A sports training system based on sports sensation

Abstract

The present invention discloses a sports training system based on sports sensation, comprising a storage unit, an electromyographic signal acquisition unit, a data analysis unit and a display unit, wherein: the storage unit has a plurality of preset limb movement schemes built in, and these schemes can be demonstrated by the display unit; the electromyographic signal acquisition unit is used to collect the electromyographic signals of the trainee when completing the movement according to the demonstrated movements, and send them to the data analysis unit; the data analysis unit is used to receive the electromyographic signals and analyze them; the storage unit can also be used to store the electromyographic signals and the analysis results thereof; the data analysis unit can generate a virtual limb movement trajectory according to the trainee's electromyographic signals, and display it through the display unit, and judge the training quality by comparing the preset movement scheme with the virtual limb movement trajectory generated by the electromyographic signals.

Description

Technical Field

[0001] The present invention relates to a training system for improving motor function, and in particular to a sports training system based on sports sensation. Background Art

[0002] Various neurological diseases and aging can lead to motor dysfunction. Under the control of the nervous system, the body's skeletal muscles can contract to complete coordinated limb (body) movements. Normal movement control requires a reasonable time-space relationship for muscle contraction. Errors in the time-space relationship of muscle contraction manifest as stiff, uneven, and inflexible movements, and even stereotyped and coarse pathological movement patterns. A reasonable force application method controlled by the nervous system is the prerequisite for the correct time-space relationship of muscle contraction.

[0003] Surface electromyography (EMG) signals can reflect muscle contraction to a certain extent and are an effective and commonly used indicator for assessing muscle contraction status. In numerous studies, EMG signals have been used to noninvasively assess the temporal-spatial relationship characteristics of muscle contraction. Surface electromyography signals also provide immediate feedback and are commonly used in muscle movement control training.

[0004] However, there are two problems with existing limb motor function training based on surface electromyography signals. First, most of them only use the amplitude of the electromyography to indicate the intensity of the myoelectric contraction, but there is a lack of reliable indicators to judge the quality of the movement. Since there is not always a strict linear correspondence between the amplitude of the surface electromyography signal and muscle contraction, it is unreliable to rely solely on the amplitude of the electromyography wave to judge the intensity of muscle contraction. Second, it ignores the motor-sensory integration characteristics of human movement control. The motor learning process of normal people requires sensory information to provide feedforward and feedback information. If the electromyography-guided biofeedback system only relies on the intensity of the electromyography signal and ignores the sensory information of the body, it may not be conducive to the improvement of motor function. Summary of the Invention

[0005] The applicant has found that since various neurological diseases and aging can impair sensory input, emphasizing the use of sensory information during exercise training can promote motor-sensory integration, which may improve the trainee's motor function and promote the recovery of their sensory function at the same time.

[0006] To this end, the present invention proposes a surface electromyography (sEMG) exercise training system based on motor sensation, which can set a variety of exercise difficulties through demonstration trajectories. As the trainee moves their limbs according to the demonstrated trajectory, the changes in the surface electromyography signals of the agonist muscles of the limb are collected to generate the movement of the virtual limb displayed on the screen. The quality index is obtained by calculating the characteristics of this movement trajectory and the demonstration trajectory. In this process, the trainee needs to experience the feeling of their own limbs during exercise, thereby establishing a motor-sensory integrated force application method. By diligently trying to improve the quality index, their own motor control ability is improved.

[0007] In order to achieve the purpose of the present invention, the following technical solutions are adopted:

[0008] A sports training system based on motor sensation includes an electromyographic signal acquisition unit, a data analysis unit, a storage unit and a display unit, wherein: the storage unit has a plurality of preset limb movement schemes built in, and these schemes can be demonstrated by the display unit; the electromyographic signal acquisition unit is used to collect the electromyographic signals of the trainee when completing the movement according to the demonstrated movements, and send them to the data analysis unit; the data analysis unit is used to receive and analyze the electromyographic signals; the storage unit can also be used to store the electromyographic signals and the analysis results; the data analysis unit can generate a virtual limb movement trajectory based on the trainee's electromyographic signals and display it through the display unit, and can also judge the training quality by comparing the preset movement scheme with the virtual limb movement trajectory generated by the electromyographic signals.

