Intelligent dumbbell training plan dynamic generation and action quality scoring method and system

Abstract

The invention relates to the technical field of intelligent fitness equipment and exercise training control, and provides a training plan dynamic generation and action quality scoring method and system for an intelligent dumbbell. The method comprises the following steps: acquiring action data of a user during training and personalized parameters input by the user in real time through an acceleration sensor, an angular velocity sensor and a pressure sensor which are arranged in the intelligent dumbbell; inputting the action data and the personalized parameters into a preset training plan generation algorithm module so as to generate a stage training plan comprising the action type, each group of training duration, intermittent time and a weight adjustment strategy; an action quality scoring model is constructed, a standard action template is pre-stored in the scoring model, and the standard action template comprises standard trajectory curves, speed threshold ranges, amplitude reference values and force exerting symmetry parameters corresponding to different training actions; and performing multi-dimensional matching analysis on the action data and a standard action template, and dynamically correcting the score in combination with the current fatigue parameter of the user.

Description

Technical Field

[0001] The present application relates to the field of intelligent fitness equipment and sports training control technology, and in particular to a method and system for dynamically generating a training plan and scoring movement quality for an intelligent dumbbell. Background Art

[0002] With the increasing prevalence of smart fitness devices, users are increasingly demanding personalized, scientific training. Traditional dumbbell training plans are typically user-defined or rely on fixed templates, which present two major technical bottlenecks. First, training plans lack dynamic adaptability. Existing solutions often generate fixed plans based on static user fitness parameters (such as weight and age). These plans are unable to accurately perceive changes in fitness, force stability, and movement standardization during training. This results in a mismatch between training intensity and actual ability, potentially leading to poor training results and the risk of sports injuries. Second, movement quality assessment methods are limited. Existing technologies often use a single sensor (such as an accelerometer) to capture movement trajectories, providing only a simple assessment of movement completion. These methods lack multi-dimensional analysis of core standard indicators such as movement trajectory consistency, speed uniformity, and force symmetry. Furthermore, they don't dynamically adjust scores based on the user's real-time fatigue status, making it difficult to accurately assess movement quality and adjust training plans accordingly. Furthermore, existing smart dumbbell systems generally lack a closed-loop control mechanism for "training plan generation - movement quality feedback - dynamic plan adjustment." This lacks effective linkage between training plans and actual training results, making adaptive optimization of training plans impossible.

[0003] To address these issues, existing technologies offer several improvements, such as using sensors to collect motion data for motion recognition or generating basic training plans based on user fitness parameters. However, none of these approaches deeply integrate real-time motion data, personalized parameters, historical training fitness, and fatigue assessment. Furthermore, none of these approaches have established a comprehensive technical system integrating dynamic training plan generation, multi-dimensional motion quality scoring, and closed-loop feedback regulation. Therefore, providing an intelligent method that can dynamically adjust training plans based on a user's real-time status and accurately assess motion quality using multi-dimensional data has become a pressing technical challenge in this field. Summary of the Invention

[0004] This application provides a method and system for dynamically generating training plans and scoring movement quality for smart dumbbells, aiming to solve some improvement solutions in the existing technology, such as collecting motion data through sensors for motion recognition, or generating basic training plans based on user physical parameters, but none of them deeply integrates real-time motion data, personalized parameters, historical training fitness and fatigue status assessment, and has not built a complete technical system integrating "dynamic training plan generation, multi-dimensional movement quality scoring, and closed-loop feedback adjustment".

[0005] In a first aspect, an embodiment of the present application provides a method for dynamically generating a training plan and scoring movement quality for a smart dumbbell, which is applied to the smart dumbbell; the method comprises:

[0006] The smart dumbbells use built-in accelerometers, angular velocity sensors, and pressure sensors to collect real-time motion data from users during training. The motion data includes dumbbell motion trajectory, speed, amplitude, grip pressure, and force angle. The smart dumbbells also capture personalized parameters entered by the user, including age, weight, body fat percentage, muscle strength level, training goals, and historical training records.

[0007] Inputting movement data and personalized parameters into a preset training plan generation algorithm module, the module dynamically generates a phased training plan based on the user's movement data, personalized parameters, historical training fitness, and training goals, including movement type, training duration per set, rest time, and weight adjustment strategy. The module also adaptively adjusts the weight adjustment strategy based on the user's real-time force stability and historical strength growth curve.

[0008] A movement quality scoring model is constructed, which pre-stores standard movement templates. The standard movement templates contain standard trajectory curves, speed threshold ranges, amplitude reference values, and force symmetry parameters corresponding to different training movements. The movement data and the standard movement templates are matched and analyzed in multiple dimensions. The movement standard score is calculated from four dimensions: movement trajectory consistency, speed uniformity, amplitude compliance rate, and left-right force symmetry. The user's current fatigue parameters are obtained based on the continuous training duration, historical heart rate data, and grip pressure fluctuation amplitude. The score is dynamically corrected based on the user's current fatigue parameters.

[0009] In some embodiments, after the score is dynamically corrected in combination with the user's current fatigue parameter, it also includes: during the training process, the action quality score result is fed back to the training plan generation algorithm module in real time. If the action standard score is lower than the preset threshold for multiple consecutive times, the training plan dynamic adjustment mechanism is triggered to automatically reduce the training weight or shorten the single set training time until the action standard score returns to a reasonable range, forming a closed-loop optimization control of the training plan and action quality.

[0010] In some embodiments, the real-time collection of user motion data during training by the built-in acceleration sensor, angular velocity sensor and pressure sensor of the smart dumbbell includes: using the acceleration sensor to collect the acceleration data of the dumbbell in three-dimensional space in real time, using the angular velocity sensor to collect the angular velocity data of the dumbbell around three rotation axes in real time, and using the pressure sensor to collect the grip pressure data of the user when holding the dumbbell in real time; based on the acceleration data and angular velocity data, the motion trajectory, motion speed and motion amplitude of the dumbbell are calculated, and based on the acceleration data, angular velocity data and grip pressure data, the force angle when the user exerts force is calculated.

[0011] In some embodiments, the method dynamically generates a phased training plan based on user motion data, personalized parameters, historical training fitness and training goals, including action type, duration of each training group, interval time and weight adjustment strategy, including: the training plan generation algorithm module first analyzes the action completion, muscle recovery cycle and training effect data in the user's historical training records to determine the historical training fitness, and matches the appropriate action type from the preset action library in combination with the real-time physical state reflected by the user's current motion data, the training goals and muscle strength level in the personalized parameters; sets the duration of each training group according to the fatigue tolerance time of similar actions in the user's historical training, and sets the interval time between groups according to the recovery law of sports physiology; uses the weight adjustment strategy as a variable parameter of the training plan, so that the generated phased training plan includes a weight change strategy that is dynamically adjusted with the training stage.

[0012] In some embodiments, the weight adjustment strategy is adaptively adjusted according to the user's real-time force stability and historical strength growth curve, including: judging the user's real-time force stability by analyzing the fluctuation amplitude of the grip pressure data and force angle data collected in real time, if the fluctuation amplitude is within the preset stability threshold, the force is determined to be stable, otherwise it is determined to be unstable; calling the weight increase data of the same action type in the user's historical training records to form a historical strength growth curve, when the force is stable and the historical strength growth curve shows an upward trend, gradually increasing the training weight according to the preset incremental rules, when the force is unstable or the historical strength growth curve shows a plateau period, maintaining the current training weight or reducing the training weight according to the preset reduction rules.

[0013] In some embodiments, the construction of the movement quality scoring model includes: collecting standard movement data of professional trainees when completing different training movements through sports biomechanics experiments, the standard movement data including standard movement trajectory, standard movement speed range, standard movement amplitude and left and right force symmetry reference value; classifying the standard movement data according to the movement type, generating a standard movement template containing the corresponding standard trajectory curve, speed threshold range, amplitude reference value and force symmetry parameters, and storing the standard movement template in the database of the movement quality scoring model.

[0014] In some embodiments, the motion data is matched with the standard motion template in multiple dimensions for analysis, and the motion standard score is calculated from four dimensions: motion trajectory consistency, speed uniformity, amplitude compliance rate, and left-right force symmetry, including: calculating the coordinate point fitting degree of the dumbbell motion trajectory collected in real time and the standard trajectory curve in the standard motion template to obtain the motion trajectory consistency score; comparing the real-time motion speed with the speed threshold range in the standard motion template, and calculating the proportion of time the speed value is within the threshold range as the speed uniformity score; comparing the real-time motion amplitude with the amplitude reference value in the standard motion template, and calculating the proportion of the number of movements that reach or exceed the amplitude reference value as the amplitude compliance rate score; obtaining the grip pressure data and force angle data of the left and right smart dumbbells, and calculating the degree of deviation of the difference value of the data on both sides from the symmetry reference value as the left-right force symmetry score; and calculating the total motion standard score by combining the scores of the four dimensions.

