Athletic performance analysis method combined with wearable device

Abstract

The invention relates to the technical field of exercise analysis, and particularly discloses an exercise performance analysis method combined with wearable equipment, a first exercise set can be used in cooperation with warm-up time, and the first exercise set voluntarily builds a stepped training combination according to a pre-built action coefficient, so that when a trainer trains high-difficulty actions, the training efficiency is improved, and the training efficiency is improved. The process from easy to difficult is achieved through a series of gradient action combinations with action coefficient differences not exceeding a preset safety value, so that it is guaranteed that part of trainees directly performing training can obtain the basic warm-up effect, sport injuries are avoided as much as possible, the second exercise set is a set of target actions and supplementary actions, and the training efficiency is improved. Physical fatigue and psychological fatigue caused by continuous repetition can be avoided in the process of specially training the target action, and other actions can be reviewed at the same time.

Description

Technical Field

[0001] The present invention relates to the technical field of motion analysis. Specifically, it relates to a method for analyzing motion performance combined with wearable devices. Background Art

[0002] Badminton is one of the popular sports. However, when beginners start playing badminton and hope to improve their skills, they need long-term practice. Currently, the training process is usually arranged by teachers or coaches based on experience, controlling the training intensity. Obviously, the empirical arrangement is difficult to be specific to each person and each item, and the training efficiency is limited differently for different people. Especially when doing the training of receiving the ball, teachers or coaches usually arrange continuous practice of a single action. The practice of a single action is likely to cause physical and mental fatigue and reduce the training efficiency, while freely performing action training often fails to achieve the expected training effect. That is, there is currently a lack of a method for analyzing based on the motion performance of learners to intelligently arrange training content to improve training efficiency. Summary of the Invention

[0003] The purpose of the present invention is to provide a method for analyzing motion performance combined with wearable devices, and solve the following technical problems: How to analyze based on the motion performance of learners to intelligently arrange training content to improve training efficiency.

[0004] The purpose of the present invention can be achieved by the following technical solutions: A method for analyzing motion performance combined with wearable devices includes the following steps: Obtain the trainer's profile and collect the trainer's training data, then construct a training action library and set action coefficients based on the training data of the training actions. The action coefficient is in a proportional relationship with the physical exertion of the current trainer to complete the training action; Input at least one target action and select several training actions from the training action library based on the target action to construct a first action set. The first action set includes at least two training actions, and the difference between any two adjacent action coefficients in the sequence after arranging the action coefficients of each training action in descending order does not exceed a preset safety value; Set the warm-up time and start training in sequence from the training action with the smallest action coefficient according to the actions in the first action set within the warm-up time; Set the number of training rounds. Before the start of the number of training rounds, set supplementary actions, which together with the target action form a second motion set. Select the number of repetitions for the target action in the second motion set according to the preset repetition rule, and randomly select a target action to complete the number of repetitions and then perform at least one supplementary action; Adjust the action coefficients of the training actions in the training action library according to the training data of the target action in the training rounds.

[0005] Through the above technical solution, the first motion set and the second motion set are obtained based on the training performance of the trainer, and the subsequent training process of the trainer is controlled according to the first motion set and the second motion set. The first motion set of the present invention is used in cooperation with the warm-up time. According to the pre-constructed motion coefficients, the first motion set can automatically build a stepped training combination, so that when the trainer trains high-difficulty actions, through a series of gradient action combinations with the difference in motion coefficients not exceeding the preset safety value, the process from easy to difficult is realized, so as to ensure that some trainers who directly conduct training can obtain a basic warm-up effect and avoid sports injuries as much as possible. The second motion set is a set of target actions and supplementary actions, which can avoid physical and mental fatigue caused by continuous repetition during the process of special training on target actions, and at the same time take into account the review of other actions.

[0006] As a further technical solution of the present invention: the construction process of the motion coefficients includes: Through the formula:

[0007] Obtain the motion coefficient , where is the distance between the preset landing point of the current training action and the starting position of the trainer, is the basic distance value preset based on the body data in the trainer's file, is the bounce weight coefficient, is the expected bounce height of the current training action, is the preset basic height, is a judgment function about value. If then output 0, otherwise output , is the number of limbs of the limb data adopted, is the weight value of the i-th adopted limb data in the process of completing the current training action, and i is a non-zero positive integer not greater than n, is the score value calculated based on the limb data generated during the trainer's training.