[0009] The aforementioned motion training system, wherein: the system presets multiple demonstration trajectories D for training. For a specified limb, a training program (DT) demonstrates the limb trajectory with an animation on the screen, and its content is determined by the following parameters: the axis of limb movement, i.e., the static point; the active part other than the static point; the starting point, end point, and trajectory of the active part of the limb movement, the pause point, and the pause time; the speed of movement;

[0010]

[0011] In the first row, D is the demonstration trajectory of the limb; DT is the demonstration plan; P s is the axis of limb movement, that is, the coordinate position of the limb static point, VAL1 is P s The value of P m is a set of coordinate values ​​of the starting position of the active part outside the static point, VAL2 is P m The value of P a 1 is in P m The set of coordinate positions when the entire movement reaches the preset end point, VAL3 is P a The value of 1; P a 2 is the moving speed of the active part during exercise, VAL4 is Pa The value of 2; P a 3 is the coordinate set of the active part when it pauses in the motion trajectory, VAL5 is P a The value of 3; P a 4 is the length of time the active part pauses in the motion trajectory, VAL6 is P a The second and third rows are the conditions that VAL4 should meet (one of the two conditions is met), where r is the length between the farthest moving point of the limb and the static point, v is the velocity of the farthest moving point of the limb relative to the vertical direction of the line connecting the static point and this moving point, t is time, ω is the angular velocity of the line connecting the farthest moving point of the limb and the static point with the static point as the center, and α is the corresponding angular acceleration.

[0012] The sports training system, wherein: the data analysis unit is used to convert the original value R of the electromyographic signal into A:

[0013]

[0014] Where n is the number of original values ​​of the electromyographic signal sampled within a period of time.

[0015] The sports training system, wherein: the speed and direction of the change of the limb movement trajectory are determined by the change amplitude of the electromyographic signal per unit time:

[0016]

[0017] This means that when A meets the conditions in the first or second row, the speed is determined by the third row, and the direction is determined by the fourth row. If A does not meet the conditions in the first or second row, the speed is 0.

[0018] Among them, R is the original value of the electromyographic signal; A is the converted electromyographic signal value; n is the number of original values ​​of the electromyographic signal sampled within a period of time (such as 50ms); j is the number of time periods of continuous sampling in the relaxed state before exerting force (for example, maintaining relaxation for 5s, divided into 10 segments), from 1 to j, each segment corresponds to an A value, from A1 to A j ; A1 to A j Note that A1 to A in the first and second rows j ,as well as and R are the data values ​​in this relaxed state, while A is the data value in the force state. J in the third row is the number of consecutive sampling time periods after the limb starts to exert force, ranging from 1 to J; Δt is the time length of the electromyographic signal change. In the fourth row, θ is a built-in coefficient. When the value in the brackets is positive, the limb in the animation moves toward the end of the preset trajectory; when it is zero, the limb stops moving; when it is a negative value, the limb moves toward the starting point of the trajectory. A in the third and fourth rows i+1 and A i Represents the data values ​​of two adjacent time periods under the force state.

[0019] In the aforementioned sports training system, after the trainee completes an action, the degree of conformity M between the demonstration trajectory D and the limb motion trajectory L is used as a quantitative indicator of motion control ability. In a given time period P (assuming there are 1…p time points in total), the quality index M between the demonstration trajectory D and the myoelectrically driven limb motion trajectory L′ is p for

[0020]

[0021] Where D is the demonstration trajectory preset by the system (where the coordinates of each point in the limb are x (1-n) y (1-m) ); L is the virtual limb trajectory driven by the electromyographic signal, where the coordinates of each point in the limb are X (1-n) Y (1-m) ), considering that the starting positions of D and L are different, the coordinates XY of L are corrected to X'Y', which is the same as the starting position of D. P is a given time period (with 1…p time points). μ is the mean.

