Ping-pong ball hitting force quantitative training device based on pressure sensing

Abstract

The invention discloses a ping-pong ball hitting strength quantitative training device based on pressure sensing, and belongs to the field of ping-pong training, the ping-pong ball hitting strength quantitative training device comprises a multi-dimensional pressure sensing fusion module, a three-in-one sensing layout of a racket, a grab handle and a table top is adopted, and comprehensive collection and accurate quantification of strength data are achieved; the data processing module carries an embedded chip, integrates a multi-sensor data fusion algorithm and a machine learning model, and realizes real-time processing and intelligent analysis of data; and the multi-module real-time feedback module adopts a multi-mode feedback mode of racket end instant prompt, terminal visualization and voice guidance, so that a trainer can perceive the force exerting deviation in real time and quickly adjust the action. According to the table tennis ball hitting force quantitative training device based on pressure sensing, multi-dimensional pressure sensing fusion and real-time intelligent feedback are combined, the limitation of traditional single sensing and passive data collection is broken through, and accurate quantification of ball hitting force and intelligent distinguishing of ball hitting types are achieved.

Description

Technical Field

[0001] This invention relates to the field of table tennis training technology, and in particular to a table tennis hitting force quantification training device based on pressure sensing. Background Technology

[0002] In current table tennis training, the assessment of hitting power relies heavily on the coach's subjective judgment, lacking objective quantitative standards; beginners find it difficult to perceive changes in their own power, easily developing incorrect power habits; existing quantitative devices mostly focus on collecting single power data, failing to achieve a closed loop of "data collection-real-time feedback," and lack the ability to differentiate the power of different hitting types (forehand attack, backhand flick, topspin, etc.). Summary of the Invention

[0003] The purpose of this invention is to provide a table tennis hitting force quantification training device based on pressure sensing. By combining multi-dimensional pressure sensing fusion and real-time intelligent feedback, it breaks through the limitations of traditional single sensing and passive data collection, and realizes accurate quantification of hitting force and intelligent differentiation of hitting type.

[0004] To achieve the above objectives, the present invention provides a pressure-sensing-based training device for quantifying the force of a table tennis shot, comprising: The multi-dimensional pressure sensing fusion module adopts a three-in-one sensing layout of ping-pong paddle, ping-pong paddle handle, and ping-pong table surface to achieve comprehensive collection and accurate quantification of force data. The data processing module, equipped with an embedded chip, integrates multi-sensor data fusion algorithms and machine learning models to achieve real-time data processing and intelligent analysis. The multi-module real-time feedback module uses a multi-modal feedback approach, including real-time prompts at the racket end, terminal visualization, and voice guidance, to allow trainees to perceive force deviations in real time and quickly adjust their movements. The multi-dimensional pressure sensing fusion module includes: The racket pressure sensing unit is used to monitor the finger pressure of the trainee's penhold grip and shakehand grip, as well as the pressure at the point of impact on the racket face; The grip pressure sensing unit is used to monitor the finger pressure of the trainee's penhold grip and shakehand grip, as well as changes in grip strength. The table tennis table surface pressure sensing unit adopts a distributed layout near the net, in the middle, and far from the net to monitor the rebound pressure of the ping-pong ball after landing. The data processing module includes a data fusion and force quantification unit and a shot type recognition unit; The multi-module real-time feedback module includes an LED instant prompt unit at the racket end, a terminal visual interface unit, and an intelligent voice guidance unit.

[0005] Preferably, the racket pressure sensing unit includes a racket face impact pressure sensing area and a finger pressure sensing area, with the finger pressure sensing area located at the end of the racket face rubber near the handle. The racket face impact pressure sensing area is used to monitor and collect the pressure distribution, pressure peak, and pressure duration at the moment of impact. The impact point is identified by the pressure distribution characteristics. At the same time, the racket face impact pressure sensing area monitors the pressure on the index and ring fingers of the trainee's penhold grip. The finger pressure sensing area is used to monitor the pressure of the index finger and thumb in the shakehand grip, and the pressure of the index finger, thumb, and little finger in the penhold grip.

