Portable console intelligent training system and method

Abstract

The invention provides an intelligent training system and method for a portable control console, and relates to the technical field of man-machine interaction and intelligent training, and the method comprises the steps of original trajectory data collection, operation feature extraction, capability score vector construction, user capability vector generation, capability trend analysis and personalized training path construction. The method comprises the following steps: acquiring attitude angular velocity, displacement acceleration and key data of a user in a training process through a portable console, extracting multi-dimensional operation characteristics, performing alignment and evaluation in combination with a scoring rule and a dynamic time warping method, and generating ability scores and user ability vectors; and further analyzing a capability development trend, identifying a capability short board, and matching a training task to generate a personalized path. The technical problems that an existing training system does not have portability, data acquisition dimensions in the training process are insufficient, capability evaluation is rough, and trend analysis and personalized training path recommendation are lacked are solved.

Description

Technical Field

[0001] The present application relates to the field of human-computer interaction and intelligent training technology, and in particular to a portable console intelligent training system and method. Background Art

[0002] With the development of human-computer interaction technology and intelligent training systems, virtual training and operational capability assessment technologies are being introduced in an increasing number of fields to improve training efficiency and enhance personnel operational capabilities. In scenarios such as complex equipment operation, skill training, and remote control, an operator's operational stability, response speed, and control accuracy have become important indicators for assessing their operational proficiency. Traditional training systems often rely on fixed equipment, manual scoring, or crude statistical data. These systems suffer from poor portability, a single evaluation dimension, and difficulty in continuously tracking changes in user capabilities, making them unable to meet the needs of precise and personalized training.

[0003] Some existing systems have attempted to incorporate trajectory analysis, sensor data collection, and scoring mechanisms for operational capability assessment. However, these systems generally lack trend modeling and capability gap identification mechanisms based on historical user data, making it difficult to develop differentiated training strategies for users. Furthermore, fixed training platforms limit the flexibility of usage scenarios and hinder the implementation of multi-location, multi-frequency daily training.

[0004] Therefore, there is an urgent need for an intelligent training system with a portable structure that can collect multi-dimensional operation data and realize dynamic ability assessment and personalized training recommendations to improve the accuracy, adaptability and practicality of training. Summary of the Invention

[0005] The purpose of this application is to provide a portable console intelligent training system and method to solve the technical problems existing in the prior art, such as the lack of portability of the training system, insufficient dimensions of data collection during the training process, rough ability assessment, and lack of trend analysis and personalized training path recommendation.

[0006] In view of the above problems, the present application provides a portable console intelligent training system and method.

[0007] In the first aspect, the present application provides a portable console intelligent training system for executing a portable console intelligent training method, including: an original trajectory data acquisition module, the original trajectory data acquisition module is used to collect original trajectory data of a user performing a training task on a portable console, the portable console has posture perception and key input capabilities, and the original trajectory data includes posture angular velocity, displacement acceleration, key time series and operation instruction number; an operation feature parameter set extraction module, the operation feature parameter set extraction module is used to preprocess the original trajectory data, segment based on the key time series, obtain multiple trajectory segments corresponding to user operation behaviors, and extract an operation feature parameter set for each trajectory segment, the operation feature parameter set includes posture fluctuation, operation stability, response delay and trajectory coherence; an ability scoring vector construction module, the ability scoring vector construction module is used to match the operation feature parameter set with the scoring rule, determine the reference standard trajectory sequence, and use the dynamic time warping DTW method for time series alignment. , calculate the scoring index, and construct the capability scoring vector in combination with the operation feature parameter set; a user capability vector generation module, the user capability vector generation module is used to normalize each scoring dimension and embed a time tag through the capability scoring vector to generate a user capability vector, the user capability vector is the user's comprehensive operation level in the dimensions of operation stability, response speed, control accuracy and execution accuracy; a capability development trend map generation module, the capability development trend map generation module is used to call the user capability vectors of multiple time points in the user historical training record database, construct a capability evolution time series, and generate a capability development trend map based on the change slope and fluctuation trend of each capability dimension; a personalized training path construction module, the personalized training path construction module is used to identify the user's capability shortboard dimension through the capability dimension with a downward or slow growth trend in the capability development trend map, and screen the training tasks corresponding to the capability shortboard dimension from the training task database to generate a personalized training task path for the capability shortboard dimension.

[0008] In the second aspect, the present application provides a portable console intelligent training method, which is implemented through a portable console intelligent training system, including: collecting original trajectory data of a user performing a training task on a portable console, wherein the portable console has posture perception and key input capabilities, and the original trajectory data includes posture angular velocity, displacement acceleration, key time series and operation instruction number; preprocessing the original trajectory data, segmenting it based on the key time series, obtaining multiple trajectory segments corresponding to user operation behaviors, extracting an operation feature parameter set for each trajectory segment, wherein the operation feature parameter set includes posture fluctuation, operation stability, response delay and trajectory coherence; matching the operation feature parameter set with the scoring rule, determining the reference standard trajectory sequence, and using the dynamic time warping DTW method for timing Align, calculate scoring indicators, and construct a capability scoring vector in combination with the operation feature parameter set; through the capability scoring vector, normalize each scoring dimension and embed time tags to generate a user capability vector, which is the user's comprehensive operation level in the dimensions of operation stability, response speed, control accuracy and execution accuracy; call the user capability vectors at multiple time points in the user's historical training record database, construct a capability evolution time series, and generate a capability development trend map based on the change slope and fluctuation trend of each capability dimension; through the capability dimensions with a downward or slow growth trend in the capability development trend map, identify the user's capability shortboard dimension, and screen the training tasks corresponding to the capability shortboard dimension from the training task database to generate a personalized training task path for the capability shortboard dimension.