[0015] In some embodiments, the method of obtaining the user's current fatigue parameters based on continuous training time, historical heart rate data, and grip pressure fluctuation amplitude includes: counting the uninterrupted continuous training time in the user's current training stage, and marking the fatigue risk when the continuous training time exceeds the preset fatigue warning time; retrieving the heart rate data in the user's historical training records, analyzing the deviation between the current training heart rate and the historical average heart rate to determine the degree of cardiopulmonary fatigue; calculating the fluctuation amplitude of real-time grip pressure data, and determining that muscle control ability has decreased when the fluctuation amplitude exceeds a preset stability threshold; comprehensively considering the continuous training time, heart rate deviation amplitude, and grip pressure fluctuation amplitude, and determining the user's current fatigue parameters through preset fatigue assessment rules.

[0016] In some embodiments, the score is dynamically corrected in combination with the user's current fatigue parameter, including: setting a correction coefficient for the action standard score according to the user's current fatigue parameter, and adaptively relaxing the score thresholds for action trajectory consistency and speed uniformity when the fatigue parameter shows that the user is in a mild fatigue state; when the fatigue parameter shows that the user is in a moderate fatigue state or above, in addition to relaxing the score threshold, an additional judgment on the safety of the action trajectory is added, and if there is a force deviation that may cause sports injury, the action standard score is directly reduced; through the correspondence between the fatigue parameter and the correction rule, dynamic adjustment of the action quality score is achieved.

[0017] In a second aspect, the present application provides a system for dynamically generating a training plan and scoring movement quality for a smart dumbbell, which is applied to the smart dumbbell. The system includes:

[0018] A parameter acquisition unit is used to collect real-time motion data of the user during training through the built-in acceleration sensor, angular velocity sensor, and pressure sensor of the smart dumbbell. The motion data includes dumbbell motion trajectory, motion speed, motion amplitude, grip pressure, and force angle; and obtain personalized parameters input by the user, including age, weight, body fat percentage, muscle strength level, training goals, and historical training records;

[0019] A strategy acquisition unit, configured to input motion data and personalized parameters into a preset training plan generation algorithm module, dynamically generating a phased training plan including movement type, duration of each training set, rest time, and weight adjustment strategy based on the user's motion data, personalized parameters, historical training fitness, and training goals, and adaptively adjusting the weight adjustment strategy based on the user's real-time force stability and historical strength growth curve;

[0020] A dynamic correction unit is used to construct a movement quality scoring model. The scoring model pre-stores standard movement templates, which contain standard trajectory curves, speed threshold ranges, amplitude reference values, and force symmetry parameters corresponding to different training movements. The movement data is matched with the standard movement templates in multiple dimensions for analysis, and the movement standard score is calculated from four dimensions: movement trajectory consistency, speed uniformity, amplitude compliance rate, and left and right force symmetry. The user's current fatigue parameters are obtained based on the continuous training time, historical heart rate data, and grip pressure fluctuation amplitude, and the score is dynamically corrected in combination with the user's current fatigue parameters.

[0021] The embodiment of the present application provides a method and system for dynamically generating a training plan and scoring movement quality for a smart dumbbell. It obtains muscle response signals under static compression through multiple sensors such as pressure, vibration, and displacement, extracts multi-dimensional characteristic parameters such as pressure peak, deformation recovery time, and vibration attenuation coefficient, and accurately quantifies muscle hardness and elastic properties, breaking through the limitations of traditional single signal detection. An intelligent matching model based on machine learning is constructed, which integrates muscle hardness characteristics, user body data, and striking preferences, outputs personalized parameter combinations, and dynamically adjusts parameters through real-time feedback signals during the striking process, forming a closed-loop control of "detection-modeling-execution-feedback", and realizing for the first time real-time intelligent adaptation of smart dumbbell parameters to muscle status. Through pre-strike verification, abnormal state monitoring, and parameter protection mechanisms, striking safety is ensured while personalized adjustment is made, solving the defects of "one-size-fits-all" parameter settings or reliance on subjective judgment in the prior art.

[0022] This application generates striking parameters that are in line with individual differences based on multi-dimensional muscle hardness characteristics and user body data, avoiding the blindness of traditional manual adjustment and improving muscle relaxation effects. Parameters are optimized in real time through feedback signals during the striking process, so that the striking force, frequency, etc. always match the real-time state of the muscles (such as automatically increasing the amplitude when the muscles are stiff after exercise, and intelligently reducing the force when fatigued muscles are over-pressed), significantly improving comfort and safety. Users do not need to manually adjust parameters frequently, and the device automatically completes detection, modeling, and dynamic control, lowering the threshold for use and adapting to the differentiated needs of different sports scenarios (such as pre-exercise activation, post-exercise recovery, and daily relaxation). Through pre-strike verification, abnormal vibration monitoring, and parameter protection mechanisms, the risk of muscle injury caused by improper parameters can be effectively avoided, broadening the range of people applicable to the device (such as the elderly and sports novices).

[0023] In summary, the present invention fills the technical gap in the field of integrated control of "precise detection-intelligent matching-dynamic feedback" of existing smart dumbbells, and provides a new idea for the research and development of intelligent muscle relaxation equipment.

[0024] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0026] Figure 1 This is a schematic flow chart of the steps of a method for dynamically generating a training plan and scoring movement quality for a smart dumbbell provided in one embodiment of the present application;

[0027] Figure 2 This is a schematic diagram of the structure of a smart dumbbell provided in one embodiment of the present application;

[0028] Figure 3 This is a schematic block diagram of the structure of a system for dynamically generating a training plan and scoring movement quality for smart dumbbells provided in one embodiment of the present application;

[0029] Figure 4 This is a schematic block diagram of the structure of the smart dumbbell provided in one embodiment of the present application.

[0030] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0032] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, combined, or partially merged, so the actual execution order may vary depending on the actual situation.

[0033] It should be understood that, in order to clearly describe the technical solutions of the embodiments of the present invention, in the embodiments of the present invention, terms such as "first" and "second" are used to distinguish between identical or similar items having substantially the same functions and effects. Those skilled in the art will understand that terms such as "first" and "second" do not limit the quantity or order of execution, and that terms such as "first" and "second" do not necessarily define differences.

[0034] It should be understood that the terms used in this specification are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in this specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0035] It will also be understood that the term "and / or" as used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0036] Smart dumbbells, a device that uses high-frequency vibrations to stimulate muscles to relieve fatigue and promote blood circulation, have been widely used in sports rehabilitation, fitness, and relaxation scenarios. Existing smart dumbbells' striking parameters (such as frequency, force, and amplitude) are typically manually adjusted by the user or run based on fixed preset modes (such as "relaxation mode" and "deep striking mode"). However, this type of adjustment method has significant drawbacks:

[0037] Lack of personalized adaptation: Muscle hardness and basic body data (such as age, weight, and muscle mass) vary significantly between users, and the same user's muscle state changes dynamically in different scenarios (such as before and after exercise, and in static and fatigued states). Existing smart dumbbells cannot accurately sense real-time muscle hardness and rely solely on the user's subjective experience to adjust parameters, which can easily lead to poor striking effects (such as insufficient force to achieve deep relaxation, and excessive frequency causing muscle damage).

[0038] Single detection method: Traditional solutions only obtain limited data through pressure sensors or simple contact detection, and do not integrate multi-dimensional signals such as pressure, vibration, and displacement to fully characterize muscle hardness, elasticity, and viscosity characteristics, resulting in insufficient detection accuracy.

[0039] Lack of dynamic feedback mechanism: Existing smart dumbbells are unable to collect real-time feedback signals of muscle state changes (such as electromyographic signals and vibration attenuation characteristics) during the striking process, making it difficult to dynamically adjust parameters based on real-time muscle response. This results in a disconnect between parameters and muscle state during the striking process, making it difficult to strike a balance between safety and comfort.

[0040] While a few existing solutions involve sensor-assisted adjustment, these solutions often rely solely on processing a single signal (such as pressure) and fail to form a complete closed loop of "multi-dimensional detection, intelligent modeling, and dynamic matching." For example, some devices use only pressure sensors to control the upper limit of striking force, failing to incorporate vibration, displacement, and other signals to quantify muscle hardness. Others rely on preset empirical formulas to match parameters, failing to build personalized models through machine learning and failing to adapt to the complex differences in human muscle mass.

[0041] To resolve the above, please refer to Figure 1 The embodiment of the present application provides a method for dynamically generating a training plan and scoring the quality of an action for a smart dumbbell, which is applied to Figure 2 At the same time, it should be noted that each piece of information involved in the method provided in this application is extracted with the authorization of the relevant user and in compliance with relevant regulations, and will not infringe on the user's privacy.

[0042] The provided method for dynamically generating a training plan and scoring movement quality for a smart dumbbell includes steps S101 to S103. The details are as follows:

[0043] Step S101. Use the built-in acceleration sensor, angular velocity sensor, and pressure sensor of the smart dumbbell to collect the user's motion data during training in real time. The motion data includes the dumbbell motion trajectory, motion speed, motion amplitude, grip pressure, and force angle; obtain the personalized parameters input by the user, which include age, weight, body fat percentage, muscle strength level, training goals, and historical training records.