[0006] Preferably, the racket pressure sensing unit adopts a flexible pressure sensing array, which is adapted to the curved shape of the table tennis racket rubber. The flexible pressure sensing array is embedded between the table tennis racket rubber and the racket blade. The flexible pressure sensing array includes several piezoelectric thin film sensors.

[0007] Preferably, the grip pressure sensing unit includes a racket blade pressure sensing area and a grip shaft pressure sensing area. The racket blade pressure sensing area is located in the blade area between the racket rubber and the grip shaft. The racket blade pressure sensing area is used to monitor the pressure of the trainee's index finger in the shakehand grip and the pressure of the trainee's thumb and index finger in the penhold grip. The pressure sensing area of ​​the grip bar is used to monitor the trainee's hand grip strength and the pressure of the trainee's middle finger, ring finger, and little finger when using a horizontal grip. The racket blade pressure sensing area is evenly distributed with several pressure monitoring points, each pressure monitoring point is embedded with a pressure sensor, and the grip shaft pressure sensing area is embedded with several sets of ring pressure sensors. The racket pressure sensing unit and the grip pressure sensing unit jointly collect the changes in the grip force of the trainee when hitting the ball.

[0008] Preferably, the tabletop pressure sensing unit includes a near-net sensing group, an intermediate sensor, and a far-net sensing group, all of which are symmetrically arranged on both sides of the central net on the tabletop. The near-net sensor group includes near-net sensors symmetrically arranged about the vertical center line of the table surface to monitor the pressure in the near-net bounce area. The near-net sensors include near-net sensor one and near-net sensor two, which are located on the same horizontal line of the table surface near the net. Near-net sensor one is located at the edge of the side of the table surface, and near-net sensor two is located in the middle between near-net sensor one and the vertical center line of the table surface. The intermediate sensor is located between the near-net sensor group and the far-net sensor group, and is used to monitor the pressure of the ball rebounding in the middle area on one side of the table surface; The far-net sensor group includes far-net sensors symmetrically arranged about the vertical center line of the table surface to monitor the pressure in the far-net ball bounce area. The far-net sensors include far-net sensor one and far-net sensor two, which are located on the same horizontal line of the table surface near the end of the table. Far-net sensor one is located at the edge of the side of the table surface, and far-net sensor two is located in the middle between far-net sensor one and the vertical center line of the table surface.

[0009] Preferably, the near-grid sensor, intermediate sensor, and far-grid sensor all employ embedded miniature pressure sensors, which are evenly distributed.

[0010] Preferably, the data fusion and force quantification unit uses a weighted fusion algorithm to fuse and calculate the peak racket face pressure, peak grip force, and table surface rebound pressure data. Combined with pre-calibrated correction coefficients for different racket grip types and rubber types, it outputs the hitting force value, as shown in the following formula: The formula for calculating the initial fusion strength value is as follows: ; in, This is the initial fusion strength value. , , These are the weighting coefficients for racket face, grip, and table surface pressure data, respectively. This represents the peak pressure on the racket face. This represents the peak force exerted by the grip. This represents the peak of the rebound pressure on the platform. The initial fusion strength calculation formula has been revised, and the revised precise strength calculation formula is as follows: ; in, To correct the precise force value, This is a correction factor for racket type. This is a correction factor for the rubber type. The formula for calculating the impulse of a shot is: ; in, To generate momentum for the shot, The duration of the pressure applied to the slapping surface.

[0011] Preferably, the shot type recognition unit adopts a random forest algorithm based on machine learning. It uses racket face pressure distribution features, grip force curve features, and the rhythm of shot force changes as input features to pre-train recognition models for forehand attack, backhand flick, forehand topspin, and backhand topspin. The core formulas are the feature vector construction formula and the random forest voting decision formula, as follows: Formula for constructing input feature vectors: ; in, The input feature vector is 12-dimensional; The pressure distribution characteristics of the racket face are represented by the horizontal and vertical coordinates of the pressure peak position, the distribution entropy, and the peak concentration, respectively. The force curve characteristics of the grip are represented by the peak force of the thumb, the peak force of the index finger, the time of occurrence of the peak force, and the amplitude of curve fluctuation. The rhythm characteristics of the change in hitting force are represented by the rate of increase in force, the rate of decrease in force, the duration of peak stability, and the variance of force change, respectively. Random Forest Voting Decision Formula: ; in, The type of shot for the final output; It is a collection of 6 common shot types; The number of decision trees in the random forest; For the first Each decision tree is used to process the input feature vector. The prediction results; The indicator function is to count the number of votes for a certain type of shot predicted by all decision trees, and the type with the most votes is the final identification result.

