtained; otherwise, the psychological training weight is obtained.

[0074] A random forest regression model was constructed based on environmental data and physiological data. The model formula is as follows:

[0075] Y 心率 , Y 呼吸频率 =f(light intensity, wind speed, noise),

[0076] Among them, Y 心率 is the target heart rate, Y 呼吸频率 is the target respiratory rate, f is the random forest model;

[0077] The environmental data is used as the input of the random forest regression model, and the physiological data is used as the output of the random forest regression model. The random forest regression model is trained to obtain the feature importance parameters of the random forest regression model.

[0078] If any of the feature importance parameters of the random forest regression model is greater than the feature importance threshold, the environmental data affects the physiological data, otherwise the environmental data does not affect the physiological data.

[0079] In this step, the random forest regression model is a non-parametric model that does not have strict assumptions about the distribution of data and can handle complex nonlinear relationships and interactions between variables. The relationship between environmental data (such as light intensity, noise level, wind speed) and physiological data (such as heart rate, breathing rate) is usually a complex nonlinear relationship. For example: the noise level may only significantly affect the heart rate after exceeding a certain threshold. The effect of wind speed on heart rate may be nonlinear and increasing. The random forest regression model is trained with training samples (environmental data and physiological data). In this process, the random forest regression model can reflect the association between physiological data and environmental data. For the multi-output random forest regression model, the feature importance of each training sample can be viewed separately.

[0080] S3 integrates the performance data, the physiological data, and the environmental data according to the influence degree among the performance data, the physiological data, and the environmental data to obtain the training state data;

[0081] Specifically, the performance data, physiological data, and environmental data are fused according to the obtained different types of weights to obtain the training state data.

[0082] In this step, the contribution of different types of data to the training strategy is obviously different. If these data are simply mixed and the weights are ignored, the model may overfit some data and ignore other more critical information. It is necessary to determine which are the main factors affecting the shooter's performance based on the various types of data in step S2. For example, if environmental data is the main factor affecting the shooter's performance, it is necessary to increase the weight of environmental data. According to the weights of other data, training status data = k1 performance data + k2 physiological data + k3 environmental data, k3> k1, k2.

[0083] S4 inputs the training status data into the training strategy model to obtain the shooter's optimized training strategy, and dynamically adjusts the simulation parameters of the current training environment according to the optimized training strategy.

[0084] Furthermore, the training strategy model is obtained through the following steps:

[0085] The performance data, physiological data, and environmental data are used as state variables, and the adjustment of simulation parameters of the training environment, training rhythm adjustment, and training content adjustment are used as action variables to build a training strategy model;

[0086] Construct a reward function based on shooting hit rate, shooting reaction time, heart rate, and aiming deviation to optimize the training strategy model.

[0087] Furthermore, the calculation formula for constructing the reward function based on the shooting hit rate, shooting reaction time, heart rate, and aiming trajectory is as follows:

[0088] R = α1 hit rate - α2 reaction time - α3 heart rate deviation + α4 aiming deviation,

[0089] Among them, α1, α2, )3, and α4 are weight parameters.

[0090] In this step, the fused data (training state data) is the result of performance data, physiological data, and environmental data after weight adjustment, which can comprehensively reflect the shooter's current training state. The training strategy model is based on this comprehensive state to analyze the shooter's current ability level, adaptability, and weak links, and formulate an optimized training strategy. Data fusion only integrates multiple data types into an input vector according to weights, but it cannot directly guide the optimization of the training strategy. The training strategy model not only requires the current training state data, but also needs to combine historical data (such as training progress) and training goals to formulate a dynamic adjustment strategy. Therefore, the reinforcement learning model is used to construct the training strategy model, and the performance data, physiological data, and environmental data are the current state variables. The reward function is the core of the reinforcement learning model and determines the direction of model optimization. Positive indicators (such as hit rate) have a positive contribution to the training effect, so they are represented by a positive sign in the reward function, and negative indicators (such as reaction time and heart rate deviation) have a negative impact on the training effect, so they are represented by a negative sign. Action variables are simulation parameter adjustment variables of the training environment, such as adjustment of light intensity; training rhythm adjustment, such as adjustment of the number of shootings; training content adjustment, such as changing static shooting to dynamic shooting.