[0009] The technical solution provided in this application has at least the following technical effects or advantages: Multi-source operation data is collected through a portable operating console, key features are extracted and capability scores are quantified. Trend analysis is combined to identify capability shortcomings, and personalized training paths are generated based on task adaptability, thereby achieving portable, efficient and accurate capability assessment and personalized training recommendations, and improving the intelligence level and practicality of the training system.

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, which can be implemented in accordance with the contents of the description, and to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically listed below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in this application or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely exemplary, and a person of ordinary skill in the art can obtain other drawings based on the provided drawings without creative work.

[0012] Figure 1 This is a structural diagram of a portable console intelligent training system for this application; Figure 2 This is a flow chart of a portable console intelligent training method for this application.

[0013] Explanation of the accompanying symbols: original trajectory data acquisition module 1, operation feature parameter set extraction module 2, ability score vector construction module 3, user ability vector generation module 4, ability development trend map generation module 5, personalized training path construction module 6. DETAILED DESCRIPTION

[0014] This application provides a portable console intelligent training system and method, addressing existing technical issues such as the lack of portability, insufficient data collection during training, crude ability assessment, and a lack of trend analysis and personalized training path recommendations. This system achieves portable, efficient, and accurate ability assessment and personalized training recommendations, enhancing the intelligence and practicality of the training system.

[0015] Below, the technical solutions in this application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of this application, rather than all the embodiments of this application. It should be understood that this application is not limited to the example embodiments described herein. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. It should also be noted that, for the convenience of description, only the parts related to this application, rather than all of them, are shown in the accompanying drawings.

[0016] For example, see the attached Figure 1 The present application provides a portable console intelligent training system, specifically comprising: The original trajectory data acquisition module 1 is used to collect the original trajectory data of the user when performing training tasks on a portable console. The portable console has posture perception and key input capabilities. The original trajectory data includes posture angular velocity, displacement acceleration, key time sequence and operation instruction number.

[0017] Specifically, multi-dimensional raw trajectory data is collected while a user performs training tasks on a portable console. The console integrates a posture sensing module (gyroscope, accelerometer) and a key input unit, enabling real-time acquisition of the user's posture angular velocity data (i.e., angular velocity around each axis), displacement acceleration data (i.e., acceleration information in each direction), key press time series (i.e., timestamp of each key press event), and operation command number (indicating the currently executed task or action instruction). The raw trajectory data is stored as a time series, with posture angular velocity and displacement acceleration uniformly recorded as a three-axis vector sequence at the sampling frequency. The key press time series records the key press code and its trigger time in an event-driven manner, and the operation command number is bound to the trajectory segment identifier.

[0018] Operation feature parameter set extraction module 2 is used to preprocess the raw trajectory data, segment it based on the key press time sequence, obtain multiple trajectory segments corresponding to user operation behaviors, and extract an operation feature parameter set for each trajectory segment. The operation feature parameter set includes posture fluctuation, operation stability, response delay and trajectory coherence.

[0019] Furthermore, the raw trajectory data is pre-processed and segmented based on the key time sequence to obtain multiple trajectory segments corresponding to user operation behaviors, and an operation feature parameter set is extracted for each trajectory segment, including: Preprocessing the raw trajectory data includes performing weighted median filtering to denoise the attitude angular velocity and displacement acceleration data, aligning the sampling time using linear interpolation, and normalizing the amplitude of the trajectory spatial components based on the z-score standard deviation normalization method to generate normalized trajectory data; Based on the key pressing time points and the operation instruction number change time points in the key time series, the normalized trajectory data is segmented into time series to extract multiple trajectory segments corresponding to the user operation; For each of the trajectory segments, a temporal correspondence between trajectory changes and instruction responses is determined based on the key press time sequence within the corresponding time period, and an operation feature parameter set is extracted.

[0020] Furthermore, for each of the trajectory segments, determining the temporal correspondence between trajectory changes and command responses based on the key press time sequence within the corresponding time period, and extracting an operation feature parameter set, includes: Based on the component sequence of the attitude angular velocity in the X, Y, and Z axes in the trajectory segment, the mean, standard deviation, and coefficient of variation of each dimension are calculated. The mean is used to reflect the average attitude direction, and the standard deviation and coefficient of variation are used to characterize the degree of fluctuation of attitude control, thereby obtaining attitude fluctuation characteristics; Based on the components of the displacement acceleration in the X, Y, and Z axes in the trajectory segment, the amplitude range is calculated, and the degree of acceleration fluctuation during the operation is evaluated by combining the standard deviation of each axis with the acceleration change rate at adjacent time points to obtain the operational stability characteristics; Based on the posture angular velocity component in the trajectory segment, a sliding window method is used to analyze the angular velocity change rate of adjacent moments. When the change rate exceeds a preset mutation threshold, the corresponding moment is determined to be a posture mutation time point. The posture mutation time point is matched with the key triggering time point in the key time sequence, and the time difference between the two is calculated to obtain the response delay feature; Based on the accumulation of spatial offset distances between consecutive trajectory points in the trajectory segment, the consistency of trajectory point direction and amplitude changes in adjacent time periods is compared, and the trajectory turning angle change rate and path smoothness index are calculated to obtain trajectory continuity features.