[0012] Preferably, the racket end LED instant prompt unit includes a green LED light, a yellow LED light, and a red LED light set at the end of the racket handle, which correspond to the force being up to standard and the hitting type being matched, the force deviation being small, the force deviation being large, or the hitting type being incorrect, respectively. The terminal visualization interface unit transmits real-time data to a mobile phone or tablet APP via Bluetooth. The interface displays the hitting force value, hitting impulse, hitting point position, hitting type, and grip force curve data in real time, and shows the trend of continuous hitting force changes in the form of a line graph. The intelligent voice guidance unit outputs optimization suggestions in real time via voice through a mobile phone or tablet app.

[0013] Preferably, it also includes a cloud data management module, which is used to synchronize training data to a cloud server to achieve permanent storage of historical data and synchronization across multiple devices.

[0014] Therefore, the present invention employs the above-mentioned pressure-sensing-based table tennis hitting force quantification training device, which has the following beneficial effects: 1. Multi-dimensional sensor fusion: Breaking through the limitations of single sensors, it improves the accuracy of force quantification and the ability to identify the type of shot by coordinating the monitoring of racket face, grip, table surface, near net, middle, and far net. 2. Multimodal real-time feedback: Combining LED prompts, visualized data, and voice guidance, it provides real-time feedback to help trainees quickly adjust their movements; 3. Personalized adaptation: Supports modification of different racket / rubber types and dynamic adjustment of training goals to adapt to the different training needs of players from beginners to professional players.

[0015] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the device structure of an embodiment of a table tennis hitting force quantification training device based on pressure sensing according to the present invention. Figure 2 This is a front structural diagram of a table tennis penhold grip according to an embodiment of a table tennis hitting force quantification training device based on pressure sensing according to the present invention. Figure 3 This is a schematic diagram of the reverse side structure of a ping-pong penhold grip according to an embodiment of the ping-pong hitting force quantification training device based on pressure sensing of the present invention. Figure 4 This is a front structural diagram of a table tennis racket grip according to an embodiment of a table tennis hitting force quantification training device based on pressure sensing according to the present invention. Figure 5 This is a schematic diagram of the reverse side structure of a table tennis racket grip according to an embodiment of a table tennis hitting force quantification training device based on pressure sensing according to the present invention. Figure 6 This is a schematic diagram of the sensor distribution on the table surface of a table tennis ball striking force quantification training device based on pressure sensing, according to an embodiment of the present invention.

[0017] In the diagram: 1. Racket pressure sensing unit; 2. Grip pressure sensing unit; 3. Racket face hitting pressure sensing area; 4. Finger pressure sensing area; 5. Racket blade pressure sensing area; 6. Grip shaft pressure sensing area; 7. Near-net sensor; 8. Middle sensor; 9. Far-net sensor; 10. Near-net sensor one; 11. Near-net sensor two; 12. Far-net sensor one; 13. Far-net sensor two; 14. Vertical center line of the table surface; 15. Horizontal line of the table surface. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages disclosed in the embodiments of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the embodiments of the present invention and are not intended to limit the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments in this application without creative effort are within the scope of protection of this application. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout.

[0019] It should be noted that the terms “comprising” and “having”, and any variations thereof, are intended to cover non-exclusive inclusion, such as a process, method, apparatus, product, or server that includes a series of steps or units, not necessarily limited to those steps or units that are explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to such processes, methods, products, or apparatus.

[0020] Similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0021] In the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship in which the product of this invention is usually placed when in use. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting this invention.