[0044] Specifically, this step uses the multiple types of sensors built into the smart dumbbells to collect user training movement data in real time and obtain user personalized basic information, providing a data basis for subsequent training plan generation and movement evaluation.

[0045] Motion data collection uses accelerometers and angular velocity sensors (such as MEMS inertial measurement units) to monitor the dumbbell's three-dimensional motion trajectory (X / Y / Z axis displacement), motion speed (tangential velocity, angular velocity), and force angle (change in joint motion angle) in real time. Pressure sensors (such as piezoresistive or capacitive sensors) collect grip pressure (contact force between the palm and the dumbbell) and force symmetry (pressure difference between the left and right grips).

[0046] Personalized parameters are obtained through static data actively entered by the user: age, weight, body fat percentage, muscle strength level (such as through grip strength tests or historical training data annotation), and training goals (such as muscle gain, fat loss, and rehabilitation training). Historical training records include past training plans, exercise completion, fatigue recovery data, injury history, etc., and are stored locally on the device or in a cloud account.

[0047] Sensor data is collected via acceleration / angular velocity sensors at a sampling rate exceeding 100Hz. A Kalman filter algorithm fuses the data to eliminate motion noise and generate a smooth motion trajectory. Pressure sensors are integrated into the dumbbell grip surface, using an array layout (e.g., three pressure nodes on each side) to calculate the distribution and dynamic changes of grip pressure in real time.

[0048] Users can input basic data through the dumbbell's accompanying app or the device's touch screen. Historical training records are automatically synchronized from the user's account, supporting manual editing or access to third-party health data platforms (such as Apple Health).

[0049] Data pre-processing generates an action sequence containing timestamps (such as a time series array of [trajectory coordinates, speed, pressure, angle]) by normalizing and time-aligning the raw sensor data.

[0050] This technology breaks through the limitations of traditional single-pressure detection and fuses acceleration, angular velocity, and pressure sensors to comprehensively capture movement details (such as trajectory deviation and uneven force application), providing a data foundation for precise analysis. By combining the user's static physical data (age, weight), dynamic training goals (rehabilitation / muscle gain), and historical training habits, it avoids "one-size-fits-all" parameter settings and addresses the lack of individual adaptation in existing technologies. Real-time acquisition of grip pressure and force angle indirectly reflects real-time muscle tension (for example, a sudden increase in pressure may indicate muscle fatigue), providing real-time signals for subsequent fatigue assessment and plan adjustments.

[0051] Step S102. Input the motion data and personalized parameters into the preset training plan generation algorithm module to dynamically generate a phased training plan including the motion type, duration of each training set, interval time and weight adjustment strategy based on the user's motion data, personalized parameters, historical training fitness and training goals, and adaptively adjust the weight adjustment strategy according to the user's real-time force stability and historical strength growth curve.

[0052] Specifically, this step generates an algorithm module through a training plan, integrates the user's real-time motion data, personalized parameters, historical training results and training goals, generates a dynamically adjustable training plan, and optimizes the weight adjustment strategy based on force stability and strength growth trends.

[0053] Core Algorithm Module: Input parameters: Movement data (trajectory, speed, pressure), personalized parameters (age, muscle strength level), historical training fitness (past plan completion rate, muscle recovery speed), training goals (e.g., "deep muscle stimulation" corresponds to a high-frequency percussion strategy). Output: Phased plan including movement type (e.g., bicep curl, shoulder press), training duration per set (e.g., 30 seconds / set), rest interval (e.g., 60 seconds), and weight adjustment strategy (dynamically adjusting dumbbell resistance or vibration frequency based on real-time strength).

[0054] The adaptive adjustment mechanism determines the user's current strength state based on real-time force stability (such as trajectory fluctuation amplitude and pressure uniformity). If force stability decreases by 20% over three consecutive sets, the weight or frequency of the next set will be automatically reduced. Combined with historical strength growth curves (such as the rate of strength improvement over the past four weeks), a machine learning model predicts the optimal load at the current time, preventing overtraining or undertraining.

[0055] The training plan generation algorithm uses a hierarchical planning strategy: first, a basic framework is determined based on training objectives (e.g., prioritizing low loads and long rest intervals for rehabilitation training). Then, reinforcement learning algorithms (e.g., Q-learning) are used to optimize movement types and durations based on historical training fitness. Weight adjustment strategy: The dumbbells have built-in adjustable resistance modules (e.g., electromagnetic damping or spring structures). The algorithm dynamically calculates the resistance coefficient based on real-time pressure data and historical strength levels. For example, the safe load limit for the current strength level is 15 kg, initially set to 80%, or 12 kg, and then gradually adjusted based on force stability.

[0056] The historical training fitness calculation defines the fitness indicators: completion rate (number of planned action sets completed / preset number of sets), recovery index (heart rate recovery speed 24 hours after training), injury risk factor (based on historical action trajectory deviation data), and generates a comprehensive fitness score (0-100 points) through weighted average.

[0057] Real-time feedback adjustment: After each set of training, the algorithm module automatically analyzes the current action data. If it detects an abnormal force angle (such as exceeding the standard value by ±15°) or an excessive pressure fluctuation, it triggers the weight adjustment logic (such as reducing the resistance by 5%) and records it as historical training data.

[0058] Breaking the limitations of traditional fixed presets, this system integrates real-time motion data and historical training habits to generate personalized training plans, resolving the existing reliance on subjective experience for parameter adjustment. Dynamic weight adjustment based on force stability and historical strength growth prevents under-force (poor results) or overload (muscle damage), improving training safety and efficiency. Through a closed loop of "data collection-plan generation-real-time adjustment," training plans are dynamically optimized as the user's muscle state changes (such as post-exercise fatigue), completing a complete "detection-modeling-matching" process and addressing the lack of dynamic feedback in existing technologies.

[0059] Step S103. Construct a movement quality scoring model, which pre-stores standard movement templates, and the standard movement templates contain standard trajectory curves, speed threshold ranges, amplitude reference values, and force symmetry parameters corresponding to different training movements; perform multi-dimensional matching analysis on the movement data and the standard movement templates, calculate the movement standard score from four dimensions: movement trajectory matching, speed uniformity, amplitude compliance rate, and left and right force symmetry; obtain the user's current fatigue parameters based on the continuous training time, historical heart rate data, and grip pressure fluctuation amplitude, and dynamically correct the score based on the user's current fatigue parameters.

[0060] Specifically, this step constructs an action quality scoring model, which realizes action normative evaluation and fatigue status perception by matching standard action templates in multiple dimensions and combining the user's real-time fatigue correction score.

[0061] The standard movement template presets standard trajectory curves for different training movements (such as curls and presses) (generated based on data from sports physiology experts or a large number of sample training), speed threshold ranges (such as concentric contraction speed 2-4° / s), amplitude reference values ​​(lower limit of joint range of motion, such as 90° elbow bending), and force symmetry parameters (the difference in grip pressure between the left and right hands is ≤10%).

[0062] Scoring dimensions and calculations include: Movement trajectory consistency: Calculate the average Euclidean distance between the real-time trajectory and the standard curve, and then normalize the score (0-100 points). Speed ​​uniformity: Count the percentage of time that the movement speed deviates from the threshold range, and reversely map it to a score (e.g., if the time exceeds the limit ≤ 10%, the score is 90 points or above). Range compliance rate: Calculate the percentage of times in each set of movements that the joint range of motion reaches the baseline value (e.g., 8 curls out of 10 meet the standard for 80 points). Left and right force symmetry: Calculate the percentage of left and right pressure difference using pressure sensor data, and deduct points if it exceeds the threshold (e.g., 15%).

[0063] Fatigue parameter corrections include: Fatigue index: Combining continuous training duration (exceeding 45 minutes triggers a fatigue warning), historical heart rate data (obtained via an external heart rate monitor or the device's built-in PPG sensor; a heart rate variability ≥ 20% indicates fatigue), and grip pressure fluctuation amplitude (a pressure standard deviation exceeding 30% of the baseline value is considered unstable), a fatigue coefficient (0-1) is generated using a fuzzy logic algorithm, and the score is dynamically corrected (for example, allowing a 5% relaxation of the trajectory fit tolerance when fatigue occurs).

[0064] Standard templates are constructed by having sports rehabilitation experts demonstrate standard movements, collecting high-precision trajectory data (error ≤ 1mm), and combining them with sports biomechanics models (such as the Newton-Euler equation) to calculate speed and amplitude benchmark values ​​to form a multi-movement template library.

[0065] The real-time scoring algorithm uses the dynamic time warping (DTW) algorithm to match real-time action sequences with standard templates to solve the alignment problem of different users' action rhythm differences; and calculates the scores of each dimension in real time through a sliding window (such as a 1-second window).

[0066] Fatigue fusion correction is achieved by establishing a fatigue assessment model: inputting continuous training duration (T), heart rate variability (HRV), and pressure fluctuation coefficient (σ), and outputting a correction factor (e.g., when fatigue = 0.8, the baseline score value is reduced by 10%), avoiding excessive demands on movement standardization in a fatigued state and balancing safety and training effects.