[0022] The sports training system, wherein: the quality index M p , or it can be calculated using the following method:

[0023]

[0024] in

[0025]

[0026] Where v is continuous LTL p The number of values, G is the parameter to be solved, q is the order of the polynomial used for solution, H is the noise, and the subscripts of the parameters in the first row represent their respective dimensions, such as This parameter has 2v+1 rows and q columns. n is a positive integer starting from 1. The fourth row is P(LTL′ p ) is the fitting solution of LTL′ p The estimated value of . B is the auxiliary matrix set for solution. The superscript T in the fifth row indicates the matrix transpose. From f(LTL′p ) Solve for LTL′ p After calculating the value of M P The value of .

[0027] The length of the demonstration trajectory is DTL, and the length of the myoelectrically driven limb motion trajectory is LTL. P is a given time period (1…p time points). μ is the mean. Relative to the static point, at time point p, the length of the moving point in the demonstration trajectory D is DTL. p , the length of the limb moving in the electromyographically driven trajectory L is LTL p .

[0028] The sports training system, wherein: the difficulty C of the subsequent training program can be selected according to the quality of the user's completion of a training program:

[0029]

[0030] Where n is the number of repetitions of the training program (e.g. 30 times). The challenge C is given by the number of repetitions n of the training program, the number of M in each training p value, M calculated with different window widths when a given training scheme is repeated multiple times p The slope of the derived value is determined by multiple parameters: For example, given a training program repeated 30 times, when most of the M p When the value is greater than 50% and the overall trend is upward, or the Slope value is greater than 0.6, C = 1; when most M p (D p ,L p ) value is less than 50, or when the overall trend is downward, or when the Slope value is <0.6, C = -1; in all other cases, C = 0. The Slope value eliminates the drift effect of MP values ​​within the same training plan, considering only the autocorrelation of MP value trends. Therefore, the uniqueness of this method lies in determining the difficulty C of subsequent training plans based on both the intuitive drift trend of MP values ​​and the autocorrelation of MP value trends. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 is a system block diagram of the present invention;

[0032] Attachment Figure 2 It is a system workflow diagram of the present invention. DETAILED DESCRIPTION

[0033] The following is combined with Figure 1-2, the specific embodiments of the present invention are described in detail. The embodiments are exemplary and are only used to explain the present invention, and should not be understood as limiting the present invention. Obviously, the embodiments described in the present invention are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative work shall fall within the scope of protection of the present invention.

[0034] References to "one embodiment" or "some embodiments" in this specification mean that one or more embodiments of the present invention include the specific features, structures, or characteristics described in conjunction with that embodiment. Thus, the terms "including," "comprising," "having," and their variations throughout this specification mean "including but not limited to," unless otherwise specifically stated.

[0035] like Figure 1 、 2 As shown, the sEMG (surface electromyography) sports training system based on sports sensation of the present invention includes: an electromyography signal acquisition unit, a data analysis unit, a storage unit and a display unit.

[0036] By fixing the surface electrodes on the surface of the active muscles of the trainee's training limb, the electromyographic signal acquisition unit can collect the electromyographic signals of the trainee during exercise and send them to the communication unit (such as a Bluetooth receiving unit) of the data analysis unit through the communication unit (such as a Bluetooth transmission unit). The Bluetooth transmission unit of the data analysis unit receives the electromyographic signals, converts the electromyographic signals into digital signals through the analog-to-digital sensor ADC, and sends them to the data analysis unit, which receives and analyzes them. The storage unit is used to store the electromyographic signals and the analysis results. The data analysis unit is also used to form the trainee's limb movement trajectory according to the analysis results and display it through the display unit.