[0022] In the description of this invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," and "connect" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0023] Example: like Figure 1 As shown, the table tennis hitting force quantification training device based on pressure sensing of the present invention includes: The multi-dimensional pressure sensing fusion module adopts a three-in-one sensing layout of ping-pong paddle, ping-pong paddle handle, and ping-pong table surface to achieve comprehensive collection and accurate quantification of force data. The data processing module is equipped with a high-performance STM32H743 embedded chip (located inside the handle of the ping-pong paddle), and integrates multi-sensor data fusion algorithms and machine learning models to achieve real-time data processing and intelligent analysis. The multi-module real-time feedback module uses a multi-modal feedback approach, including real-time prompts at the racket end, terminal visualization, and voice guidance, allowing trainees to perceive force deviations in real time and quickly adjust their movements.

[0024] The multi-dimensional pressure sensing fusion module includes racket pressure sensing unit 1, grip pressure sensing unit 2, and table surface pressure sensing unit.

[0025] like Figure 2 , Figure 3 , Figure 4 , Figure 5 As shown, the racket pressure sensing unit 1 is used to monitor the finger pressure of the trainee using a penhold grip and a shakehand grip, as well as the pressure at the point of impact on the racket face. The racket pressure sensing unit 1 includes a racket face impact pressure sensing area 3 and a finger pressure sensing area 4, with the finger pressure sensing area 4 located near the end of the rubber surface close to the handle. The racket face impact pressure sensing area 3 monitors and collects the pressure distribution, peak pressure, and pressure duration at the moment of impact, identifying the point of impact through pressure distribution characteristics. Simultaneously, the racket face impact pressure sensing area 3 monitors the pressure of the trainee's index and ring fingers using a penhold grip. The finger pressure sensing area 4 monitors the pressure of the trainee's index and thumb using a shakehand grip, and the pressure of the trainee's index, thumb, and little fingers using a penhold grip. The racket pressure sensing unit 1 employs a flexible pressure sensor array, adapted to the curved shape of the table tennis racket rubber surface. A flexible pressure sensor array is embedded between the rubber and the blade of a table tennis racket. The flexible pressure sensor array includes several piezoelectric thin film sensors (using PVDF piezoelectric thin film sensors with a thickness of <0.1mm, which does not affect the original feel of the racket).

[0026] The grip pressure sensing unit 2 is used to monitor the finger pressure of the trainee using both penhold and shakehand grips, as well as changes in grip strength. The grip pressure sensing unit 2 includes a racket blade pressure sensing area 5 and a grip shaft pressure sensing area 6. The racket blade pressure sensing area 5 is located in the racket blade area between the racket rubber and the grip shaft. The racket blade pressure sensing area 5 monitors the pressure of the trainee's index finger in a shakehand grip and the pressure of the thumb and index finger in a penhold grip. The grip shaft pressure sensing area 6 monitors the trainee's hand grip strength and the pressure of the middle, ring, and little fingers in a shakehand grip. The racket blade pressure sensing area 5 has several pressure monitoring points evenly distributed, each embedded with a pressure sensor. The grip shaft pressure sensing area 6 has several sets of ring-shaped pressure sensors embedded in it. The racket pressure sensing unit 1 and the grip pressure sensing unit 2 jointly collect changes in the trainee's grip strength during hitting the ball. The grip strength curves for different types of shots (such as forehand attack which requires the index finger to generate the main force, and backhand flick which requires the thumb to generate the main force) are significantly different. By integrating racket face pressure data and grip strength curves, intelligent recognition of shot type can be achieved (recognition accuracy ≥90%).