[0067] Through multi-dimensional quantitative scoring (trajectory, speed, amplitude, and symmetry), this approach addresses the subjective judgment inherent in traditional methods of assessing movement standards, helping users correct incorrect posture (e.g., muscle compensation caused by trajectory deviation) in real time. Dynamic scoring is achieved by combining physiological signals (heart rate) and movement signals (pressure fluctuations), mitigating the risk of injury caused by forcing movement standards during fatigue and improving training comfort and safety. This approach transcends the limitations of single-mode pressure detection by integrating multimodal data such as trajectory, speed, and pressure to comprehensively characterize muscle movement (e.g., vibration attenuation characteristics can be indirectly reflected through trajectory fluctuations), thus addressing the limitations of existing technologies with their single detection method.

[0068] Through the closed-loop design of S101-S103, this method realizes the complete process of "multi-dimensional data collection → intelligent model planning → dynamic feedback adjustment", and specifically solves the problems of existing smart dumbbells such as lack of personalized adaptation, single detection, and missing feedback. Compared with traditional solutions, its core innovation lies in: building a three-dimensional perception system of muscle status and movement details through the collaboration of acceleration, angular velocity, and pressure sensors; generating adaptive training plans based on the user's static physical signs, dynamic goals, and historical training data, rather than relying on fixed preset modes; and achieving dynamic matching of parameters and muscle status through movement scoring and fatigue correction during training, taking into account effectiveness, safety, and comfort. Through technological innovation, this solution has promoted the upgrade of smart dumbbells from "manual adjustment tools" to "intelligent adaptive training systems", significantly improving the scientificity and safety of sports rehabilitation and fitness relaxation.

[0069] In some embodiments, after the score is dynamically corrected in combination with the user's current fatigue parameter, it also includes: during the training process, the action quality score result is fed back to the training plan generation algorithm module in real time. If the action standard score is lower than the preset threshold for multiple consecutive times, the training plan dynamic adjustment mechanism is triggered to automatically reduce the training weight or shorten the single set training time until the action standard score returns to a reasonable range, forming a closed-loop optimization control of the training plan and action quality.

[0070] After dynamic corrections to movement quality scores, a closed-loop feedback mechanism is established between training plans and movement quality. The movement standard scores are fed back to the training plan generation module in real time. If the scores fall below the preset threshold multiple times in a row, the plan is automatically adjusted, reducing the training load (weight / duration) until the scores rise again, forming a closed-loop control system of "score-adjustment-optimization."

[0071] The feedback trigger condition is achieved by setting a threshold for the action standard score (such as 70 points). When the score of a single set is lower than the threshold for three consecutive times, it is determined that the current training load exceeds the user's ability range.

[0072] Adjustment strategies include: Weight adjustment: Reduce the next training weight according to a preset reduction rule (such as 10% of the current weight), or switch to low vibration frequency mode (for smart dumbbell hitting scenarios). Duration adjustment: Shorten the duration of a single training set from 40 seconds to 30 seconds, while increasing the rest time (such as from 60 seconds to 90 seconds) to give muscles more time to recover. Closed-loop mechanism: After each adjustment, the system continues to monitor the subsequent exercise score. If the score rises above 80 points for two consecutive sets, the adjustment stops; if it continues to fall below the threshold, the load reduction strategy is triggered again (to a minimum of 50% of the initial load).

[0073] This prevents movement distortion caused by excessive load, reduces the risk of muscle strain and joint injury, and is particularly suitable for beginners or those experiencing fatigue. Dynamically calibrating training plans through real-time scoring feedback addresses the disconnect between plans and actual execution in traditional approaches, improving the adaptability of plan implementation. This complete closed loop of "data collection - quality assessment - plan adjustment" empowers the training system with self-optimization capabilities, rather than a one-way output plan.

[0074] In some embodiments, the real-time collection of user motion data during training by the built-in acceleration sensor, angular velocity sensor and pressure sensor of the smart dumbbell includes: using the acceleration sensor to collect the acceleration data of the dumbbell in three-dimensional space in real time, using the angular velocity sensor to collect the angular velocity data of the dumbbell around three rotation axes in real time, and using the pressure sensor to collect the grip pressure data of the user when holding the dumbbell in real time; based on the acceleration data and angular velocity data, the motion trajectory, motion speed and motion amplitude of the dumbbell are calculated, and based on the acceleration data, angular velocity data and grip pressure data, the force angle when the user exerts force is calculated.

[0075] By clarifying the specific implementation method of multi-sensor data collection: the acceleration sensor collects three-dimensional acceleration, the angular velocity sensor collects three-axis angular velocity, and the pressure sensor collects grip pressure; by fusing acceleration and angular velocity data to solve the motion trajectory, speed, and amplitude, and combining the three to calculate the force angle.

[0076] Sensor Data Collection: The accelerometer collects acceleration data along the X / Y / Z axes at a frequency of 200 Hz, including the gravitational acceleration component. The angular velocity sensor collects angular velocity (° / s) about the X / Y / Z axes for calculating the dumbbell's rotational position. A pressure sensor (such as an FSR thin-film sensor) is embedded in the handle, with two contacts on each side, to collect grip force (N) in real time.

[0077] Data Calculation Algorithm: Motion Trajectory: Acceleration and angular velocity data are fused through complementary filtering to eliminate gravity interference, resulting in the dumbbell's position coordinates (x, y, z) in three-dimensional space. The trajectory curve is generated by integrating the velocity data. Force Angle: Based on the dumbbell's posture matrix (derived from the integral of angular velocity) and the grip pressure distribution, the angle between the force direction and the body's coronal / sagittal planes (e.g., the angle between the forearm and upper arm during a curl) is calculated.

[0078] Through three-axis sensor fusion, millimeter-level trajectory and angle measurement accuracy (error ≤ 2°) is achieved, providing richer movement details (such as trajectory deviation direction and force tilt angle) than traditional single-axis pressure sensors. This provides high-precision input for subsequent movement scoring (trajectory consistency, force symmetry) and planning adjustments (weight strategy), solving the problem of crude parameter adjustment caused by a single detection method.

[0079] In some embodiments, the method dynamically generates a phased training plan based on user motion data, personalized parameters, historical training fitness and training goals, including action type, duration of each training group, interval time and weight adjustment strategy, including: the training plan generation algorithm module first analyzes the action completion, muscle recovery cycle and training effect data in the user's historical training records to determine the historical training fitness, and matches the appropriate action type from the preset action library in combination with the real-time physical state reflected by the user's current motion data, the training goals and muscle strength level in the personalized parameters; sets the duration of each training group according to the fatigue tolerance time of similar actions in the user's historical training, and sets the interval time between groups according to the recovery law of sports physiology; uses the weight adjustment strategy as a variable parameter of the training plan, so that the generated phased training plan includes a weight change strategy that is dynamically adjusted with the training stage.

[0080] By refining the training plan generation logic: matching action types based on historical training fitness (action completion, recovery cycle, effect data), setting group duration according to fatigue tolerance, setting interval time according to sports physiology laws, and embedding weight strategy as a dynamic parameter into the plan.

[0081] Historical training fitness analysis calculates the fitness score (0-100 points) by extracting the "action completion rate" (such as the percentage of completed sets of curls in the past 10 training sessions), "recovery time" (the time it takes for muscle soreness to disappear after training), and "strength growth rate" (the increase in maximum load per week) from historical records through the hierarchical analysis method.

[0082] Exercise Type Matching: The preset exercise library includes over 20 basic exercises (such as curls, presses, and lateral raises). Each exercise is labeled with a difficulty level (1-5) and the primary muscle group trained (such as biceps and deltoids). Based on the user's muscle strength level (e.g., beginner ≤ 10kg) and training goals (compound exercises prioritize muscle growth), 3-5 suitable exercises are selected.

[0083] Duration and Interval Settings: Set duration: Refer to historical tolerance for similar exercises (e.g., if a user has averaged 35 seconds per set of curls in the past, set the current set duration to 30 seconds, leaving a 5-second safety margin). Interval duration: Automatically adjusts based on the excess post-exercise oxygen consumption (EPOC) model, using the principle of "long intervals (90 seconds) for high-intensity exercises and short intervals (60 seconds) for low-intensity exercises."

[0084] Avoid the blindness of "universal templates" by leveraging historical data to accurately match user capabilities (e.g., avoiding difficult movements for beginners) to improve program feasibility. Set intervals based on exercise physiology (e.g., fully restoring the ATP-CP system) to ensure training effectiveness (e.g., focusing on between-set recovery for muscle growth) and reduce the risk of excessive fatigue.

[0085] In some embodiments, the weight adjustment strategy is adaptively adjusted according to the user's real-time force stability and historical strength growth curve, including: judging the user's real-time force stability by analyzing the fluctuation amplitude of the grip pressure data and force angle data collected in real time, if the fluctuation amplitude is within the preset stability threshold, the force is determined to be stable, otherwise it is determined to be unstable; calling the weight increase data of the same action type in the user's historical training records to form a historical strength growth curve, when the force is stable and the historical strength growth curve shows an upward trend, gradually increasing the training weight according to the preset incremental rules, when the force is unstable or the historical strength growth curve shows a plateau period, maintaining the current training weight or reducing the training weight according to the preset reduction rules.