[0021] Specifically, the operation feature parameter set extraction module is used to pre-process the original trajectory data collected by the portable console and extract the operation features. The original trajectory data includes: three-axis attitude angular velocity sequence , three-axis displacement acceleration series , as well as the key time sequence and operation command number sequence. First, weighted median filtering is performed on the attitude angular velocity and acceleration data to remove high-frequency noise. Then, linear interpolation is used to align the sampling points of different channels. Then, the z-score normalization method is applied to the data of each sensor channel for amplitude normalization. The processing formula is: in, Represents the value at any moment in the original data sequence, Represents the mean value of the channel over the entire time series, and the calculation formula is: , It represents the standard deviation and is calculated as: =1 to ), the normalization process converts the trajectory data of different users or different sampling conditions into a uniform distribution with a mean of 0 and a standard deviation of 1. Next, based on the time points of the key press events and operation instruction number changes recorded in the key time series, the normalized trajectory data is divided into multiple trajectory segments, each segment representing an independent operation behavior of the user. In each trajectory segment, the following operation feature parameters are extracted: posture fluctuation feature: the normalized sequence of posture angular velocity 、 、 , calculate the mean respectively , standard deviation and coefficient of variation , where the coefficient of variation is defined as: Used to quantify the stability of the attitude control process, The larger the value, the more severe the fluctuation and the more unstable the posture. Operational stability characteristics: 、 、 , calculate the normalized RMS on each axis , defined as: ( =1 to )in Indicates the number of sampling points in the trajectory segment. >Preset stability threshold (e.g. 1.5), then it is considered that there is significant fluctuation on the axis and the operation is unstable. This threshold can be set by statistical samples of standard user groups. In addition, the maximum amplitude range on each axis is calculated, that is, , as a supplementary criterion. Response delay characteristics: Sliding window analysis of angular velocity change rate , If the mutation threshold is exceeded (e.g. 2.0), it is recorded as a posture mutation point. . Compare this time point with the most recent key trigger time point Perform matching and define the response delay as: This indicator is used to evaluate the timeliness of instruction execution. Trajectory continuity feature: based on the spatial distance between adjacent sampling points in the trajectory segment ,in, 、 、 Represents the trajectory at time The three-axis coordinate components of express The spatial displacement between a moment and the previous moment. The total length of the path is obtained by accumulating the total length of the path in the entire trajectory segment. Furthermore, the rate of change of the trajectory direction at consecutive moments, that is, the rate of change of the turning angle, is calculated. The variance of the angles between trajectory points is calculated as a path smoothness indicator (e.g., the angle variance between trajectory points) to measure whether the trajectory curvature changes smoothly and consistently. The path length, turning angle change rate, and angle variance are combined to form a trajectory coherence feature. This feature characterizes the smoothness and stability of the user's actions during operation. Higher coherence indicates a more natural and mature operation, while lower coherence may indicate instability or hesitation.

[0022] The capability scoring vector construction module 3 is used to match the operation feature parameter set with the scoring rules, determine the reference standard trajectory sequence, use the dynamic time warping (DTW) method to perform time series alignment, calculate the scoring index, and construct the capability scoring vector in combination with the operation feature parameter set.

[0023] Further, including: Match the user's operational feature parameter set extracted in the current training task with the scoring rules in the training scoring standard database. Determine the reference standard trajectory sequence by comparing the operational feature index with the threshold set by the scoring rule item to see if it falls within the scoring level range. The scoring rules include trajectory deviation tolerance threshold, posture control deviation threshold, time rhythm tolerance, and operational stability threshold. Dynamic time warping (DTW) is used to align the user trajectory segments with the reference standard trajectory sequence to obtain an alignment path. Based on the alignment path, scoring indicators are calculated, including trajectory space deviation score, time rhythm matching score, and posture matching score, and matching degree correction is performed according to the scoring rules; The operation characteristic parameter set is compared with the scoring rules, and the posture fluctuation score, operation stability score, response delay score and trajectory coherence score are determined according to the relationship between the characteristic value of each dimension and the set threshold; The trajectory spatial deviation score, time rhythm matching score, posture matching score, posture fluctuation score, operation stability score, response delay score and trajectory coherence score are weighted and combined according to preset weight parameters to generate an ability score vector.

[0024] Furthermore, based on the alignment path, scoring indicators are calculated, including trajectory space deviation score, time rhythm matching score, and posture matching score, and matching degree correction is performed according to the scoring rules, including: Based on each pair of matching trajectory points in the alignment path , respectively obtain the first The location of the sampling points The corresponding reference standard trajectory sequence Alignment points , and calculate the trajectory space deviation score , the formula is: ; in is the total number of trajectory point pairs in the alignment path, For the For matching trajectory point pairs, For the k The spatial distance between points, The trajectory deviation tolerance threshold set in the scoring rule; Extract the timestamp difference of each pair of matching points from the alignment path and calculate the tempo matching score , and its calculation formula is: ; in, is the total number of trajectory point pairs in the alignment path, and The user trajectory segment and the reference standard trajectory sequence are respectively The timestamp of the matching points, is the time rhythm tolerance; Based on the alignment path, the attitude angle parameter difference corresponding to each pair of trajectory points is extracted, and the attitude angle parameter difference includes the pitch angle between the user trajectory segment and the reference standard trajectory sequence. , yaw angle and roll angle differences in Calculate pose matching score , and its calculation formula is: ; in, is the total number of trajectory point pairs in the alignment path, Respectively The pitch, yaw and roll angles of each user trajectory segment, For the The pitch, yaw and roll angles of a reference standard trajectory sequence, is the preset maximum attitude deviation tolerance value; The trajectory spatial deviation score , time rhythm matching score Matching score with posture Together they constitute the dimension elements of the capability score vector.