[0008] As a further technical solution of the present invention: the process of obtaining the weight value of the i-th adopted limb data includes: Set a fluctuation range for each involved limb data based on the current training action, and then compare the corresponding data in the training data with the fluctuation range; If the corresponding data after a certain completion of the training action in the training data all fall within the fluctuation range, it is output that the limb corresponding to the current limb data is qualified in the current training action; Select the nearest The consecutive completion process of the i-th time, and output the number of times the i-th limb is qualified in the training action of that time , let .

[0009] As a further technical solution of the present invention: the scoring value The acquisition process includes: Through the formula:

[0010] Among them, and are preset proportion coefficients, which are constants, is the mean value of the current limb swing amplitude of the i-th limb in the current training action within a training phase, is the minimum value of the current limb swing amplitude of the i-th limb in the current training action within a training phase, is the maximum value of the current limb swing amplitude of the i-th limb in the current training action within a training phase, is the mean value of the completion time of the current limb swing amplitude of the i-th limb in the current training action within a training phase, is the maximum value of the completion time of the current limb swing amplitude of the i-th limb in the current training action within a training phase, is the minimum value of the completion time of the current limb swing amplitude of the i-th limb in the current training action within a training phase.

[0011] As a further technical solution of the present invention: the adjustment process of the action coefficient includes: Set index parameters for the current training action, and the index parameters are the limb data generated by several limbs in the current training action; If all the index parameters are judged to be qualified in the current training action, output the current training action as a qualified action, otherwise output it as an unqualified action; Sort a certain target action within a training phase according to the completion time to obtain a completion queue; Construct an identification sequence based on the unqualified actions, and the identification sequence is a segment sequence intercepted from the completion queue with the unqualified action as the center; Substitute all the completed target actions and the corresponding data generated in the identification sequence into the acquisition formula of the action coefficient and use the calculation result as the reference value of the current unqualified action; If the reference values of at least three unqualified actions are all greater than the action coefficient originally set for the target action, calculate and obtain a new action coefficient based on the training data within the corresponding training phase, overwrite the original action coefficient in the training action library and reset the fluctuation range; If the reference values ​​of at least three unqualified actions are all smaller than the action coefficient originally set for the target action, a new action coefficient is calculated based on the training data in the corresponding training phase, and the original action coefficient is overwritten in the training action library and the fluctuation range is reset.

[0012] Through the above technical scheme: an adjustment process of weight values, scoring values ​​and action coefficients is provided. The action coefficient of the present invention will continue to change with the training performance of the trainee. The better the trainee's performance of a certain training action is, the lower the action coefficient corresponding to the action will be after it is determined that the action coefficient needs to be adjusted. That is, real-time feedback control of the subsequent training process can be achieved based on the trainee's specific training performance, thereby ensuring that each exercise plan generated based on the exercise performance is highly relevant to the trainee's progress, and by monitoring the source of the indicator parameters of unqualified actions, targeted correction suggestions can be provided to achieve scientific and rapid progress.

[0013] As a further technical solution of the present invention: the process of selecting a supplementary action includes: The action coefficient selection interval is set according to the exercise time and the last completed training action, and the training actions whose action coefficients fall into the selection interval are used as the supplementary action set; When a supplementary action is needed, a training action is randomly selected from the supplementary action set as the supplementary action.

[0014] Through the above technical scheme: a scheme for selecting supplementary actions is provided. The supplementary actions of the present invention are used as adjustments and reviews in the process of repeated training of target actions. They are to avoid physical and mental fatigue caused by continuous repetition in the process of special training of target actions, and to review other actions at the same time. The setting of the supplementary actions is based on a certain range of the action coefficient of the target action. The exercise difficulty is equivalent to or even lower than that of the target action, and will not burden the training of the target action.

[0015] As a further technical solution of the present invention: the preset repetition rules include: Set a maximum number of repetitions for each exercise; The number of repetitions is randomly selected from zero to the maximum number of repetitions.

[0016] As a further technical solution of the present invention: the process of resetting the fluctuation range includes: Obtain the difference between the new action coefficient and the covered action coefficient, and record it as the adjustment index; Obtaining adjustment values ​​from a preset standard table according to the adjustment index; The endpoint values ​​of the original fluctuation range are added to the adjustment value to obtain a new fluctuation range to complete the resetting of the fluctuation range.

[0017] As a further technical solution of the present invention: it also includes: After a training round is completed, the training data of all target actions are obtained, and the corresponding data are substituted into the formula for obtaining the action coefficient and the calculation results are recorded as the training index of the corresponding target action; like , then it is recommended to proceed to the next training round.