[0091] Specifically, the training status data is input into the training strategy model to obtain environmental simulation parameters, training rhythm parameters, and training content parameters. The environmental simulation parameters include: light intensity, wind speed, and noise. The training rhythm parameters include: number of shootings, shooting intervals, and shooting duration. The training content parameters include: static shooting, dynamic shooting, and static and dynamic combined shooting.

[0092] In this step, environmental simulation parameters (such as light intensity, wind speed, and noise) directly affect the training difficulty and adaptability of the shooter. The model senses the impact of the current environment on the shooter by inputting the shooter's training status data. For example, when the noise level is too high and the hit rate drops, the model can identify the adverse effects of noise on the shooter. The model adjusts parameters such as light intensity, wind speed, and noise level through output actions to optimize the training environment. For example, when the shooter performs well in a quiet environment, the model can increase the noise level to train the shooter's anti-interference ability. The reward function evaluates whether the adjustment of environmental parameters helps to improve the training effect through indicators such as hit rate and reaction time. The model finds the optimal combination of environmental parameters for the shooter through multiple trial and error learning.

[0093] The training rhythm parameters (such as the number of shots, the interval between shots, and the duration of shots) determine the intensity and rhythm of the training. A reasonable training rhythm can improve the efficiency of training while preventing the shooter from becoming fatigued due to overtraining. The model senses whether the shooter is in the best condition by inputting the shooter's physiological data (such as heart rate deviation) and performance data (such as hit rate and reaction time). For example, if the shooter's heart rate is high, the model will identify that the shooter may be in a state of fatigue. The model dynamically adjusts the training rhythm by outputting actions, for example, increasing the training interval to help the shooter recover from fatigue. The reward function comprehensively evaluates the effect of adjusting the training rhythm. For example, if the hit rate is significantly improved after increasing the training intensity, the model will strengthen this adjustment strategy.

[0094] The training content parameters (such as static shooting, dynamic shooting, and static and dynamic combined shooting) determine the type and complexity of the training task. By dynamically adjusting the training content, the shooter's weak links can be optimized while improving the shooter's overall ability. The model determines whether the current training content is suitable by inputting the shooter's performance data (such as hit rate, aiming deviation) and training goals. For example: If the shooter performs stably in static shooting, but the hit rate decreases in dynamic shooting, the model will identify dynamic shooting as the shooter's weak link. The model dynamically adjusts the training content by outputting actions. For example: Increase the proportion of dynamic shooting to train the shooter's dynamic aiming ability. The reward function evaluates the effect of the adjustment of the training content. For example: If the hit rate gradually increases after adding dynamic shooting, the model will strengthen this adjustment strategy.

[0095] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0096] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1A device that provides the functions specified in a block or multiple blocks.

[0097] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0098] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0099] It should be noted that in the claims, any reference signs placed between brackets shall not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claim. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention may be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third etc. does not indicate any order. These words may be interpreted as names.

[0100] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0101] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

[0102] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0103] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms should not be understood as necessarily referring to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

Claims

1. A shooting training method based on training big data, characterized in that: include: obtaining performance data and physiological data of the shooter, shooting environment data, and calculating a performance score of the shooter based on the performance data of the shooter; According to the shooter's performance score, combined with the shooter's physiological data and shooting environment data, the influence degree between the performance data, physiological data and environmental data is analyzed; The performance data, the physiological data, and the environmental data are integrated according to the influence degree among the performance data, the physiological data, and the environmental data to obtain the training status data; The training status data is input into the training strategy model to obtain the shooter's optimized training strategy, and the current shooting training method is dynamically adjusted according to the optimized training strategy.