[0025] Specifically, the capability scoring vector construction module is used to match the set of operational feature parameters extracted by the user in the portable console training task with the preset scoring rules, and to achieve temporal alignment of the user trajectory and the standard trajectory based on the dynamic time warping DTW method to quantify their operational capability performance. First, the operational feature parameter set obtained by the user in the current training task is input into the system, and the system accesses the training scoring standard database. According to the tolerance intervals of each dimension indicator set in the scoring rules, the reference standard trajectory sequence that adapts to the task type, action type and difficulty level is matched. The scoring rules are a set of judgment criteria for mapping low-dimensional operational feature values ​​into structured scoring results, which are preset in the system training scoring standard database. The setting of the scoring rules is based on a large amount of expert operation sample data analysis and experience domain knowledge modeling, covering different task types, action categories and training difficulty levels. The scoring rules include: trajectory deviation tolerance threshold Used to evaluate the degree of deviation between the user's trajectory and the standard trajectory in space. Its value is set based on the average deviation range of the standard trajectory samples. If the trajectory deviation is less than the threshold, a higher score is given. If it exceeds the threshold, the score is deducted proportionally, reflecting the operation accuracy. Posture control deviation threshold Used to evaluate the changing trend and fluctuation range of the posture angular velocity (Pitch, Roll, Yaw) to determine whether the user's control is stable. The threshold is determined based on the posture angular velocity characteristics (such as mean, standard deviation, coefficient of variation) and the results of expert data modeling to map the posture fluctuation score. Time rhythm tolerance , which is used to assess the degree of synchronization between user operation events (such as sudden posture changes and trajectory turns) and the system's expected action timing. Its tolerance range is set based on the task rhythm requirements, reflecting the user's sense of rhythm and reaction speed, and is converted into a response delay score. The operation stability threshold is used to assess the smoothness of the user's acceleration changes during the operation, reflecting their operation consistency and force control level. This threshold is set based on the statistical characteristics of the amplitude range, change rate, and standard deviation of each axial acceleration component in the standard operation sample, and is used to limit the degree of fluctuation within a reasonable range. When the acceleration changes in the user's operation fall within the threshold range, the action output is relatively stable and the score is high. If there are frequent violent fluctuations or sudden acceleration changes (i.e., exceeding the threshold range), the operation is considered unstable and a lower score is given according to the set deduction rules. This threshold comprehensively reflects the operation smoothness and control accuracy and is the key basis for calculating the operation stability score. Secondly, the user trajectory segment and the matching reference standard trajectory sequence are temporally aligned using the DTW algorithm to obtain the optimal alignment path W, where each pair of alignment points is composed of the user trajectory points. With reference trajectory point Based on the alignment path, multiple scoring indicators are calculated: ① Trajectory space deviation score , the formula is ; where N is the total number of trajectory point pairs in the alignment path, For the For matching trajectory point pairs, For the The spatial distance between points, is the trajectory deviation tolerance threshold set in the scoring rule, The value range is [0,1], the closer to 1, the closer it is to the reference trajectory; ② Time rhythm matching score , calculated based on the timestamp difference of the alignment points: ;in and The user trajectory segment and the reference standard trajectory are respectively The timestamp of the matching points, is the time rhythm tolerance; ③ posture matching score , extract the difference in pitch angle, yaw angle and roll angle of each pair of alignment points, and calculate the formula: ;in is the total number of trajectory point pairs in the alignment path, Respectively The pitch, yaw and roll angles of each user trajectory segment, For the The pitch, yaw and roll angles of a reference standard trajectory sequence, The system then scores each feature in the operational characteristic parameter set according to a scoring rule, mapping the feature values ​​to capability scores. The scoring rule includes preset numerical thresholds for different operational dimensions, which are used to map the feature parameters into quantitative scores. For the component sequences of the attitude angular velocity along the X, Y, and Z axes, the mean, standard deviation, and coefficient of variation are calculated to construct the attitude fluctuation feature. This feature value is then matched with the attitude control deviation threshold set in the scoring rule to determine the scoring range within which it falls, thereby obtaining the attitude fluctuation score. Specifically, when the attitude fluctuation range exceeds the set attitude control deviation threshold, the score is correspondingly lowered; when the attitude fluctuation is small and remains within the normal fluctuation range, a higher score is given. For the component sequences of the displacement acceleration along each axis, the amplitude range, acceleration change rate, and standard deviation are calculated to construct the operational stability feature. The amplitude range reflects the amplitude range of the acceleration change, the change rate reflects the degree of acceleration change over a short period of time, and the standard deviation characterizes the acceleration fluctuation. These calculation results are compared with the operational stability scoring criteria to obtain the operational stability score. If the acceleration fluctuates significantly and exceeds the operational stability threshold in the pre-set scoring rules, the operational stability score is low. Conversely, if the acceleration is stable and the fluctuation range is small, the angular velocity change rate is analyzed through a sliding window to extract the time points of posture changes that exceed the pre-set mutation threshold. The time difference between this and the command trigger time point in the key press time sequence is calculated to obtain the response delay feature. This time difference is compared with the tempo tolerance threshold in the scoring rules and mapped to a response delay score based on the scoring threshold. A smaller time difference is assigned a higher response delay score. Regarding trajectory coherence, trajectory coherence features are constructed by calculating the trajectory direction change rate, path smoothness, and spatial offset consistency indicators. Combined with the trajectory offset tolerance threshold set in the scoring rules, the trajectory coherence score is completed. A smooth and highly consistent trajectory is scored higher; a drastic trajectory with large deviations is scored lower. Finally, the above scoring results are combined with the spatial, temporal, and posture matching scores calculated during the DTW alignment process and weighted according to pre-set weight parameters to form a user's ability score vector for the training task, which serves as the basis for subsequent ability assessment and path recommendation.

[0026] The user capability vector generation module 4 is used to normalize each scoring dimension and embed a time tag through the capability scoring vector to generate a user capability vector. The user capability vector is the user's comprehensive operation level in the dimensions of operation stability, response speed, control precision and execution accuracy.