[0086] By clarifying the adaptive logic of the weight adjustment strategy: judging the force stability through the fluctuations of grip pressure and force angle, and dynamically adjusting the load in combination with the historical strength growth curve (weight increase data), it is divided into three modes: "stable improvement", "unstable maintenance" and "plateau adjustment".

[0087] Force stability is determined by calculating the standard deviation of grip pressure (σ_pressure) and the fluctuation range of force angle (Δangle). If σ_pressure < 5N and Δangle < 5°, the force is considered stable; otherwise, it is unstable. Historical strength growth curve analysis: Weight data from the past 8 weeks of the same exercise are extracted and fitted with a growth curve (e.g., a linear regression model). If the slope is > 0 for 4 consecutive weeks, it is considered an upward trend; if the slope is ≤ 0 for 2 consecutive weeks, it is considered a plateau.

[0088] Dynamic adjustment rules include: Stability + Ascendance: Increase the weight in 5% increments (e.g., from 20kg to 21kg), no more than twice a week. Instability / Plateau: Maintain the current weight for two weeks. If instability persists, reduce the weight by 10% to avoid overloading and causing deformation.

[0089] Dynamic overloading based on actual capacity growth aligns with the principle of progressive muscle adaptation (a key principle for muscle growth), avoiding the blindness of fixed increments in traditional programs. Through real-time stability monitoring, the weight is automatically reduced before exhaustion, preventing injuries caused by a "lost control during the last set" of movements. This is particularly suitable for unsupervised home training scenarios.

[0090] In some embodiments, the construction of the movement quality scoring model includes: collecting standard movement data of professional trainees when completing different training movements through sports biomechanics experiments, the standard movement data including standard movement trajectory, standard movement speed range, standard movement amplitude and left and right force symmetry reference value; classifying the standard movement data according to the movement type, generating a standard movement template containing the corresponding standard trajectory curve, speed threshold range, amplitude reference value and force symmetry parameters, and storing the standard movement template in the database of the movement quality scoring model.

[0091] The process of constructing standardized movement templates collects standard data (trajectory, speed, amplitude, symmetry) of professional trainees through sports biomechanics experiments, generates templates by movement type classification, and stores them in the scoring model database.

[0092] The data collection experiment recruited 10 professional athletes and rehabilitation therapists. A high-precision motion capture system (e.g., Vicon) was used to capture standard movement data and simultaneously record sensor data (accuracy: ±1mm for trajectory and ±1° for angle). Fifty samples were collected for each movement, and after removing outliers, the average value was taken as the standard trajectory. The speed range was set to ±2σ (standard deviation) as the threshold (e.g., for a standard speed of 2° / s, the threshold range was 1.6-2.4° / s).

[0093] Template generation and storage: Create folders by movement type (e.g., "dumbbell curl," "shoulder press"). Each template contains: a standard trajectory curve (.csv format, time-coordinate sequence); speed threshold range (min_speed, max_speed); amplitude reference value (e.g., elbow flexion angle ≥ 90° during curls); and force symmetry parameters (left and right pressure difference ≤ 15N).

[0094] Template data is derived from professional sports biomechanics experiments, avoiding biases based on subjective experience and ensuring that scoring criteria align with the laws of human movement (for example, trajectory alignment directly reflects the correctness of muscle force paths). The template library is expandable (supporting the import of user-defined movements) to accommodate diverse training needs (rehabilitation, strength training, and functional training), enhancing system versatility.

[0095] In some embodiments, the motion data is matched with the standard motion template in multiple dimensions for analysis, and the motion standard score is calculated from four dimensions: motion trajectory consistency, speed uniformity, amplitude compliance rate, and left-right force symmetry, including: calculating the coordinate point fitting degree of the dumbbell motion trajectory collected in real time and the standard trajectory curve in the standard motion template to obtain the motion trajectory consistency score; comparing the real-time motion speed with the speed threshold range in the standard motion template, and calculating the proportion of time the speed value is within the threshold range as the speed uniformity score; comparing the real-time motion amplitude with the amplitude reference value in the standard motion template, and calculating the proportion of the number of movements that reach or exceed the amplitude reference value as the amplitude compliance rate score; obtaining the grip pressure data and force angle data of the left and right smart dumbbells, and calculating the degree of deviation of the difference value of the data on both sides from the symmetry reference value as the left-right force symmetry score; and calculating the total motion standard score by combining the scores of the four dimensions.

[0096] By clarifying the specific calculation methods of the four scoring dimensions: trajectory consistency is calculated by coordinate fitting, speed uniformity is the proportion of time within the threshold, amplitude compliance rate is the proportion of times the amplitude meets the standard, and the symmetry score is based on the deviation of the left and right data from the baseline value.

[0097] The trajectory fit (0-100 points) is calculated by aligning the real-time trajectory with the time series of the standard template using the dynamic time warping (DTW) algorithm, calculating the Euclidean distance of each frame coordinate, taking the average and normalizing it (the smaller the distance, the higher the score, the base distance corresponds to 70 points, and 1 point is added for each 1 mm reduction).

[0098] Speed ​​uniformity (0-100 points) is calculated by counting the proportion of time in a single set of movements that the speed is within the standard threshold range (for example, the total duration is 30 seconds, the time to reach the standard is 25 seconds, the score = 25 / 30×100≈83 points).

[0099] The range of motion (0-100 points) is determined by calculating whether the range of motion is ≥ the benchmark value (e.g., curl ≥ 90°) each time the movement is completed. The number of times the range of motion is achieved is divided by the total number of times × 100 (e.g., 8 out of 10 times, 80 points).

[0100] Left-right symmetry (0-100 points) is calculated by calculating the ratio of the absolute value of the grip pressure difference between the left and right hands to the symmetry reference value (e.g., if the reference difference is 15N and the actual difference is 20N, the score is 100-[(20-15) / 15×10]≈97 points. The greater the difference, the more points will be deducted).

[0101] By breaking down movement standards into calculable, objective metrics, we avoid the subjectivity of manual scoring (such as errors in the coach's visual judgment). Users can view weaknesses in each dimension in real time (for example, trajectory deviation is primarily in the X-axis). Scoring across four dimensions accurately identifies movement problems (e.g., uneven speed may be due to poor force application habits, and poor symmetry may indicate left-right muscle strength imbalance), guiding targeted corrections.

[0102] In some embodiments, the method of obtaining the user's current fatigue parameters based on continuous training time, historical heart rate data, and grip pressure fluctuation amplitude includes: counting the uninterrupted continuous training time in the user's current training stage, and marking the fatigue risk when the continuous training time exceeds the preset fatigue warning time; retrieving the heart rate data in the user's historical training records, analyzing the deviation between the current training heart rate and the historical average heart rate to determine the degree of cardiopulmonary fatigue; calculating the fluctuation amplitude of real-time grip pressure data, and determining that muscle control ability has decreased when the fluctuation amplitude exceeds a preset stability threshold; comprehensively considering the continuous training time, heart rate deviation amplitude, and grip pressure fluctuation amplitude, and determining the user's current fatigue parameters through preset fatigue assessment rules.

[0103] The calculation method of fatigue parameters is defined: comprehensive continuous training time (risk of exceeding the warning time), heart rate deviation amplitude (cardiopulmonary fatigue), grip pressure fluctuation (muscle control ability), and fatigue level is generated through preset rules (0-1, the higher the value, the deeper the fatigue level).

[0104] Continuous training duration: Set the fatigue warning duration (e.g. 45 minutes). For every 10 minutes exceeding the fatigue warning duration, the fatigue level increases by 0.1 (e.g. 55 minutes → 0.2).

[0105] Heart Rate Deviation Analysis: Retrieve your average heart rate from historical training sessions (HR_avg) and calculate the difference between your current heart rate (HR_current) and HR_avg (ΔHR% = (HR_current - HR_avg) / HR_avg). If ΔHR% exceeds 20%, increase the fatigue level by 0.1 for every 5% difference.

[0106] Pressure fluctuations were calculated by taking the standard deviation of grip pressure (σ_pressure) during a single set of movements and comparing it with the baseline value (σ0 in a fatigue-free state). When σ_pressure / σ0 > 1.5, fatigue was increased by 0.1 for every 0.1 difference above 1.5.

[0107] Comprehensive evaluation rule: Fatigue = duration weight × 0.4 + heart rate weight × 0.3 + pressure fluctuation weight × 0.3. The three factors are integrated through the fuzzy logic algorithm to output a continuous value between 0 and 1 (such as mild fatigue 0.3-0.5, moderate fatigue 0.5-0.7, severe fatigue >0.7).

[0108] This combination of physiological signals (heart rate) and movement signals (pressure fluctuations) provides a more comprehensive picture than heart rate monitoring alone (e.g., during static fatigue, heart rate may be normal, but muscle control may have decreased). It also dynamically captures early signs of fatigue (e.g., increased pressure fluctuations may indicate muscle compensation), avoiding overtraining, as traditional methods rely solely on subjective fatigue (which can be misjudged by the user).