[0037] The data analysis unit drives the movement of the virtual limb displayed on the screen based on the received digitized EMG signals and calculates the consistency of the limb (torso) movement driven by these EMG signals with the demonstration trajectory used, which is known as the "quality index." The "quality index" reflects the body's motor control ability and provides feedback to the user. Based on this feedback, the user experiences the movement sensation and adjusts their movement style, forming a closed-loop mechanism of "practice-feedback-re-practice."

[0038] This system uses screen animation displayed by the display unit to show the limb (torso) movement of the upper or lower limbs. The user follows the limb (torso) movement (direction, speed, amplitude) displayed by the screen animation to complete the actual limb (torso) movement.

[0039] Fluctuations in the Quality Index indicate changes in the trainee's ability to control their electromyographic signals. A higher Quality Index indicates improved force application, leading to better motor control. Therefore, this system can train the user's force application. Trainees should understand their own force application and continuously adjust it based on feedback from the Quality Index. This enhances the integration between proprioception and motor output during exercise.

[0040] The limb (torso) trajectories demonstrated in the animation include: left (or right) shoulder girdle extension, shoulder girdle retraction, shoulder girdle elevation, shoulder girdle descent, shoulder flexion, shoulder extension, shoulder adduction, shoulder abduction, horizontal shoulder adduction, horizontal shoulder abduction, shoulder internal rotation, shoulder external rotation, elbow flexion, elbow extension, wrist flexion, joint extension, radial wrist deviation, ulnar wrist deviation, finger group flexion, finger group extension, single finger flexion, single finger extension, opposition, thumb radial abduction, thumb palmar abduction, hip extension, hip flexion, knee extension, knee flexion, ankle dorsiflexion, ankle plantar flexion, ankle inversion, ankle eversion, trunk flexion, trunk extension, etc.

[0041] This system provides dual-track displays for upper limb, lower limb, or other body movements: (1) a demonstration track of limb (body) movement, displayed as an animation (e.g., displayed in blue). The trainee moves their limbs following this preset track displayed on the screen; (2) an EMG-driven limb movement track (e.g., displayed in red). Surface EMG signals collected from the trainee's muscles drive another limb movement track, namely, an EMG-driven virtual movement track. The degree of matching between the two tracks reflects the trainee's ability to control their own limb movements.

[0042] The direction and speed of the virtual motion trajectory driven by myoelectric signals are driven by the collected myoelectric signals of the trainee. For example, within two adjacent fixed time windows (such as 100ms), if the amplitude value in the latter time window is higher than the amplitude value in the previous time window, the limb movement displayed on the screen animation will follow the direction of the demonstration trajectory; if the amplitude values ​​are equal, the limb will be stationary; if the amplitude value decreases, the limb will move towards the starting position.

[0043] The EMG signal value A used here is converted from the original EMG signal value R (calculation formula 1):

[0044]

[0045] Where n is the number of original values ​​of the electromyographic signal sampled within a period of time.

[0046] The speed and direction of the virtual limb motion trajectory are determined by the amplitude of the myoelectric signal change per unit time (Equation 2):

[0047]

[0048] This means that when A meets the conditions in the first or second row, the speed is determined by the third row, and the direction is determined by the fourth row. If A does not meet the conditions in the first or second row, the speed is 0.

[0049] Among them, R is the original value of the electromyographic signal; A is the converted electromyographic signal value; n is the number of original values ​​of the electromyographic signal sampled within a preset period of time (such as 50ms); j is the number of time periods of continuous sampling in the relaxed state before exerting force (for example, maintaining relaxation for 5s, divided into 10 segments), from 1 to j, each segment corresponds to an A value, from A1 to A j ; A1 to A j In particular, A1 to A in the first and second rows j ,as well as and R are the data values ​​in this relaxed state, while A is the data value in the force state. J in the third row is the number of consecutive sampling time periods after the limb starts to exert force, ranging from 1 to J; Δt is the time length of the electromyographic signal change. In the fourth row, θ is a built-in coefficient. When the value in the brackets is positive, the limb in the animation moves toward the end of the preset trajectory; when it is zero, the limb stops moving; when it is a negative value, the limb moves toward the starting point of the trajectory. A in the third and fourth rows i+1 and A i are the data values ​​of two adjacent periods under the force state.