[0027] Furthermore, the capability scoring vector is used to normalize each scoring dimension and embed a time tag to generate a user capability vector, including: Performing linear normalization processing on each dimension of the capability score vector to obtain a normalized capability score vector; Based on the original trajectory data and the operation timestamp information contained in the key press time sequence, the normalized ability score vector is matched with each operation time to generate a user ability vector containing a time tag.

[0028] Specifically, the user capability vector generation module is used to construct a user capability vector based on the capability score vector to reflect the user's comprehensive capability performance in dimensions such as operational stability, response speed, control accuracy, and execution accuracy, as well as their changing trends over time. The capability score vector is composed of multiple sub-scores, which respectively reflect the evaluation results of posture fluctuation, operational stability, response delay, trajectory coherence, trajectory space deviation, time rhythm matching, and posture matching. In order to achieve comparative and trend analysis across time periods and users, the system performs linear normalization on each dimension in the capability score vector. The normalization formula is: ,in is the current rating value, and The normalized score values ​​are the historical minimum and maximum values ​​of the score dimension, which can be obtained based on statistics of similar tasks. , ensuring that different dimensions are comparable and the same dimension can be tracked. Subsequently, the system embeds the normalized score vector into a time tag, which can be derived from the original trajectory data and the operation start timestamp in the key time sequence. , realizing user capability vector Each user capability vector represents the user's operational capability status when executing a certain training task.

[0029] The capability development trend map generation module 5 is used to call the user capability vectors at multiple time points in the user historical training record database, construct a capability evolution time series, and generate a capability development trend map based on the change slope and fluctuation trend of each capability dimension.

[0030] Furthermore, the user capability vectors at multiple time points in the user historical training record database are called to construct a capability evolution time series, and based on the change slope and fluctuation trend of each capability dimension, a capability development trend map is generated, including: Extract user capability vectors corresponding to multiple training time points in the user historical training record database; Arranging the user capability vectors in chronological order to construct a capability evolution time series; Based on the preset sliding window length and step size, a sliding window difference calculation is performed on the scores of each dimension in the capability evolution time series to obtain the slope of the score change between consecutive time points; In each sliding window, the standard deviation of the scores of each dimension is calculated as a volatility indicator; Based on the score sequence of each capability dimension, combined with the score change slope and the corresponding volatility index, a weighted linear regression method is used for trend fitting. The weighting coefficient is the inverse of the volatility index within the sliding window. The trend fitting results of each dimension are combined to generate a user capability development trend map, which is a collection of curves showing how the scores of each dimension change over time.

[0031] Specifically, the capability development trend map generation module is used to extract the user capability vectors at multiple time points from the user's historical training record database to construct a capability score time series of the user's operating capability evolution over time, and generate a capability development trend map through trend analysis methods to visually reflect the user's growth trajectory in dimensions such as operating stability, response speed, control precision and execution accuracy. The system first extracts multiple training time points from the user's historical training records. Corresponding user capability vector , among which The vector is represented as Indicates that the user is at time Yu Di Normalized score values ​​on capability scoring dimensions (such as operational stability). Arrange these capability vectors in ascending order of time to construct capability evolution time series. For each capability scoring dimension , extract its time-varying score sequence , and perform sliding window processing on the sequence, setting the window length to , the sliding step length is (For example , ). In each sliding window, calculate the score change slope and volatility index of the current window. The calculation method of score change slope is: for the sliding window starting at A window at a certain point in time , the score change slope is defined as: ,in, Indicates the The growth rate of the score dimension in the current window is used to measure the improvement speed of the user's ability dimension. At the same time, the score volatility index in the window is calculated, that is, the score standard deviation: ,in, is the mean of the scores within the window. Volatility indicator Used to measure the stability of the scoring dimension in the current window. In order to comprehensively reflect the trend and stability of the ability score, the system uses a weighted linear regression method to calculate the score sequence of each dimension. Perform trend fitting, the fitting function form is: ; The regression weight is set to the inverse of the volatility of the current window , enhancing the contribution of stable score intervals to the fitted trend. Ultimately, the system combines the score fitting curves for all dimensions to generate a user capability development trend map. This map is a collection of two-dimensional time-capability coordinate curves, dynamically visualizing the user's development trajectory across multiple capability dimensions. This method considers both the rate of change and volatility of scores, improving the accuracy of trend modeling and the power of analytical interpretation. It helps identify operational dimensions where users are making slow or unstable progress, supporting subsequent training plan optimization and personalized capability intervention.

[0032] The personalized training path construction module 6 is used to identify the user's ability short board dimension through the ability dimension that has a downward or slow growth trend in the ability development trend map, and to filter the training tasks corresponding to the ability short board dimension from the training task database to generate a personalized training task path for the ability short board dimension.

[0033] Furthermore, the method of identifying the user's ability short board dimension by using the ability dimension that has a downward or slow growth trend in the ability development trend map, and screening the training tasks corresponding to the ability short board dimension from the training task database to generate a personalized training task path for the ability short board dimension includes: Extract the score change slope of each capability dimension in the capability development trend map, and mark the scoring dimension with a positive slope less than the preset threshold or a negative value as a capability shortcoming dimension; Screening training tasks that match the capability shortcoming dimension from a training task database to form a candidate task set; Calculating a fitness score for each task in the candidate task set, the fitness score is based on the overlap ratio between the training target label and the user's ability short board dimension label, the coverage degree between the operational characteristic parameter set and the ability short board characteristic parameter in the training task, and the matching between the training rhythm and the user's historical training rhythm; The candidate task set is sorted according to the fitness score, and several tasks are selected to form the personalized training task path.