[0027] like Figure 6As shown, the table tennis table pressure sensing unit adopts a distributed layout of near-net, middle, and far-net areas to monitor the rebound pressure of the table tennis ball after landing. The table tennis table pressure sensing unit includes a near-net sensing group, a middle sensor 8, and a far-net sensing group, all symmetrically arranged on both sides of the middle net on the table tennis table surface. The near-net sensing group includes near-net sensors 7 symmetrically arranged about the vertical center line 14 of the table tennis table surface, used to monitor the pressure in the rebound area of ​​short balls and fast push balls hit near the net. Near-net sensors 7 include near-net sensor one 10 and near-net sensor two 11, located on the same horizontal line 15 of the table tennis table surface near the net. Near-net sensor one 10 is located at the side edge of the table tennis table surface, and near-net sensor two 11 is located in the middle position between near-net sensor one 10 and the vertical center line 14 of the table tennis table surface. Near-net sensor 10 and near-net sensor 21 are located 5-8cm from the net. The intermediate sensor 8 is positioned between the near-net sensor group 7 and the far-net sensor group 9, monitoring the pressure of the ball's rebound in the central area of ​​one side of the table. The far-net sensor group includes far-net sensors 9 symmetrically arranged about the vertical center line 14 of the table surface, monitoring the pressure in the far-net rebound area. Far-net sensors 9 include far-net sensor 12 and far-net sensor 23, located on the same horizontal line 15 of the table surface near the end of the table. Far-net sensor 12 is located at the edge of the side of the table, and far-net sensor 23 is located midway between far-net sensor 12 and the vertical center line 14 of the table surface. The longitudinal distance between the near-net and far-net sensors 9 is no less than 22cm to avoid mutual interference during pressure signal transmission and ensure independent data acquisition. Near-net sensor 7, intermediate sensor 8, and far-net sensor 9 all employ embedded miniature pressure sensors, which are evenly distributed. All sensors are embedded, with their surfaces flush with the table tennis table surface and a protrusion height of ≤0.1mm. This ensures that they do not affect the normal rebound trajectory of the table tennis ball or the trainee's hitting action, enabling interference-free real-time feedback.

[0028] The data processing module includes a data fusion and force quantification unit and a shot type recognition unit. The data fusion and force quantification unit uses a weighted fusion algorithm to calculate and fuse data on racket face pressure peak, grip force peak, and table rebound pressure. Combined with pre-calibrated correction coefficients for different racket types (penhold / shakehand) and different rubber types (inverted / short pips / long pips), it outputs the shot force value (error ≤ 5%), as shown in the following formula: The formula for calculating the initial fusion strength value is as follows: ; in, This represents the initial fusion strength value (unit: N). , , These are the weighting coefficients for racket face, grip, and table surface pressure data, respectively (based on experimental calibration, the default values ​​are 0.6, 0.2, and 0.2, respectively, and can be fine-tuned according to the training scenario). Peak pressure on the racket face (unit: N); Peak force exerted on the grip (unit: N); Peak value of the rebound pressure on the platform (unit: N).

[0029] The initial fusion strength calculation formula has been revised, and the revised precise strength calculation formula is as follows: ; in, The corrected precision force value (unit: N). Correction factor for racket type (penhold racket) Horizontal grip , Correction factor for rubber type (inverted rubber) positive rubber Long rubber (To match the elastic properties of different rubbers).

[0030] The formula for calculating impact momentum (reflecting explosive power characteristics) is as follows: ; in, The impulse of the ball (unit: N·s). Duration of slapping pressure (unit: seconds).

[0031] The shot type recognition unit employs a machine learning-based random forest algorithm, using racket face pressure distribution features, grip force curve features, and the rhythm of shot force changes as input features. It pre-trains recognition models for six common shot types: forehand attack, backhand flick, forehand topspin, and backhand topspin. The core formulas are the feature vector construction formula and the random forest voting decision formula, as follows: Formula for constructing input feature vectors: ; in, The input feature vector is 12-dimensional; The pressure distribution characteristics of the racket face are represented by the horizontal and vertical coordinates of the pressure peak position, the distribution entropy, and the peak concentration, respectively. The force curve characteristics of the grip are represented by the peak force of the thumb, the peak force of the index finger, the time of occurrence of the peak force, and the amplitude of curve fluctuation. The rhythm characteristics of the change in hitting force are represented by the following: the rate of increase in force, the rate of decrease in force, the duration of peak stability, and the variance of force change.