[0050] The preset demonstration trajectory includes a variety of movement modes: slow, smooth and uniform movement (for example, from knee flexion 90 degrees to extension 180 degrees, it takes 4 seconds, and the speed is uniform within this range of joint motion), slow and smooth variable speed movement (for example, from knee flexion 90 degrees to extension 180 degrees, it takes 4 seconds, 1-2 seconds for acceleration period, the speed gradually increases, 3-4 seconds for deceleration period, the speed gradually slows down), pause movement (for example, from knee flexion 90 degrees to extension 180 degrees, it takes 5 seconds, including 500ms pauses at extension to 120 degrees and 150 degrees respectively), and fast movement (for example, from knee flexion 90 degrees to extension 180 degrees, it takes 1 second to complete).

[0051] The degree of match between the demonstration trajectory and the myoelectrically driven limb motion trajectory, that is, the degree of match between the spatial positions of the two during the entire motion process. If the starting positions of the two completely overlap, then if the two motion trajectories overlap, then the two overlap at any point in time. If the starting positions are different, then when the starting positions are corrected to be the same (the starting position of the limb motion trajectory is translated to the starting position of the demonstration trajectory), the two trajectories overlap. Taking the knee joint starting from 90 degrees of flexion and extending to 180 degrees of knee extension as an example, this movement is a single-joint single-degree-of-freedom movement. Both the demonstration trajectory (D) and the myoelectrically driven limb motion trajectory (L) change with time in the human sagittal plane space. During the time period when the entire motion trajectory is completed, the quality index M(D, L) of the myoelectrically driven limb motion trajectory is (calculation formula 3):

[0052]

[0053] Where i is the i-th time point within the time period. μ is the mean value of L or D. Here, the trajectory of the calf in the human sagittal plane is used as an example. Both D and L are planar projections of the calf and foot, consisting of a matrix of points in the human sagittal plane, rather than a single value.

[0054] The horizontal coordinate of the sagittal plane is defined as x (from 1 to n) and the vertical coordinate is y (from 1 to m), where n is the maximum horizontal coordinate and m is the maximum vertical coordinate. The coordinates of all points in the sagittal plane space are the following set:

[0055]

[0056] At time point i, for any point in the above coordinate set, the demonstration trajectory D takes a value of 1 where it is covered and a value of 0 where it is not covered (Calculation Formula 4):

[0057]

[0058] The first row of brackets in the brackets represents the set of coordinates, and the following three rows represent the conditions for the values ​​corresponding to the coordinates. For any training scheme, the time-varying D shown in (Equation 4) i Value is a set containing 0,1 values.

[0059] Similarly, at time point i, at any point in the coordinate set of the projected two-dimensional plane, the value of the myoelectrically driven limb motion trajectory L is 1 where it is covered, and 0 where it is not covered (Equation 5):

[0060]

[0061] Wherein, the horizontal axis is X (from 1 to n) and the vertical axis is Y (from 1 to m). The time-varying L shown in (Calculation Formula 5)i Value is also a set containing 0,1 values.

[0062] According to the system default, D i and L i The corresponding elements in these two sets have the same shape and size in the two-dimensional plane of the human body sagittal plane (i.e., for example, for the calf, whether it is demonstration or myoelectric drive, the shape and size of the two calves are the same). Their starting positions can be set to overlap or not overlap; if not overlap, D i and L i The starting position of is a (non-rotational) translation relationship. That is, for D i The starting position and L i The starting position of all corresponding elements x a y b and X a′ Y b′ The relationship between the two is X a′ =x a +h,Y b′ =y b +j,(h and j are x a y b and X a′ Y b′ The distance on the x-axis and y-axis, so 0≤|h| <n,0≤|j|<m)。

[0063] When D i and L i If the starting positions do not coincide, if you need to i The position is first corrected to L' i , so that the starting positions of the two trajectories coincide, then (Formula 6):