[0034] Furthermore, the calculation of the fitness score includes: For each training task in the candidate task set, extract its task label set and compare it with the user ability short board dimension set D. Calculate the ratio of the number of intersection items to the total number of items in the user ability short board dimension set to obtain the label overlap ratio. ; Count the intersection ratios between the operational characteristic parameter sets involved in the training operations in the candidate task set and the operational characteristic parameter sets corresponding to the capability short board dimensions to obtain the parameter coverage ratio. , as a feature coverage indicator; Calculate the Euclidean distance between the training time of the candidate task set and the average training time in the user's historical training records, and take the reciprocal as the rhythm matching score ; According to the preset weight 、 、 , for label overlap ratio , parameter coverage ratio and rhythm matching score Perform weighted summation to generate the final fitness score .

[0035] Specifically, based on the ability dimensions that have a downward or slow growth trend in the user's ability development trend map, the user's ability shortcomings are identified, and appropriate tasks are selected from the training task database based on this, to build a targeted personalized training path. First, the module extracts the change slope of each ability dimension score in the user's ability development trend map, and uses the sliding window method to analyze the time series user ability vector. Specifically, set the sliding window length w and step size s, and score each dimension Calculate the mean difference within adjacent windows and divide by the time interval to get the score slope: ;in For the Dimension capability score at the start time of the window If < (in The preset improvement threshold for this dimension, such as 0.01, is set empirically. Then this capability dimension is marked as the user capability short board dimension set D, which is recorded as Next, tasks matching the capability short board dimension D are selected from the training task database to form a candidate task set. Each training task Collection with task tags and the set of operating characteristic parameters involved For each candidate task, calculate the fitness score The fitness score comprehensively considers the overlap ratio between the training target label and the user's ability short board dimension label, the coverage degree between the operation feature parameter set in the training task and the ability short board feature parameter, and the matching between the training rhythm and the user's historical training rhythm. The fitness score consists of three parts: label overlap ratio : Defined as , used to measure the degree of match between task objectives and short board dimensions; parameter coverage ratio : Defined as ,in Is all with D A collection of operational characteristic parameters associated with the middle dimension, used to measure the degree of coverage between task operation content and short board parameters; rhythm matching score :By comparing the task training time The average duration of user historical training The Euclidean distance calculation formula is Used to measure whether the training rhythm matches the user's habits and the final fitness score The above three items are weighted by preset weights Weighted combination 。

[0036] In summary, the portable console intelligent training system provided by this application has the following technical effects: The original trajectory data acquisition module 1 is used to collect the original trajectory data of the user during the training task on the portable console. The portable console has posture perception and key input capabilities. The original trajectory data includes posture angular velocity, displacement acceleration, key time sequence and operation instruction number; Operation feature parameter set extraction module 2, the operation feature parameter set extraction module 2 is used to pre-process the original trajectory data, segment it based on the key press time sequence, obtain multiple trajectory segments corresponding to user operation behaviors, and extract an operation feature parameter set for each trajectory segment. The operation feature parameter set includes posture fluctuation, operation stability, response delay and trajectory coherence; Ability scoring vector construction module 3, which is used to match the operation feature parameter set with the scoring rules, determine the reference standard trajectory sequence, use the dynamic time warping (DTW) method to perform time series alignment, calculate the scoring index, and construct the ability scoring vector based on the operation feature parameter set; A user capability vector generation module 4 is configured to normalize each scoring dimension and embed a time tag using the capability scoring vector to generate a user capability vector. The user capability vector represents the user's comprehensive operational level in terms of operational stability, response speed, control precision, and execution accuracy. A capability development trend map generation module 5 is used to call the user capability vectors at multiple time points in the user historical training record database, construct a capability evolution time series, and generate a capability development trend map based on the change slope and fluctuation trend of each capability dimension; The personalized training path construction module 6 is used to identify the user's ability short board dimension through the ability dimension that has a downward or slow growth trend in the ability development trend map, and to filter the training tasks corresponding to the ability short board dimension from the training task database to generate a personalized training task path for the ability short board dimension.

[0037] In the second embodiment, based on the same inventive concept as the portable console intelligent training system in the above embodiment, the present application also provides a portable console intelligent training method, please refer to the attached Figure 2 ,include: The original trajectory data of the user performing the training task on the portable console is collected. The portable console has the ability of posture perception and key input. The original trajectory data includes posture angular velocity, displacement acceleration, key time sequence and operation instruction number; the original trajectory data is preprocessed and segmented based on the key time sequence to obtain multiple trajectory segments corresponding to the user's operation behavior. The operation feature parameter set is extracted for each trajectory segment. The operation feature parameter set includes posture fluctuation, operation stability, response delay and trajectory coherence; the operation feature parameter set is matched with the scoring rule to determine the reference standard trajectory sequence, and the dynamic time warping (DTW) method is used for time sequence alignment to calculate the scoring index. The ability evaluation system is constructed by combining the operation feature parameter set. The method comprises the following steps: first, normalizing the scoring dimension and embedding the time tag into the capability scoring vector; using the capability scoring vector, normalizing each scoring dimension and embedding the time tag to generate a user capability vector, which is the user's comprehensive operation level in the dimensions of operation stability, response speed, control accuracy and execution accuracy; calling the user capability vectors at multiple time points in the user's historical training record database, constructing a capability evolution time series, and generating a capability development trend map based on the change slope and fluctuation trend of each capability dimension; identifying the user's capability shortboard dimension through the capability dimension with a downward or slow growth trend in the capability development trend map, and screening the training tasks corresponding to the capability shortboard dimension from the training task database to generate a personalized training task path for the capability shortboard dimension.

[0038] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The portable console intelligent training system and specific examples in the aforementioned embodiment one are also applicable to the portable console intelligent training method in this embodiment. Through the aforementioned detailed description of a portable console intelligent training system, those skilled in the art can clearly understand the portable console intelligent training method in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here.