[0109] In some embodiments, the score is dynamically corrected in combination with the user's current fatigue parameter, including: setting a correction coefficient for the action standard score according to the user's current fatigue parameter, and adaptively relaxing the score thresholds for action trajectory consistency and speed uniformity when the fatigue parameter shows that the user is in a mild fatigue state; when the fatigue parameter shows that the user is in a moderate fatigue state or above, in addition to relaxing the score threshold, an additional judgment on the safety of the action trajectory is added, and if there is a force deviation that may cause sports injury, the action standard score is directly reduced; through the correspondence between the fatigue parameter and the correction rule, dynamic adjustment of the action quality score is achieved.

[0110] By clarifying the dynamic correction rules for scores based on fatigue level: different correction strategies are set according to the degree of fatigue (mild / moderate / severe), the trajectory / speed threshold is relaxed for mild fatigue, and safety judgment is increased for moderate and above fatigue, and adaptive adjustment of scores is achieved through correction coefficients.

[0111] Correction Factor Settings: Mild Fatigue (0.3 ≤ Fatigue < 0.5): The trajectory consistency threshold is relaxed by 5% (e.g., passing score from 70 to 65 points), and the speed uniformity threshold is relaxed by 3% of the time percentage. Moderate Fatigue (0.5 ≤ Fatigue < 0.7): In addition to relaxing the thresholds, if a track shows joint hyperextension (e.g., elbow extension angle > 185°), 10 points will be deducted per session, triggering a safety alert. Severe Fatigue (≥ 0.7): The current training session is terminated, prompting a rest period, and the standard movement score is reduced by 50% (to prevent continued training while fatigued).

[0112] The dynamic adjustment mechanism automatically loads the corresponding correction rules based on the real-time fatigue parameters after each set of movements is completed, generating a corrected score (for example, the original trajectory score is 68 points, which is corrected to 72 points under mild fatigue, meeting the standard).

[0113] This approach avoids the frustration or forced persistence caused by strict grading even when fatigued, balancing training intensity and safety (e.g., allowing for some deviation in movement during mild fatigue to ensure training continuity). It also incorporates safety assessments (e.g., hyperextension angle detection) for moderate or higher fatigue, resulting in direct point deductions and warnings. This approach also provides protection against dangerous movements (e.g., joint locks) that can occur during fatigue, making it more intelligent than traditional solutions.

[0114] See also Figure 3 As shown, Figure 31 is a schematic diagram of the structure of the intelligent dumbbell training plan dynamic generation and movement quality scoring system 200 provided in an embodiment of the present application. The intelligent dumbbell training plan dynamic generation and movement quality scoring system 200 is used to execute the steps of the intelligent dumbbell training plan dynamic generation and movement quality scoring method shown in the above embodiments. The intelligent dumbbell training plan dynamic generation and movement quality scoring system 200 can be a single server or a server cluster, or the intelligent dumbbell training plan dynamic generation and movement quality scoring system 200 can be a terminal, which can be a handheld terminal, a laptop computer, a wearable device, or a robot.

[0115] like Figure 3 As shown, the intelligent dumbbell training plan dynamic generation and movement quality scoring system 200 includes:

[0116] The parameter acquisition unit 201 is configured to collect, in real time, the user's motion data during training using the built-in accelerometer, angular velocity sensor, and pressure sensor of the smart dumbbell. The motion data includes the dumbbell's motion trajectory, motion speed, motion amplitude, grip pressure, and force angle. The unit also acquires personalized parameters input by the user, including age, weight, body fat percentage, muscle strength level, training goals, and historical training records.

[0117] Strategy acquisition unit 202 is configured to input motion data and personalized parameters into a preset training plan generation algorithm module to dynamically generate a phased training plan including motion type, duration of each training set, rest time, and weight adjustment strategy based on the user's motion data, personalized parameters, historical training fitness, and training goals, and adaptively adjust the weight adjustment strategy based on the user's real-time force stability and historical strength growth curve;

[0118] The dynamic correction unit 203 is used to construct a movement quality scoring model. The scoring model pre-stores a standard movement template, which contains a standard trajectory curve, speed threshold range, amplitude reference value and force symmetry parameters corresponding to different training movements; the movement data is matched with the standard movement template in multiple dimensions for analysis, and the movement standard score is calculated from four dimensions: movement trajectory consistency, speed uniformity, amplitude compliance rate and left and right force symmetry; the user's current fatigue parameter is obtained according to the continuous training time, historical heart rate data and grip pressure fluctuation amplitude, and the score is dynamically corrected in combination with the user's current fatigue parameter.

[0119] In some embodiments, after the score is dynamically corrected in combination with the user's current fatigue parameter, it also includes: during the training process, the action quality score result is fed back to the training plan generation algorithm module in real time. If the action standard score is lower than the preset threshold for multiple consecutive times, the training plan dynamic adjustment mechanism is triggered to automatically reduce the training weight or shorten the single set training time until the action standard score returns to a reasonable range, forming a closed-loop optimization control of the training plan and action quality.

[0120] In some embodiments, the real-time collection of user motion data during training by the built-in acceleration sensor, angular velocity sensor and pressure sensor of the smart dumbbell includes: using the acceleration sensor to collect the acceleration data of the dumbbell in three-dimensional space in real time, using the angular velocity sensor to collect the angular velocity data of the dumbbell around three rotation axes in real time, and using the pressure sensor to collect the grip pressure data of the user when holding the dumbbell in real time; based on the acceleration data and angular velocity data, the motion trajectory, motion speed and motion amplitude of the dumbbell are calculated, and based on the acceleration data, angular velocity data and grip pressure data, the force angle when the user exerts force is calculated.

[0121] In some embodiments, the method dynamically generates a phased training plan based on user motion data, personalized parameters, historical training fitness and training goals, including action type, duration of each training group, interval time and weight adjustment strategy, including: the training plan generation algorithm module first analyzes the action completion, muscle recovery cycle and training effect data in the user's historical training records to determine the historical training fitness, and matches the appropriate action type from the preset action library in combination with the real-time physical state reflected by the user's current motion data, the training goals and muscle strength level in the personalized parameters; sets the duration of each training group according to the fatigue tolerance time of similar actions in the user's historical training, and sets the interval time between groups according to the recovery law of sports physiology; uses the weight adjustment strategy as a variable parameter of the training plan, so that the generated phased training plan includes a weight change strategy that is dynamically adjusted with the training stage.

[0122] In some embodiments, the weight adjustment strategy is adaptively adjusted according to the user's real-time force stability and historical strength growth curve, including: judging the user's real-time force stability by analyzing the fluctuation amplitude of the grip pressure data and force angle data collected in real time, if the fluctuation amplitude is within the preset stability threshold, the force is determined to be stable, otherwise it is determined to be unstable; calling the weight increase data of the same action type in the user's historical training records to form a historical strength growth curve, when the force is stable and the historical strength growth curve shows an upward trend, gradually increasing the training weight according to the preset incremental rules, when the force is unstable or the historical strength growth curve shows a plateau period, maintaining the current training weight or reducing the training weight according to the preset reduction rules.

[0123] In some embodiments, the construction of the movement quality scoring model includes: collecting standard movement data of professional trainees when completing different training movements through sports biomechanics experiments, the standard movement data including standard movement trajectory, standard movement speed range, standard movement amplitude and left and right force symmetry reference value; classifying the standard movement data according to the movement type, generating a standard movement template containing the corresponding standard trajectory curve, speed threshold range, amplitude reference value and force symmetry parameters, and storing the standard movement template in the database of the movement quality scoring model.

[0124] In some embodiments, the motion data is matched with the standard motion template in multiple dimensions for analysis, and the motion standard score is calculated from four dimensions: motion trajectory consistency, speed uniformity, amplitude compliance rate, and left-right force symmetry, including: calculating the coordinate point fitting degree of the dumbbell motion trajectory collected in real time and the standard trajectory curve in the standard motion template to obtain the motion trajectory consistency score; comparing the real-time motion speed with the speed threshold range in the standard motion template, and calculating the proportion of time the speed value is within the threshold range as the speed uniformity score; comparing the real-time motion amplitude with the amplitude reference value in the standard motion template, and calculating the proportion of the number of movements that reach or exceed the amplitude reference value as the amplitude compliance rate score; obtaining the grip pressure data and force angle data of the left and right smart dumbbells, and calculating the degree of deviation of the difference value of the data on both sides from the symmetry reference value as the left-right force symmetry score; and calculating the total motion standard score by combining the scores of the four dimensions.