[0064]

[0065] X′1=X1-h,…X′ n =X n -h

[0066] Y1′=Y1-j,…Y′ m =Y m -j

[0067] The system includes multiple training demonstration scenarios D. For a given limb, a training scenario (DT) uses an on-screen animation to demonstrate the limb's trajectory D. Its content is determined by the following parameters: the axis of the limb's movement, or the stationary point; the active part of the limb, excluding the stationary point; the starting point, end point, and trajectory of the active part of the limb's movement; the pause points and pause duration; and the speed of movement. The difficulty of the training task can be varied by changing the location and number of pause points, pause duration, and movement speed. A training scenario is determined by (Equation 7):

[0068]

[0069] In the first row, D is the demonstration trajectory of the limb; DT is the demonstration plan (training plan); P s is the axis of limb movement, that is, the coordinate position of the limb static point, VAL1 is P s The value of P m is a set of coordinate values ​​of the starting position of the active part outside the static point, VAL2 is P m The value of P a 1 is in P m The set of coordinate positions when the entire movement reaches the preset end point, VAL3 is P a The value of 1; P a 2 is the moving speed of the active part during exercise, VAL4 is P a The value of 2; P a 3 is the coordinate set of the active part when it pauses in the motion trajectory, VAL5 is P a The value of 3; P a 4 is the length of time the active part pauses in the motion trajectory, VAL6 is P a The second and third rows are the conditions that VAL4 should meet (one of the two conditions is met), where r is the length between the distal moving point and the static point of the limb, v is the velocity of the distal moving point relative to the static point and the line connecting the distal moving point in the direction perpendicular to the moving point, t is time, ω is the angular velocity of the line connecting the distal moving point and the static point with the static point as the center, and α is the corresponding angular acceleration.

[0070] After the trainee completes an action, the demonstration trajectory D and the limb movement trajectory L may match completely (direction and speed are exactly the same), or there may be differences. The degree of matching between the two is measured by the quality index M. p Quantify. In a given time period P (assuming there are 1…p time points), the quality index M between the demonstration trajectory D and the myoelectrically driven limb motion trajectory L is p =(Formula 8):

[0071]

[0072] Where D is the demonstration trajectory preset by the system (where the coordinates of each point in the limb are x (1-n) y (1-m) ); L is the virtual limb trajectory driven by the electromyographic signal, where the coordinates of each point in the limb are X (1-n) Y (1-m) ), considering D i The starting position and L i The starting positions are different, and the coordinates XY of L are corrected to X'Y' according to (Formula 6), that is, the starting positions are the same, and P is a given time period (with 1...p time points).

[0073] In (Calculation Formula 8), The subscript p means the coordinate value corresponding to the time point p.

[0074] In (Calculation Formula 8), The meaning of is the mean of this set in a given time period P (where there are 1…p time points). Similarly, Calculated using the same method.

[0075] For a given time point i, D i and L i (or L' i ) two sets, the following transformations are possible for Equation 8:

[0076] Taking the example of knee extension achieved through quadriceps training, the calf and foot are considered as a rigid body (assuming that the ankle joint angle is constant, that is, ignoring the plantar flexion, dorsiflexion, and internal and external eversion of the ankle joint). When the knee joint is extended or flexed, the knee joint is the axis and is considered as a static point, and the coordinates are constant. The corresponding point of the ankle joint on the sagittal plane of the human body is considered as a moving point, and its coordinates at time point i1 are x i1 y i1 , the coordinate at time point i2 is x i2 y i2 , and so on, at time point i k The coordinate at time is x ik y ik At this time, the value of the demonstration trajectory D in (Calculation Formula 4) is equivalent to the value corresponding to this moving point. In a time period t1-t k The moving point trajectory D p , then (Formula 9):

[0077] D p =[x t1 y t1 … x tk y tk ]

[0078] Similarly, the value of the limb motion trajectory L driven by electromyography in (Equation 5) is also equivalent to the value corresponding to its moving point, so in a time period t1-t k The moving point trajectory between i and L i When the starting positions completely overlap, it is (Calculation Formula 10):

[0079] L p =[X t1 Y t1 … X tk Y tk ]

[0080] Such as D i and L i When the starting positions do not coincide, correction is performed according to (Formula 6).