[0039] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

[0040] Obviously, those skilled in the art may make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A portable console intelligent training system, characterized in that: The system comprises: A raw trajectory data acquisition module, which is used to collect raw trajectory data of a user performing a training task on a portable console with posture perception and key input capabilities. The raw trajectory data includes posture angular velocity, displacement acceleration, key time sequence, and operation instruction number; An operation feature parameter set extraction module is used to pre-process the raw trajectory data, segment it based on the key press time sequence, obtain multiple trajectory segments corresponding to user operation behaviors, and extract an operation feature parameter set for each trajectory segment. The operation feature parameter set includes posture fluctuation, operation stability, response delay, and trajectory coherence; A capability scoring vector construction module is used to match the operation feature parameter set with the scoring rules, determine the reference standard trajectory sequence, use the dynamic time warping (DTW) method to perform time series alignment, calculate the scoring index, and construct the capability scoring vector in combination with the operation feature parameter set; A user capability vector generation module, which is used to normalize each scoring dimension and embed a time tag using the capability scoring vector to generate a user capability vector. The user capability vector represents the user's comprehensive operational level in terms of operational stability, response speed, control precision, and execution accuracy. A capability development trend map generation module is used to call the user capability vectors at multiple time points in the user historical training record database, construct a capability evolution time series, and generate a capability development trend map based on the change slope and fluctuation trend of each capability dimension; A personalized training path construction module is used to identify the user's ability short board dimension through the ability dimension that has a downward or slow growth trend in the ability development trend map, and to filter the training tasks corresponding to the ability short board dimension from the training task database to generate a personalized training task path for the ability short board dimension.

2. A portable console intelligent training system as claimed in claim 1, characterized in that: The raw trajectory data is preprocessed, segmented based on the key time sequence, and a plurality of trajectory segments corresponding to user operation behaviors are obtained, and an operation feature parameter set is extracted for each trajectory segment, including: Preprocessing the raw trajectory data includes performing weighted median filtering to denoise the attitude angular velocity and displacement acceleration data, aligning the sampling time using linear interpolation, and normalizing the amplitude of the trajectory spatial components based on the z-score standard deviation normalization method to generate normalized trajectory data; Based on the key pressing time points and the operation instruction number change time points in the key time series, the normalized trajectory data is segmented into time series to extract multiple trajectory segments corresponding to the user operation; For each of the trajectory segments, a temporal correspondence between trajectory changes and instruction responses is determined based on the key press time sequence within the corresponding time period, and an operation feature parameter set is extracted.

3. A portable console intelligent training system as claimed in claim 2, characterized in that: For each of the trajectory segments, determining a temporal correspondence between trajectory changes and command responses based on the key press time sequence within a corresponding time period, and extracting an operation feature parameter set, includes: Based on the component sequence of the attitude angular velocity in the X, Y, and Z axes in the trajectory segment, the mean, standard deviation, and coefficient of variation of each dimension are calculated. The mean is used to reflect the average attitude direction, and the standard deviation and coefficient of variation are used to characterize the degree of fluctuation of attitude control, thereby obtaining attitude fluctuation characteristics; Based on the components of the displacement acceleration in the X, Y, and Z axes in the trajectory segment, the amplitude range is calculated, and the degree of acceleration fluctuation during the operation is evaluated by combining the standard deviation of each axis with the acceleration change rate at adjacent time points to obtain the operational stability characteristics; Based on the posture angular velocity component in the trajectory segment, a sliding window method is used to analyze the angular velocity change rate of adjacent moments. When the change rate exceeds a preset mutation threshold, the corresponding moment is determined to be a posture mutation time point. The posture mutation time point is matched with the key triggering time point in the key time sequence, and the time difference between the two is calculated to obtain the response delay feature; Based on the accumulation of spatial offset distances between consecutive trajectory points in the trajectory segment, the consistency of trajectory point direction and amplitude changes in adjacent time periods is compared, and the trajectory turning angle change rate and path smoothness index are calculated to obtain trajectory continuity features.

4. A portable console intelligent training system as claimed in claim 1, characterized in that: The process of matching the operational characteristic parameter set with the scoring rules, determining the reference standard trajectory sequence, using the dynamic time warping (DTW) method for time series alignment, calculating the scoring index, and constructing the capability scoring vector in combination with the operational characteristic parameter set includes: Match the user's operational feature parameter set extracted in the current training task with the scoring rules in the training scoring standard database. Determine the reference standard trajectory sequence by comparing the operational feature index with the threshold set by the scoring rule item to see if it falls within the scoring level range. The scoring rules include trajectory deviation tolerance threshold, posture control deviation threshold, time rhythm tolerance, and operational stability threshold. Dynamic time warping (DTW) is used to align the user trajectory segments with the reference standard trajectory sequence to obtain an alignment path. Based on the alignment path, scoring indicators are calculated, including trajectory space deviation score, time rhythm matching score, and posture matching score, and matching degree correction is performed according to the scoring rules; The operation characteristic parameter set is compared with the scoring rules, and the posture fluctuation score, operation stability score, response delay score and trajectory coherence score are determined according to the relationship between the characteristic value of each dimension and the set threshold; The trajectory spatial deviation score, time rhythm matching score, posture matching score, posture fluctuation score, operation stability score, response delay score and trajectory coherence score are weighted and combined according to preset weight parameters to generate an ability score vector.