2. A shooting training method based on training big data according to claim 1, characterized in that: The acquisition of the shooter's performance data and physiological data, and the shooting environment data includes: Obtain the shooter's shooting accuracy, reaction time, and aiming deviation; Get the shooter’s heart rate and breathing rate; Get the light intensity, wind speed, and noise around the shooting.

3. A shooting training method based on training big data according to claim 2, characterized in that: The calculation formula for calculating the shooter's performance score based on the shooter's performance data is as follows: S = w1·hit rate-w2·reaction time-w3·aiming deviation, Among them, w1, w2, w3 are weight coefficients, and S is the performance score.

4. The shooting training method based on training big data according to claim 1, characterized in that: The analyzing the influence degree between the performance data, the physiological data, and the environmental data according to the performance score of the shooter combined with the physiological data of the shooter and the environmental data of the shooting includes: If the performance score is satisfactory, the advanced training weight will be obtained; If the performance score is unsatisfactory, determine whether the physiological data affects the performance data. If the physiological data does not affect the performance data, obtain the action training weight. Otherwise, determine whether the environmental data affects the physiological data. If the environmental data affects the physiological data, the environmental training weight is obtained, otherwise the psychological training weight is obtained.

5. A shooting training method based on training big data according to claim 4, characterized in that: The determining whether the physiological data is related to the performance-affecting data includes: A multivariate linear regression model was constructed based on physiological data and performance scores. The model formula is as follows: Y 表现 =β0+β1·heart rate+β2·respiratory rate+∈ Among them, Y 表现 is the performance score, β1 and β2 are regression coefficients, and ∈ is the error term; If any regression coefficient of β1 or β2 is greater than the significance threshold, the physiological data affects the performance data; otherwise, the physiological data does not affect the performance data.

6. A shooting training method based on training big data according to claim 4, characterized in that: The determining whether the environmental data affects the physiological data includes: A random forest regression model was constructed based on environmental data and physiological data. The model formula is as follows: Y 心率 , Y 呼吸频率 =f(light intensity, wind speed, noise), Among them, Y 心率 is the target heart rate, Y 呼吸频率 is the target respiratory rate, f is the random forest model; The environmental data is used as the input of the random forest regression model, and the physiological data is used as the output of the random forest regression model. The random forest regression model is trained to obtain the feature importance parameters of the random forest regression model. If any of the feature importance parameters of the random forest regression model is greater than the feature importance threshold, the environmental data affects the physiological data, otherwise the environmental data does not affect the physiological data.

7. The shooting training method based on training big data according to claim 5, characterized in that: The fusing of the performance data, the physiological data, and the environmental data according to the degree of influence between the performance data, the physiological data, and the environmental data to obtain the training state data includes: The performance data, physiological data, and environmental data are fused according to the obtained different types of weights to obtain the training status data.

8. The shooting training method based on training big data according to claim 2, characterized in that: The training strategy model is obtained through the following steps: The performance data, physiological data, and environmental data are used as state variables, and the adjustment of simulation parameters of the training environment, training rhythm adjustment, and training content adjustment are used as action variables to build a training strategy model; Construct a reward function based on shooting hit rate, shooting reaction time, heart rate, and aiming deviation to optimize the training strategy model.

9. A shooting training method based on training big data according to claim 8, characterized in that: The calculation formula for constructing the reward function based on the shooting hit rate, shooting reaction time, heart rate, and aiming trajectory is as follows: R = α1 hit rate - α2 reaction time - α3 heart rate deviation + α4 aiming deviation, Among them, α1, α2, α3, and α4 are weight parameters.

10. The shooting training method based on training big data according to claim 8, characterized in that: The step of inputting the training state data into the training strategy model to obtain the shooter's optimized training strategy includes: The training status data is input into the training strategy model to obtain environmental simulation parameters, training rhythm parameters, and training content parameters. The environmental simulation parameters include: light intensity, wind speed, and noise. The training rhythm parameters include: number of shootings, shooting intervals, and shooting duration. The training content parameters include: static shooting, dynamic shooting, and static and dynamic combined shooting.

Images