[0125] In some embodiments, the method of obtaining the user's current fatigue parameters based on continuous training time, historical heart rate data, and grip pressure fluctuation amplitude includes: counting the uninterrupted continuous training time in the user's current training stage, and marking the fatigue risk when the continuous training time exceeds the preset fatigue warning time; retrieving the heart rate data in the user's historical training records, analyzing the deviation between the current training heart rate and the historical average heart rate to determine the degree of cardiopulmonary fatigue; calculating the fluctuation amplitude of real-time grip pressure data, and determining that muscle control ability has decreased when the fluctuation amplitude exceeds a preset stability threshold; comprehensively considering the continuous training time, heart rate deviation amplitude, and grip pressure fluctuation amplitude, and determining the user's current fatigue parameters through preset fatigue assessment rules.

[0126] In some embodiments, the score is dynamically corrected in combination with the user's current fatigue parameter, including: setting a correction coefficient for the action standard score according to the user's current fatigue parameter, and adaptively relaxing the score thresholds for action trajectory consistency and speed uniformity when the fatigue parameter shows that the user is in a mild fatigue state; when the fatigue parameter shows that the user is in a moderate fatigue state or above, in addition to relaxing the score threshold, an additional judgment on the safety of the action trajectory is added, and if there is a force deviation that may cause sports injury, the action standard score is directly reduced; through the correspondence between the fatigue parameter and the correction rule, dynamic adjustment of the action quality score is achieved.

[0127] It should be noted that those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described smart dumbbell training plan dynamic generation and movement quality scoring system and each module can refer to the corresponding contents in the above-mentioned smart dumbbell training plan dynamic generation and movement quality scoring method embodiments, and will not be repeated here.

[0128] The above-mentioned method for dynamically generating a training plan and scoring the quality of movement of the intelligent dumbbell can be implemented in the form of a computer program. The computer program can be used in a computer system such as Figure 3 Run on the device shown.

[0129] See also Figure 4 , Figure 4 : This is a schematic block diagram of the structure of the smart dumbbell provided by an embodiment of the present application. The smart dumbbell includes a processor, a memory, and a network interface connected via a device bus, wherein the memory may include a storage medium and an internal memory.

[0130] The storage medium can store an operating device and a computer program. The computer program includes program instructions, which, when executed, can cause a processor to execute any method for dynamically generating a training plan and scoring movement quality for a smart dumbbell.

[0131] The processor is used to provide computing and control capabilities to support the operation of the entire smart dumbbell.

[0132] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, the processor can execute any method for dynamically generating a training plan and scoring movement quality of the smart dumbbell.

[0133] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art will understand that Figure 4The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the terminal to which the solution of the present application is applied. The specific smart dumbbell may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0134] It should be understood that the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0135] In one embodiment, the processor is configured to execute a computer program stored in the memory to implement the following steps:

[0136] The smart dumbbells use built-in accelerometers, angular velocity sensors, and pressure sensors to collect real-time motion data from users during training. The motion data includes dumbbell motion trajectory, speed, amplitude, grip pressure, and force angle. The smart dumbbells also capture personalized parameters entered by the user, including age, weight, body fat percentage, muscle strength level, training goals, and historical training records.

[0137] Inputting movement data and personalized parameters into a preset training plan generation algorithm module, the module dynamically generates a phased training plan based on the user's movement data, personalized parameters, historical training fitness, and training goals, including movement type, training duration per set, rest time, and weight adjustment strategy. The module also adaptively adjusts the weight adjustment strategy based on the user's real-time force stability and historical strength growth curve.

[0138] A movement quality scoring model is constructed, which pre-stores standard movement templates. The standard movement templates contain standard trajectory curves, speed threshold ranges, amplitude reference values, and force symmetry parameters corresponding to different training movements. The movement data and the standard movement templates are matched and analyzed in multiple dimensions. The movement standard score is calculated from four dimensions: movement trajectory consistency, speed uniformity, amplitude compliance rate, and left-right force symmetry. The user's current fatigue parameters are obtained based on the continuous training duration, historical heart rate data, and grip pressure fluctuation amplitude. The score is dynamically corrected based on the user's current fatigue parameters.

[0139] In some embodiments, after the score is dynamically corrected in combination with the user's current fatigue parameter, it also includes: during the training process, the action quality score result is fed back to the training plan generation algorithm module in real time. If the action standard score is lower than the preset threshold for multiple consecutive times, the training plan dynamic adjustment mechanism is triggered to automatically reduce the training weight or shorten the single set training time until the action standard score returns to a reasonable range, forming a closed-loop optimization control of the training plan and action quality.

[0140] In some embodiments, the real-time collection of user motion data during training by the built-in acceleration sensor, angular velocity sensor and pressure sensor of the smart dumbbell includes: using the acceleration sensor to collect the acceleration data of the dumbbell in three-dimensional space in real time, using the angular velocity sensor to collect the angular velocity data of the dumbbell around three rotation axes in real time, and using the pressure sensor to collect the grip pressure data of the user when holding the dumbbell in real time; based on the acceleration data and angular velocity data, the motion trajectory, motion speed and motion amplitude of the dumbbell are calculated, and based on the acceleration data, angular velocity data and grip pressure data, the force angle when the user exerts force is calculated.

[0141] In some embodiments, the method dynamically generates a phased training plan based on user motion data, personalized parameters, historical training fitness and training goals, including action type, duration of each training group, interval time and weight adjustment strategy, including: the training plan generation algorithm module first analyzes the action completion, muscle recovery cycle and training effect data in the user's historical training records to determine the historical training fitness, and matches the appropriate action type from the preset action library in combination with the real-time physical state reflected by the user's current motion data, the training goals and muscle strength level in the personalized parameters; sets the duration of each training group according to the fatigue tolerance time of similar actions in the user's historical training, and sets the interval time between groups according to the recovery law of sports physiology; uses the weight adjustment strategy as a variable parameter of the training plan, so that the generated phased training plan includes a weight change strategy that is dynamically adjusted with the training stage.

[0142] In some embodiments, the weight adjustment strategy is adaptively adjusted according to the user's real-time force stability and historical strength growth curve, including: judging the user's real-time force stability by analyzing the fluctuation amplitude of the grip pressure data and force angle data collected in real time, if the fluctuation amplitude is within the preset stability threshold, the force is determined to be stable, otherwise it is determined to be unstable; calling the weight increase data of the same action type in the user's historical training records to form a historical strength growth curve, when the force is stable and the historical strength growth curve shows an upward trend, gradually increasing the training weight according to the preset incremental rules, when the force is unstable or the historical strength growth curve shows a plateau period, maintaining the current training weight or reducing the training weight according to the preset reduction rules.

[0143] In some embodiments, the construction of the movement quality scoring model includes: collecting standard movement data of professional trainees when completing different training movements through sports biomechanics experiments, the standard movement data including standard movement trajectory, standard movement speed range, standard movement amplitude and left and right force symmetry reference value; classifying the standard movement data according to the movement type, generating a standard movement template containing the corresponding standard trajectory curve, speed threshold range, amplitude reference value and force symmetry parameters, and storing the standard movement template in the database of the movement quality scoring model.

[0144] In some embodiments, the motion data is matched with the standard motion template in multiple dimensions for analysis, and the motion standard score is calculated from four dimensions: motion trajectory consistency, speed uniformity, amplitude compliance rate, and left-right force symmetry, including: calculating the coordinate point fitting degree of the dumbbell motion trajectory collected in real time and the standard trajectory curve in the standard motion template to obtain the motion trajectory consistency score; comparing the real-time motion speed with the speed threshold range in the standard motion template, and calculating the proportion of time the speed value is within the threshold range as the speed uniformity score; comparing the real-time motion amplitude with the amplitude reference value in the standard motion template, and calculating the proportion of the number of movements that reach or exceed the amplitude reference value as the amplitude compliance rate score; obtaining the grip pressure data and force angle data of the left and right smart dumbbells, and calculating the degree of deviation of the difference value of the data on both sides from the symmetry reference value as the left-right force symmetry score; and calculating the total motion standard score by combining the scores of the four dimensions.

[0145] In some embodiments, the method of obtaining the user's current fatigue parameters based on continuous training time, historical heart rate data, and grip pressure fluctuation amplitude includes: counting the uninterrupted continuous training time in the user's current training stage, and marking the fatigue risk when the continuous training time exceeds the preset fatigue warning time; retrieving the heart rate data in the user's historical training records, analyzing the deviation between the current training heart rate and the historical average heart rate to determine the degree of cardiopulmonary fatigue; calculating the fluctuation amplitude of real-time grip pressure data, and determining that muscle control ability has decreased when the fluctuation amplitude exceeds a preset stability threshold; comprehensively considering the continuous training time, heart rate deviation amplitude, and grip pressure fluctuation amplitude, and determining the user's current fatigue parameters through preset fatigue assessment rules.

[0146] In some embodiments, the score is dynamically corrected in combination with the user's current fatigue parameter, including: setting a correction coefficient for the action standard score according to the user's current fatigue parameter, and adaptively relaxing the score thresholds for action trajectory consistency and speed uniformity when the fatigue parameter shows that the user is in a mild fatigue state; when the fatigue parameter shows that the user is in a moderate fatigue state or above, in addition to relaxing the score threshold, an additional judgment on the safety of the action trajectory is added, and if there is a force deviation that may cause sports injury, the action standard score is directly reduced; through the correspondence between the fatigue parameter and the correction rule, dynamic adjustment of the action quality score is achieved.