5. A portable console intelligent training system as claimed in claim 4, characterized in that: Based on the alignment path, scoring indicators are calculated, including trajectory space deviation score, time rhythm matching score, and posture matching score, and matching degree correction is performed according to the scoring rules, including: Based on each pair of matching trajectory points in the alignment path , respectively obtain the first The location of the sampling points The corresponding reference standard trajectory sequence Alignment points , and calculate the trajectory space deviation score , the formula is: ； in is the total number of trajectory point pairs in the alignment path, For the For matching trajectory point pairs, For the k The spatial distance between points, The trajectory deviation tolerance threshold set in the scoring rule; Extract the timestamp difference of each pair of matching points from the alignment path and calculate the tempo matching score , and its calculation formula is: ； in, is the total number of trajectory point pairs in the alignment path, and The user trajectory segment and the reference standard trajectory sequence are respectively The timestamp of the matching points, is the time rhythm tolerance; Based on the alignment path, the attitude angle parameter difference corresponding to each pair of trajectory points is extracted, and the attitude angle parameter difference includes the pitch angle between the user trajectory segment and the reference standard trajectory sequence. , yaw angle and roll angle differences in Calculate pose matching score , and its calculation formula is: ； in, is the total number of trajectory point pairs in the alignment path, Respectively The pitch, yaw and roll angles of each user trajectory segment, For the The pitch, yaw and roll angles of a reference standard trajectory sequence, is the preset maximum attitude deviation tolerance value; The trajectory spatial deviation score , time rhythm matching score Matching score with posture Together they constitute the dimension elements of the capability score vector.

6. A portable console intelligent training system as claimed in claim 1, characterized in that: The process of normalizing each scoring dimension and embedding a time tag into the capability scoring vector to generate a user capability vector includes: Performing linear normalization processing on each dimension of the capability score vector to obtain a normalized capability score vector; Based on the original trajectory data and the operation timestamp information contained in the key press time sequence, the normalized ability score vector is matched with each operation time to generate a user ability vector containing a time tag.

7. A portable console intelligent training system as claimed in claim 1, characterized in that: The method calls the user capability vectors at multiple time points in the user historical training record database, constructs a capability evolution time series, and generates a capability development trend map based on the change slope and fluctuation trend of each capability dimension, including: Extract user capability vectors corresponding to multiple training time points in the user historical training record database; Arranging the user capability vectors in chronological order to construct a capability evolution time series; Based on the preset sliding window length and step size, a sliding window difference calculation is performed on the scores of each dimension in the capability evolution time series to obtain the slope of the score change between consecutive time points; In each sliding window, the standard deviation of the scores of each dimension is calculated as a volatility indicator; Based on the score sequence of each capability dimension, combined with the score change slope and the corresponding volatility index, a weighted linear regression method is used for trend fitting. The weighting coefficient is the inverse of the volatility index within the sliding window. The trend fitting results of each dimension are combined to generate a user capability development trend map, which is a collection of curves showing how the scores of each dimension change over time.

8. The portable console intelligent training system according to claim 1, characterized in that: The method of identifying the user's capability short board dimension by using the capability dimension that has a downward or slow growth trend in the capability development trend map, screening the training tasks corresponding to the capability short board dimension from the training task database, and generating a personalized training task path for the capability short board dimension includes: Extract the score change slope of each capability dimension in the capability development trend map, and mark the scoring dimension with a positive slope less than the preset threshold or a negative value as a capability shortcoming dimension; Screening training tasks that match the capability shortcoming dimension from a training task database to form a candidate task set; Calculating a fitness score for each task in the candidate task set, the fitness score is based on the overlap ratio between the training target label and the user's ability short board dimension label, the coverage degree between the operational characteristic parameter set and the ability short board characteristic parameter in the training task, and the matching between the training rhythm and the user's historical training rhythm; The candidate task set is sorted according to the fitness score, and several tasks are selected to form the personalized training task path.

9. A portable console intelligent training system as claimed in claim 8, characterized in that: The calculation of the fitness score includes: For each training task in the candidate task set, extract its task label set and compare it with the user ability short board dimension set D. Calculate the ratio of the number of intersection items to the total number of items in the user ability short board dimension set to obtain the label overlap ratio. ; Count the intersection ratios between the operational characteristic parameter sets involved in the training operations in the candidate task set and the operational characteristic parameter sets corresponding to the capability short board dimensions to obtain the parameter coverage ratio. , as a feature coverage indicator; Calculate the Euclidean distance between the training time of the candidate task set and the average training time in the user's historical training records, and take the reciprocal as the rhythm matching score ; According to the preset weight 、 、 , for label overlap ratio , parameter coverage ratio and rhythm matching score Perform weighted summation to generate the final fitness score .

10. A portable console intelligent training method, characterized in that: The method is performed by a portable console intelligent training system according to any one of claims 1 to 9, comprising: Collecting raw trajectory data of a user performing a training task on a portable console with gesture sensing and key input capabilities. The raw trajectory data includes gesture angular velocity, displacement acceleration, key time sequence, and operation instruction number. Preprocessing the raw trajectory data, segmenting it based on the key press time sequence to obtain multiple trajectory segments corresponding to user operation behaviors, and extracting an operation feature parameter set for each trajectory segment, wherein the operation feature parameter set includes posture fluctuation, operation stability, response delay, and trajectory coherence; Match the operational characteristic parameter set with the scoring rules, determine the reference standard trajectory sequence, use the dynamic time warping (DTW) method to align the time series, calculate the scoring index, and construct the capability scoring vector based on the operational characteristic parameter set; By using the capability scoring vector, each scoring dimension is normalized and a time tag is embedded to generate a user capability vector. The user capability vector represents the user's comprehensive operational level in terms of operational stability, response speed, control precision, and execution accuracy. Calling the user's ability vectors at multiple time points in the user's historical training record database, constructing an ability evolution time series, and generating an ability development trend map based on the change slope and fluctuation trend of each ability dimension; Through the ability dimensions that have a downward or slow growth trend in the ability development trend map, the user's ability shortcoming dimensions are identified, and training tasks corresponding to the ability shortcoming dimensions are screened from the training task database to generate personalized training task paths for the ability shortcoming dimensions.

Images