[0018] Beneficial effects of the present invention: The first movement set of the present invention can be used in conjunction with the warm-up time. The first movement set builds a step-by-step training combination based on the pre-constructed action coefficients, so that when the trainees are training difficult movements, a series of gradient action combinations whose action coefficient differences do not exceed the preset safety values ​​can be used to achieve a process from easy to difficult, thereby ensuring that some trainees who directly train can obtain a basic warm-up effect and avoid sports injuries as much as possible. The second movement set is a collection of target movements and supplementary movements, which can avoid physical and mental fatigue caused by continuous repetition in the process of special training of target movements, while taking into account the review of other movements.

[0019] The action coefficient of the present invention will continue to change with the training performance of the trainee. The better the trainee's performance on a certain training action, the lower the action coefficient corresponding to the action will be after it is determined that the action coefficient needs to be adjusted. That is, real-time feedback control of the subsequent training process can be achieved based on the trainee's specific training performance, thereby ensuring that each exercise plan generated based on the exercise performance is highly relevant to the trainee's progress, and by monitoring the source of the indicator parameters of unqualified actions, targeted correction suggestions can be provided to achieve scientific and rapid progress.

[0020] The supplementary action of the present invention is used as an adjustment and review in the repeated training process of the target action. It is to avoid physical and mental fatigue caused by continuous repetition in the process of special training of the target action, and to review other actions at the same time. The setting of the supplementary action is based on a certain range of the action coefficient of the target action. The exercise difficulty is equivalent to or even lower than that of the target action, and will not burden the training of the target action. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The present invention will be further described below in conjunction with the accompanying drawings.

[0022] Figure 1 It is a flow chart of the overall steps of the present invention; Figure 2 It is a flow chart of the steps of the action coefficient adjustment process of the present invention. DETAILED DESCRIPTION

[0023] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0024] Please refer to Figure 1 As shown, in one embodiment, a method for analyzing sports performance in combination with a wearable device is provided, including the following steps: S100. Obtain the trainer's profile and collect the trainer's training data, then construct a training action library and set action coefficients based on the training data of the training actions. The action coefficient is in a proportional relationship with the physical consumption of the current trainer to complete the training action. In this embodiment, the proportional relationship is a direct proportional relationship, that is, the larger the action coefficient, the greater the difficulty and consumption of completing the training action. The trainer's profile is the body data input by the trainer himself or uploaded uniformly, and at least includes the trainer's gender, height, weight, and the length data of the upper arm, forearm, thigh, and calf. The training data at least includes limb data and camera data from the wearable device; S200. Input at least one target action and select several training actions from the training action library based on the target action to construct a first action set. The first action set consists of the target action and the selected several training actions. The first action set includes at least two training actions, and the difference between any two adjacent action coefficients in the sequence after arranging the action coefficients of each training action in descending order does not exceed a preset safety value. In the selection process of the preset safety value, different training actions are first classified, and the average value of the differences between the action coefficients of different training actions in adjacent levels is used as the preset safety value. The classification is based on the action coefficient, and different classification methods result in different action coefficients; S300. Set the warm-up time and start training in sequence from the training action with the smallest action coefficient in the first action set within the warm-up time; S400. Set the number of training rounds. The number of training rounds needs to start after the warm-up time ends. Supplementary actions are set before the start of the training rounds and together with the target action form a second exercise set. The second exercise set includes the target action and the supplementary action. The number of repetitions for the target action in the second exercise set is selected according to a preset repetition rule. After randomly selecting a target action to complete the number of repetitions, at least one supplementary action is performed. In each training round, the target action is randomly selected only once. When all the target actions have been randomly selected once in a round, enter the next training round; S500. Adjust the action coefficients of the training actions in the training action library according to the training data of the target actions in the training rounds.

[0025] In this embodiment, a first exercise set and a second exercise set are obtained based on the exercise performance of the trainer, and the subsequent training process of the trainer is controlled according to the first exercise set and the second exercise set. The first exercise set of the present invention is used in cooperation with the warm-up time. According to the pre-constructed action coefficients, the first exercise set can build a stepped training combination by itself, so that when the trainer trains difficult actions, through a series of gradient action combinations with the difference of action coefficients not exceeding the preset safety value, the process from easy to difficult can be realized, so as to ensure that some trainers who directly start training can obtain a basic warm-up effect and avoid sports injuries as much as possible. The second exercise set is a set of target actions and supplementary actions, which can avoid physical and mental fatigue caused by continuous repetition during the process of special training on target actions, and at the same time take into account the review of other actions.