[0032] Random Forest Voting Decision Formula: ; in, The type of shot for the final output; It is a collection of 6 common shot types; The number of decision trees in the random forest (default 100). For the first Each decision tree is used to process the input feature vector. The prediction results; This is an indicator function (when the condition inside the parentheses is true). ,otherwise This involves counting the number of votes for a certain type of shot predicted by all decision trees, and the type with the most votes is the final identification result. When the trainee hits the ball, the chip can complete the calculation and decision of the above formula within 10ms, output the shot type, and match the standard power range of the corresponding shot type, providing a basis for subsequent targeted feedback.

[0033] The multi-module real-time feedback system includes a racket-end LED instant prompt unit, a terminal visual interface unit, and an intelligent voice guidance unit. The racket-end LED instant prompt unit includes green, yellow, and red LEDs located at the end of the racket handle, corresponding to whether the force is sufficient and the shot type is matched, small force deviation, large force deviation, or incorrect shot type, respectively. The LEDs light up instantly upon impact, allowing the trainee to quickly perceive their force application without needing to check the terminal, enabling them to focus on adjusting their movements.

[0034] The terminal's visual interface unit transmits real-time data to a mobile phone or tablet app via Bluetooth. The interface displays real-time data on the impact force, impact momentum, impact point, impact type, and grip force curve, showing the trend of force changes in consecutive shots in a line graph. Simultaneously, the app allows users to preset training goals (e.g., "maintaining forehand attack force at 8-10 N·s"). When three consecutive shots meet the goal, a "goal achievement" notification pops up, enhancing the sense of accomplishment during training.

[0035] The intelligent voice guidance unit provides real-time optimization suggestions via a mobile phone or tablet app (such as "Insufficient power, increase forearm power," "Hit point is too far to the edge, adjust racket face angle," "Grip too tight for backhand flick, relax thumb pressure"). The voice guidance is generated based on preset power standards for different shot types, combined with the trainee's personalized data deviations, making it more targeted.

[0036] This invention also includes a cloud data management module for synchronizing training data to a cloud server, enabling permanent storage of historical data and synchronization across multiple devices. Coaches can view trainees' training data through the cloud platform, generate training reports (such as "Insufficient stability of forehand attack power, forearm power training needs to be strengthened"), and push personalized training plans. Simultaneously, the device can automatically adjust the difficulty of training goals based on the trainee's historical data (e.g., a wider range of target power for beginners, narrowing the range for advanced stages), achieving progressive training.

[0037] Full-counter surface detection logic: When a ping-pong ball bounces, it generates an instantaneous pressure wave. This pressure wave propagates rapidly along the table surface (wood / composite material), and sensors in the corresponding areas can capture key data such as the pressure peak and pressure transmission time. For the core areas covered by sensors 9 near and far from the net, the sensors can directly collect pressure data. For the gaps between sensors, by using the time difference and peak attenuation of pressure signals received by adjacent sensors, combined with a preset table surface pressure transmission characteristic algorithm (based on the pressure wave propagation speed and attenuation coefficient of the table material), the actual bounce position of the ping-pong ball and the pressure magnitude at that position can be calculated, achieving full-surface, blind-spot-free rebound pressure detection. By combining racket face hitting force data with table surface rebound pressure data, the actual effect of hitting force can be verified (e.g., the difference in table surface rebound force caused by different hitting angles and different landing areas under the same racket face force), further improving the accuracy of force quantification, and providing data support for optimizing hitting angle and landing point control.

[0038] The present invention also includes communication components, electronic components, etc. integrated inside the grip bar, all of which adopt existing technologies and will not be described in detail here.

[0039] The device's workflow is as follows: (1) Device initialization: The trainee selects the racket type, rubber type, training mode (basic strength training / specialized hitting training), and presets the training target strength range; (2) Data acquisition: When the trainee hits the ball, the racket face pressure sensor array, the grip pressure sensor module, and the table surface auxiliary sensor unit (near net + middle + far net sensor 9) simultaneously collect pressure data; (3) Data processing: The embedded chip filters and fuses the collected data to output the hitting force value, hitting impulse, and hitting point position. At the same time, it identifies the hitting type through a machine learning model. (4) Real-time feedback: LED lights at the racket end provide real-time indication of the force application status, the APP interface displays real-time data, and voice outputs optimization suggestions; (5) Data synchronization and analysis: Training data is synchronized to the cloud, the device generates training progress analysis, and the coach can view the data and adjust the training plan.