[0147] The present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the processor implements the steps of the method for dynamically generating a training plan and scoring movement quality for smart dumbbells as provided in any embodiment of the present application.

[0148] The computer-readable storage medium may be an internal storage unit of the smart dumbbell described in the aforementioned embodiment, such as a hard disk or memory of the smart dumbbell. The computer-readable storage medium may also be an external storage device of the smart dumbbell, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash memory card, etc., equipped on the smart dumbbell.

[0149] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present application, and such modifications or substitutions should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A method for dynamically generating a training plan and scoring movement quality for a smart dumbbell, characterized in that: Applied to smart dumbbells; the method includes: The smart dumbbells use built-in accelerometers, angular velocity sensors, and pressure sensors to collect real-time motion data from users during training. The motion data includes dumbbell motion trajectory, speed, amplitude, grip pressure, and force angle. The smart dumbbells also capture personalized parameters entered by the user, including age, weight, body fat percentage, muscle strength level, training goals, and historical training records. Inputting movement data and personalized parameters into a preset training plan generation algorithm module, the module dynamically generates a phased training plan based on the user's movement data, personalized parameters, historical training fitness, and training goals, including movement type, training duration per set, rest time, and weight adjustment strategy. The module also adaptively adjusts the weight adjustment strategy based on the user's real-time force stability and historical strength growth curve. A movement quality scoring model is constructed, which pre-stores standard movement templates. The standard movement templates contain standard trajectory curves, speed threshold ranges, amplitude reference values, and force symmetry parameters corresponding to different training movements. The movement data and the standard movement templates are matched and analyzed in multiple dimensions. The movement standard score is calculated from four dimensions: movement trajectory consistency, speed uniformity, amplitude compliance rate, and left-right force symmetry. The user's current fatigue parameters are obtained based on the continuous training duration, historical heart rate data, and grip pressure fluctuation amplitude. The score is dynamically corrected based on the user's current fatigue parameters.

2. The method according to claim 1, characterized in that After dynamically revising the score based on the user's current fatigue parameter, the method further includes: During the training process, the movement quality score results are fed back to the training plan generation algorithm module in real time. If the movement standard score is lower than the preset threshold for multiple consecutive times, the dynamic adjustment mechanism of the training plan will be triggered, automatically reducing the training weight or shortening the duration of a single set of training until the movement standard score returns to a reasonable range, forming a closed-loop optimization control of the training plan and movement quality.

3. The method according to claim 1, characterized in that The smart dumbbells use built-in acceleration sensors, angular velocity sensors, and pressure sensors to collect real-time motion data during user training, including: The acceleration sensor is used to collect the acceleration data of the dumbbell in three-dimensional space in real time. The angular velocity sensor is used to collect the angular velocity data of the dumbbell around three rotation axes in real time. The pressure sensor is used to collect the grip pressure data when the user holds the dumbbell in real time. The motion trajectory, motion speed and motion amplitude of the dumbbell are calculated based on the acceleration data and angular velocity data, and the force angle when the user exerts force is calculated based on the acceleration data, angular velocity data and grip pressure data.

4. The method according to claim 1, wherein The method dynamically generates a phased training plan based on user action data, personalized parameters, historical training fitness, and training goals, including action types, training duration per set, rest time, and weight adjustment strategy, including: The training plan generation algorithm module first analyzes the user's historical training records for movement completion, muscle recovery cycle, and training effect data to determine historical training fitness. It then combines the user's current movement data with the real-time physical fitness status, training goals, and muscle strength level in the personalized parameters to match appropriate movement types from a preset movement library. The training duration of each group is set according to the fatigue tolerance duration of similar actions in the user's historical training, and the interval time between groups is set according to the recovery law of sports physiology; the weight adjustment strategy is used as a variable parameter of the training plan, so that the generated phased training plan includes a weight change strategy that is dynamically adjusted with the training stage.

5. The method according to claim 1, wherein The weight adjustment strategy is adaptively adjusted based on the user's real-time force stability and historical strength growth curve, including: By analyzing the fluctuation amplitude of the grip pressure data and force angle data collected in real time, the user's real-time force stability is determined. If the fluctuation amplitude is within the preset stability threshold, the force is determined to be stable; otherwise, it is determined to be unstable. The weight increase data of the same action type in the user's historical training records is retrieved to form a historical strength growth curve. When the force is stable and the historical strength growth curve shows an upward trend, the training weight is gradually increased according to the preset incremental rules. When the force is unstable or the historical strength growth curve reaches a plateau, the current training weight is maintained or the training weight is reduced according to the preset reduction rules.

6. The method according to claim 1, characterized in that The step of constructing the action quality scoring model includes: Collect standard movement data of professional trainees when completing different training movements through sports biomechanics experiments. The standard movement data includes standard movement trajectory, standard movement speed range, standard movement amplitude and left and right force symmetry benchmark value; The standard action data is classified according to the action type to generate a standard action template including the corresponding standard trajectory curve, speed threshold range, amplitude reference value and force symmetry parameter, and the standard action template is stored in the database of the action quality scoring model.

7. The method according to claim 1, characterized in that The motion data is matched with the standard motion template in multiple dimensions for analysis, and the motion standard score is calculated from four dimensions: motion trajectory consistency, speed uniformity, amplitude compliance rate, and left and right force symmetry, including: The coordinate point fitting degree of the dumbbell motion trajectory collected in real time is calculated with the standard trajectory curve in the standard motion template to obtain the motion trajectory fitting degree score; The real-time movement speed is compared with the speed threshold range in the standard movement template, and the proportion of time the speed value is within the threshold range is calculated as the speed uniformity score; the real-time movement amplitude is compared with the amplitude reference value in the standard movement template, and the proportion of movements that reach or exceed the amplitude reference value is calculated as the amplitude compliance rate score; Obtain the grip pressure data and force angle data of the left and right smart dumbbells, calculate the difference between the two sides of the data and the degree of deviation from the symmetry reference value as the left and right force symmetry score; and calculate the total score of the movement standard by combining the scores of the four dimensions.

8. The method according to claim 1, characterized in that The method of obtaining the user's current fatigue parameter based on the continuous training duration, historical heart rate data, and grip pressure fluctuation amplitude includes: Count the duration of uninterrupted continuous training in the user's current training phase, and mark fatigue risk when the continuous training duration exceeds the preset fatigue warning duration; retrieve the heart rate data from the user's historical training records, analyze the deviation between the current training heart rate and the historical average heart rate to determine the degree of cardiopulmonary fatigue; Calculate the fluctuation amplitude of real-time grip pressure data and judge that muscle control ability has declined when the fluctuation amplitude exceeds the preset stability threshold; The user's current fatigue parameters are determined by comprehensively considering the continuous training duration, heart rate deviation and grip pressure fluctuation, using preset fatigue assessment rules.

9. The method according to claim 1, characterized in that The dynamic correction of the score based on the user's current fatigue parameter includes: The correction coefficient of the action standard score is set according to the user's current fatigue parameter. When the fatigue parameter shows that the user is in a state of mild fatigue, the score thresholds for the action trajectory consistency and speed uniformity are adaptively relaxed; When the fatigue parameter indicates that the user is in a moderate or higher fatigue state, in addition to relaxing the score threshold, an additional assessment of the safety of the movement trajectory is added. If there is a force deviation that may cause sports injury, the movement standard score will be directly reduced. Through the correspondence between fatigue parameters and correction rules, dynamic adjustment of movement quality scores can be achieved.

10. A system for dynamically generating training plans and scoring movement quality for smart dumbbells, characterized in that: Applied to smart dumbbells; the system includes: A parameter acquisition unit is used to collect real-time motion data of the user during training through the built-in acceleration sensor, angular velocity sensor, and pressure sensor of the smart dumbbell. The motion data includes dumbbell motion trajectory, motion speed, motion amplitude, grip pressure, and force angle; and obtain personalized parameters input by the user, including age, weight, body fat percentage, muscle strength level, training goals, and historical training records; A strategy acquisition unit, configured to input motion data and personalized parameters into a preset training plan generation algorithm module, dynamically generating a phased training plan including movement type, duration of each training set, rest time, and weight adjustment strategy based on the user's motion data, personalized parameters, historical training fitness, and training goals, and adaptively adjusting the weight adjustment strategy based on the user's real-time force stability and historical strength growth curve; A dynamic correction unit is used to construct a movement quality scoring model. The scoring model pre-stores standard movement templates, which contain standard trajectory curves, speed threshold ranges, amplitude reference values, and force symmetry parameters corresponding to different training movements. The movement data is matched with the standard movement templates in multiple dimensions for analysis, and the movement standard score is calculated from four dimensions: movement trajectory consistency, speed uniformity, amplitude compliance rate, and left and right force symmetry. The user's current fatigue parameters are obtained based on the continuous training time, historical heart rate data, and grip pressure fluctuation amplitude, and the score is dynamically corrected in combination with the user's current fatigue parameters.

Images