[0026] The construction process of the action coefficients includes: Through the formula:

[0027] Obtain the action coefficients , where is the distance between the preset landing point of the current training action and the starting position of the trainer, is the basic distance value preset based on the body data in the trainer's file, which is a constant. For example, for a female trainer with a height of 1.6m and a weight of 42kg, the basic distance value can be set to 1.2m. It should be noted that in the present invention, a ball feeder is used to serve the ball, and the serving angle, landing point and the starting position of the trainer are specified, and the trainer is reminded of the training actions required during the current serving process by the screen. is the bounce weight coefficient, which is the ratio of the energy consumed by the bounce action under the average bounce height of the current training action to the total efficiency energy of completing the current training action, is the expected bounce height of the current training action, is the preset basic height, is a judgment function about value. If then output 0, otherwise output , is the number of limbs of the limb data used, is the weight value of the i-th limb data used in the process of completing the current training action, where i is a non-zero positive integer not greater than n, is the scoring value calculated based on the limb data generated by the trainer during training.

[0028] In this embodiment, a process for obtaining an action coefficient is provided. The action coefficient of the present invention is calculated and obtained from data generated during multiple training processes of an action, and is used to describe the proportional relationship between the action coefficient and the physical exertion of the current trainer in completing the training action.

[0029] Among them, the process for obtaining the weight value of the limb data adopted in the i-th step includes: Based on the current training action, a fluctuation range is set for each involved limb data, and then the corresponding data in the training data is compared with the fluctuation range; If the corresponding data after a certain completion of the training action in the training data all fall within the fluctuation range, it is output that the limb corresponding to the current limb data is qualified in the current training action. The limb data at least includes the swing amplitude and the time taken to complete the amplitude. The corresponding fluctuation range is set based on the standard action. For example, the fluctuation range of the swing amplitude of the forearm corresponding to the hand holding the ball is (27°, 35°), and the fluctuation range of the time taken to complete the swing amplitude is (0.8s, 1.5s). If both the swing amplitude of the forearm and the time taken to complete the swing amplitude in the current training action fall within the corresponding fluctuation range, it is determined that the corresponding limb is qualified in the current training action; Select the most recent continuous completion process of the current action within a training stage, and output the number of times that the i-th limb is qualified in the current training action , and let . A training stage includes a warm-up time and subsequent training rounds.

[0030] Among them, the process for obtaining the scoring value includes: Through the formula:

[0031] Among them, and are preset proportion coefficients, which are constants and are selected by the expert group based on experience. is the average value of the current limb swing amplitude of the i-th limb in the current training action within a training stage. is the minimum value of the current limb swing amplitude of the i-th limb in the current training action within a training stage. is the maximum value of the current limb swing amplitude of the i-th limb in the current training action within a training stage. is the average value of the completion time of the current limb swing amplitude of the i-th limb in the current training action within a training stage. is the maximum value of the completion time of the current limb swing amplitude of the i-th limb in the current training action within a training stage. is the minimum value of the completion time of the current limb swing amplitude of the i-th limb in the current training action within a training stage.

[0032] Reference Figure 2 , where the adjustment process of the action coefficient includes: S510. Set index parameters for the current training action. The index parameters are the limb data generated by several limbs in the current training action, and the index parameters of different training actions are different. Specifically, for example, in the badminton action of forehand high clear, it mainly involves the forearm corresponding to the ball-holding hand and the same-side thigh; S520. If all index parameters are judged to be qualified in the current training action, output the current training action as a qualified action; otherwise, output it as an unqualified action. It should be noted that unqualified actions do not only refer to training actions with negative development, but also include the training actions of trainers with positive development. For example, the training quality of a trainer in a certain training action continuously exceeds or is continuously lower than the current average level; S530. Sort a target action within a training phase according to the completion time to obtain a completion queue; S540. Construct an identification sequence based on unqualified actions. The identification sequence is a segment sequence intercepted from the completion queue with the unqualified action as the center, and the intercepted length is set according to the action difficulty. The greater the action difficulty, the shorter the intercepted length; S550. Substitute all the completed target actions and their corresponding data in the identification sequence into the acquisition formula of the action coefficient and use the calculation result as the reference value of the current unqualified action; S560. If the reference values of at least three unqualified actions are all greater than the action coefficient originally set for the target action, calculate and obtain a new action coefficient based on the training data within the corresponding training phase, overwrite the original action coefficient in the training action library, and reset the fluctuation range; S570. If the reference values of at least three unqualified actions are all less than the action coefficient originally set for the target action, calculate and obtain a new action coefficient based on the training data within the corresponding training phase, overwrite the original action coefficient in the training action library, and reset the fluctuation range. It should be noted that Figure 2 the judgment conditions in are "if the reference values of at least three unqualified actions are all greater than the action coefficient originally set for the target action" and "if the reference values of at least three unqualified actions are all less than the action coefficient originally set for the target action".