[0040] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A table tennis hitting force quantification training device based on pressure sensing, characterized in that, include: The multi-dimensional pressure sensing fusion module adopts a three-in-one sensing layout of ping-pong paddle, ping-pong paddle handle, and ping-pong table surface to achieve comprehensive collection and accurate quantification of force data. The data processing module, equipped with an embedded chip, integrates multi-sensor data fusion algorithms and machine learning models to achieve real-time data processing and intelligent analysis. The multi-module real-time feedback module uses a multi-modal feedback approach, including real-time prompts at the racket end, terminal visualization, and voice guidance, to allow trainees to perceive force deviations in real time and quickly adjust their movements. The multi-dimensional pressure sensing fusion module includes: The racket pressure sensing unit is used to monitor the finger pressure of the trainee's penhold grip and shakehand grip, as well as the pressure at the point of impact on the racket face; The grip pressure sensing unit is used to monitor the finger pressure of the trainee's penhold grip and shakehand grip, as well as changes in grip strength. The table tennis table surface pressure sensing unit adopts a distributed layout near the net, in the middle, and far from the net to monitor the rebound pressure of the ping-pong ball after landing. The data processing module includes a data fusion and force quantification unit and a shot type recognition unit; The multi-module real-time feedback module includes an LED instant prompt unit at the racket end, a terminal visual interface unit, and an intelligent voice guidance unit.

2. The table tennis hitting force quantification training device based on pressure sensing according to claim 1, characterized in that: The racket pressure sensing unit includes a racket face hitting pressure sensing area and a finger pressure sensing area, with the finger pressure sensing area located at the end of the racket face rubber near the handle. The racket face impact pressure sensing area is used to monitor and collect the pressure distribution, pressure peak, and pressure duration at the moment of impact. The impact point is identified by the pressure distribution characteristics. At the same time, the racket face impact pressure sensing area monitors the pressure on the index and ring fingers of the trainee's penhold grip. The finger pressure sensing area is used to monitor the pressure of the index finger and thumb in the shakehand grip, and the pressure of the index finger, thumb, and little finger in the penhold grip.

3. The table tennis hitting force quantification training device based on pressure sensing according to claim 2, characterized in that: The racket pressure sensing unit uses a flexible pressure sensing array that is adapted to the curved shape of the table tennis racket rubber. The flexible pressure sensing array is embedded between the table tennis racket rubber and the racket blade, and includes several piezoelectric thin film sensors.

4. The table tennis hitting force quantification training device based on pressure sensing according to claim 2, characterized in that: The grip pressure sensing unit includes a racket blade pressure sensing area and a grip shaft pressure sensing area. The racket blade pressure sensing area is located in the blade area between the racket rubber and the grip shaft. The racket blade pressure sensing area is used to monitor the pressure of the trainee's index finger in the shakehand grip and the pressure of the trainee's thumb and index finger in the penhold grip. The pressure sensing area of ​​the grip bar is used to monitor the trainee's hand grip strength and the pressure of the trainee's middle finger, ring finger, and little finger when using a horizontal grip. The racket blade pressure sensing area is evenly distributed with several pressure monitoring points, each pressure monitoring point is embedded with a pressure sensor, and the grip shaft pressure sensing area is embedded with several sets of ring pressure sensors. The racket pressure sensing unit and the grip pressure sensing unit jointly collect the changes in the grip force of the trainee when hitting the ball.