[0033] In this embodiment, an adjustment process for weight values, score values ​​and action coefficients is provided. The action coefficient of the present invention will continue to change with the training performance of the trainee. The better the trainee's performance of a certain training action, the lower the action coefficient corresponding to the action will be after it is determined that the action coefficient needs to be adjusted. That is, real-time feedback control of the subsequent training process can be achieved based on the trainee's specific training performance, thereby ensuring that each exercise plan generated based on the exercise performance is highly relevant to the trainee's progress, and by monitoring the source of indicator parameters of unqualified actions, targeted correction suggestions can be provided to achieve scientific and rapid progress.

[0034] The process of selecting a supplementary action includes: The action coefficient selection interval is set according to the exercise time and the last completed training action, and the training actions whose action coefficients fall into the selection interval are used as the supplementary action set; When a supplementary action is needed, a training action is randomly selected from the supplementary action set as the supplementary action.

[0035] In this embodiment, a scheme for selecting supplementary actions is provided. The supplementary actions of the present invention are used as adjustments and reviews during the repeated training of the target action. They are intended to avoid physical and mental fatigue caused by continuous repetition during the special training of the target action, while reviewing other actions. The setting of the supplementary actions is based on a certain range of the action coefficient of the target action. The exercise difficulty is equivalent to or even lower than that of the target action, and will not impose a burden on the training of the target action.

[0036] The preset recurrence rules include: Set the maximum number of repetitions for each training exercise. The maximum number of repetitions is set according to the action coefficient of the training exercise. For example, the maximum number of repetitions for training exercises with an action coefficient of 1.1-1.3 is set to 20, the maximum number of repetitions for training exercises with an action coefficient of 1.3-1.5 is set to 15, and so on. The number of repetitions is randomly selected from zero to the maximum number of repetitions.

[0037] The process of resetting the volatility range includes: Obtain the difference between the new action coefficient and the covered action coefficient, and record it as the adjustment index; Obtaining adjustment values ​​from a preset standard table according to the adjustment index; The endpoint values ​​of the original fluctuation range are added to the adjustment value respectively to obtain a new fluctuation range to complete the reset of the fluctuation range. The reset fluctuation range is a new fluctuation range that covers the original fluctuation range. By setting a new fluctuation range, trainees whose training status changes can select ranges corresponding to new indicator parameters as the status changes during the process of judging qualified movements, thereby achieving real-time feedback.

[0038] It further includes: After a training round is completed, obtain the training data of all target actions, substitute the corresponding data into the acquisition formula of the action coefficient, and record the calculation results as the training indices of the corresponding target actions respectively; If , then it is determined that it is recommended to proceed to the next training round.

[0039] The above has described an embodiment of the present invention in detail, but the content described is only a preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equivalent changes and improvements made in accordance with the scope of the present invention application shall still fall within the scope covered by the patent of the present invention.

Claims

1. A sports performance analysis method combined with a wearable device, characterized in that: The steps include: Obtaining a trainer profile and collecting the trainer's training data, then constructing a training action library and setting an action coefficient based on the training data of the training action, wherein the action coefficient is proportional to the physical exertion of the current trainer to complete the training action, and the training data at least includes limb data and camera data from a wearable device; Input at least one target action and select a number of training actions from a training action library based on the target action to construct a first action set, wherein the first action set includes at least two training actions, and the difference between any two adjacent action coefficients in a sequence after the action coefficients of the training actions are arranged in order of magnitude does not exceed a preset safety value, and the process of obtaining the preset safety value includes classifying different training actions, and taking the average of the differences between the action coefficients of different training actions in adjacent levels as the preset safety value; Set the warm-up time and train in sequence according to the movements in the first movement set during the warm-up time, starting from the training movement with the smallest movement coefficient; Setting training rounds, setting supplementary actions before the start of the training rounds, and forming a second movement set together with the target action, selecting the number of repetitions for the target action in the second movement set according to a preset repetition rule, and randomly selecting a target action to perform at least one supplementary action after completing the number of repetitions; The action coefficients of the training actions in the training action library are adjusted according to the training data of the target action in the training round.

2. A sports performance analysis method combined with a wearable device according to claim 1, characterized in that: The construction process of the action coefficient includes: By formula: Get the action coefficient ,in The distance between the preset landing point of the current training action and the trainer's starting position. It is the basic distance value preset based on the body data in the trainer's file. is the bounce weight coefficient, is the expected jumping height of the current training action, is the preset base height, About The judgment function of the value, if Then output 0, otherwise output , is the number of limbs for which limb data is used, is the weight value of the limb data used by the i-th limb in the process of completing the current training action, i is a non-zero positive integer not greater than n, It is a score value calculated based on the limb data generated by the trainee during training.

3. A sports performance analysis method combined with a wearable device according to claim 2, characterized in that: The process of obtaining the weight value of the limb data used for the i-th time includes: Based on the current training action, a fluctuation range is set for each limb data involved, and then the corresponding data in the training data is compared with the fluctuation range; If the corresponding data after a training action is completed in a certain training data falls within the fluctuation range, then the limb corresponding to the current limb data is outputted as qualified in the current training action; Select the current action within a training phase The process is completed continuously and the number of times the i-th limb passes the training action is output. ,make .

4. The method for analyzing sports performance in combination with a wearable device according to claim 2, characterized in that: Rating value The acquisition process includes: By formula: in, and is the preset specific gravity coefficient, which is a constant. is the average swing amplitude of the i-th limb in the current training action during a training phase, is the minimum swing amplitude of the ith limb in the current training action within a training phase, is the maximum swing amplitude of the ith limb in the current training action within a training phase, is the mean completion time of the swing amplitude of the i-th limb in the current training action within a training stage, is the maximum completion time of the swing amplitude of the i-th limb in the current training action within a training stage, It is the minimum completion time of the swing amplitude of the i-th limb in the current training action within a training stage.

5. The method for analyzing sports performance in combination with a wearable device according to claim 1, characterized in that: The adjustment process of the action coefficient includes: Setting indicator parameters for the current training action, where the indicator parameters are limb data generated by several limbs in the current training action; If all the index parameters are judged to be qualified in the current training action, the current training action is output as a qualified action, otherwise it is output as an unqualified action; Sort a target action in a training phase by completion time to obtain a completion queue; Building a recognition sequence based on the unqualified action, the recognition sequence is a sequence intercepted from the completion queue with the unqualified action as the center; Based on all the completed target actions in the recognition sequence and the corresponding data generated by them, the action coefficient acquisition formula is introduced and the calculation result is used as the reference value of the current unqualified action; If the reference values ​​of at least three unqualified actions are greater than the action coefficients originally set for the target action, a new action coefficient is calculated based on the training data in the corresponding training phase, and the original action coefficient is overwritten in the training action library and the fluctuation range is reset; If the reference values ​​of at least three unqualified actions are all smaller than the action coefficient originally set for the target action, a new action coefficient is calculated based on the training data in the corresponding training phase, and the original action coefficient is overwritten in the training action library and the fluctuation range is reset.

6. The sports performance analysis method combined with a wearable device according to claim 1, characterized in that: The process of selecting a supplementary action includes: The action coefficient selection interval is set according to the exercise time and the last completed training action, and the training actions whose action coefficients fall into the selection interval are used as the supplementary action set; When a supplementary action is needed, a training action is randomly selected from the supplementary action set as the supplementary action.

7. The method for analyzing sports performance in combination with a wearable device according to claim 1, characterized in that: The preset recurrence rules include: Set a maximum number of repetitions for each exercise; The number of repetitions is randomly selected from zero to the maximum number of repetitions.

8. The method for analyzing sports performance in combination with a wearable device according to claim 5, characterized in that: The process of resetting the volatility range includes: Obtain the difference between the new action coefficient and the covered action coefficient, and record it as the adjustment index; Obtaining adjustment values ​​from a preset standard table according to the adjustment index; The endpoint values ​​of the original fluctuation range are added to the adjustment value to obtain a new fluctuation range to complete the resetting of the fluctuation range.

9. The method for analyzing sports performance in combination with a wearable device according to claim 1, characterized in that: Also includes: After a training round is completed, the training data of all target actions are obtained, and the corresponding data are substituted into the formula for obtaining the action coefficient and the calculation results are recorded as the training index of the corresponding target action; like , then it is recommended to proceed to the next training round.

Images