5. The table tennis hitting force quantification training device based on pressure sensing according to claim 4, characterized in that: The tabletop pressure sensing unit includes a near-net sensing group, a middle sensor, and a far-net sensing group, all of which are symmetrically arranged on both sides of the middle net on the tabletop. The near-net sensor group includes near-net sensors symmetrically arranged about the vertical center line of the table surface to monitor the pressure in the near-net bounce area. The near-net sensors include near-net sensor one and near-net sensor two, which are located on the same horizontal line of the table surface near the net. Near-net sensor one is located at the edge of the side of the table surface, and near-net sensor two is located in the middle between near-net sensor one and the vertical center line of the table surface. The intermediate sensor is located between the near-net sensor group and the far-net sensor group, and is used to monitor the pressure of the ball rebounding in the middle area on one side of the table surface; The far-net sensor group includes far-net sensors symmetrically arranged about the vertical center line of the table surface to monitor the pressure in the far-net ball bounce area. The far-net sensors include far-net sensor one and far-net sensor two, which are located on the same horizontal line of the table surface near the end of the table. Far-net sensor one is located at the edge of the side of the table surface, and far-net sensor two is located in the middle between far-net sensor one and the vertical center line of the table surface.

6. The table tennis hitting force quantification training device based on pressure sensing according to claim 5, characterized in that: Near-field, intermediate, and far-field sensors all employ embedded miniature pressure sensors that are evenly distributed.

7. The table tennis hitting force quantification training device based on pressure sensing according to claim 1, characterized in that: The data fusion and force quantification unit uses a weighted fusion algorithm to fuse and calculate the peak racket face pressure, peak grip force, and table surface rebound pressure data. Combined with pre-calibrated correction coefficients for different racket grip types and rubber types, it outputs the hitting force value, as shown in the following formula: The formula for calculating the initial fusion strength value is as follows: ; in, This is the initial fusion strength value. , , These are the weighting coefficients for racket face, grip, and table surface pressure data, respectively. This represents the peak pressure on the racket face. This represents the peak force exerted by the grip. This represents the peak of the rebound pressure on the platform. The initial fusion strength calculation formula has been revised, and the revised precise strength calculation formula is as follows: ; in, To correct the precise force value, This is a correction factor for racket type. This is a correction factor for the rubber type. The formula for calculating the impulse of a shot is: ; in, To generate momentum for the shot, The duration of the pressure applied to the slapping surface.

8. The table tennis hitting force quantification training device based on pressure sensing according to claim 7, characterized in that: The shot type recognition unit employs a machine learning-based random forest algorithm, using racket face pressure distribution features, grip force curve features, and the rhythm of shot force changes as input features. It pre-trains recognition models for forehand attack, backhand flick, forehand topspin, and backhand topspin. The core formulas are the feature vector construction formula and the random forest voting decision formula, as follows: Formula for constructing input feature vectors: ； in, The input feature vector is 12-dimensional; The pressure distribution characteristics of the racket face are represented by the x-axis, y-axis, distribution entropy, and peak concentration of the pressure peak position, respectively. The force curve characteristics of the grip are represented by the peak force of the thumb, the peak force of the index finger, the time of occurrence of the peak force, and the amplitude of curve fluctuation. The rhythm characteristics of the change in hitting force are represented by the rate of increase in force, the rate of decrease in force, the duration of peak stability, and the variance of force change, respectively. Random Forest Voting Decision Formula: ； in, The type of shot for the final output; It is a collection of 6 common shot types; The number of decision trees in the random forest; For the first Each decision tree is used to process the input feature vector. The prediction results; The indicator function is to count the number of votes for a certain type of shot predicted by all decision trees, and the type with the most votes is the final identification result.

9. The table tennis hitting force quantification training device based on pressure sensing according to claim 1, characterized in that: The racket end LED real-time prompt unit includes a green LED light, a yellow LED light, and a red LED light located at the end of the racket handle, which correspond to the force being up to standard and the hitting type being matched, the force deviation being small, the force deviation being large, or the hitting type being incorrect, respectively. The terminal visualization interface unit transmits real-time data to a mobile phone or tablet APP via Bluetooth. The interface displays the hitting force value, hitting impulse, hitting point position, hitting type, and grip force curve data in real time, and shows the trend of continuous hitting force changes in the form of a line graph. The intelligent voice guidance unit outputs optimization suggestions in real time via voice through a mobile phone or tablet app.

10. The table tennis hitting force quantification training device based on pressure sensing according to claim 1, characterized in that: It also includes a cloud data management module, which is used to synchronize training data to the cloud server to achieve permanent storage of historical data and synchronization across multiple devices.

Images