[0081] The length of the demonstration trajectory is DTL, and the length of the myoelectrically driven limb motion trajectory is LTL. P is a given time period (with 1…p time points). μ is the mean. Relative to the static point, at time point p, the length of the moving point in the demonstration trajectory D is DTL p , the length of the limb moving in the electromyographically driven trajectory L is LTL p . Take v LTLs in a row p The value is processed by polynomial to remove noise interference according to its adjacent values ​​and converted into LTL p′ Value (Calculation Formula 11):

[0082]

[0083] Where v is continuous LTL p The number of values, G is the parameter to be solved, q is the order of the polynomial used for solution, H is the noise, and the subscripts of the parameters in the first row represent their respective dimensions, such as This parameter has 2v+1 rows and q columns. n is a positive integer starting from 1. The fourth row is P(LTL′ p ) is the fitting solution of LTL′ p The estimated value of . B is the auxiliary matrix set for solution. The superscript T in the fifth row indicates the matrix transpose. From f(LTL′ p ) Solve for LTL′ p After calculating the value of M P The value of (Calculation 12):

[0084]

[0085] Due to differences in user age, gender, physical condition, and physical status (e.g., healthy versus suffering from a disease such as stroke or sarcopenia), the system sets the time period P (1…p time points) to be adjustable. When a user possesses rapid motor and cognitive response capabilities, P is adjusted to a shorter interval; otherwise, P is adjusted to a longer interval.

[0086] After selecting a training program for the user, if it is used multiple times, M p When the value gradually increases and approaches 100%, the system recommends demonstration trajectories with higher difficulty. p If the value does not show an increasing trend or the increasing trend is very small, the system recommends a demonstration trajectory with lower difficulty. That is, the challenge C of the subsequent training program is determined by the M after multiple trainings. p The trend of value change is determined by (Formula 13):

[0087]

[0088] Where n is the number of repetitions of the training program (e.g. 30 times). The challenge C is given by the number of repetitions n of the training program, the M of each training at this repetition number n, and the p The trend of value change, M calculated with different window widths when a given training scheme is repeated multiple times p The slope of the derived value is determined by multiple parameters: For example, given a training program repeated 30 times, when most of the M p When the value is greater than 50% and the overall trend is upward, or the Slope value is greater than 0.6, C = 1; when most M p (D p ,L p ) value is less than 50, or the overall trend is downward, or the Slope value is <0.6, C = -1; in other cases, C = 0.

[0089] When C=1, the difficulty of the demonstration track (D) increases by one level. When C=-1, the difficulty of the demonstration track (D) decreases by one level. When C=0, the difficulty of the demonstration track (D) remains unchanged.

[0090] The slope value is calculated as follows:

[0091] For a given training program, the trainee completes M times (for example, 30 times) continuously, and then obtains M segments of electromyographic signals. For each segment of the signal, calculate M with a fixed window width such as 500ms. p value, then get M p Time series of values. p The time series of values ​​are spliced ​​into a whole time series according to the order of the training time, and the sequence contains N Mp Value. p The time series of values ​​is divided into A subsets, the length of each subset a is n (n = N / A), and the mean is mean a , then for all A subsets, there are A mean a value.

[0092] For any subset a among A subsets, each M p Value (for example, the kth M p value) relative to the mean of the subset a a , its cumulative deviation y k =(Equation 14):

[0093] for k=1,2,…n

[0094] For this subset a, its M p The fluctuation range of the value is (Calculation Formula 15):

[0095]

[0096] For this subset a, its M p The standard deviation of the values ​​is (calculated in Equation 16):

[0097]

[0098] Then, for this subset a, according to S a The cumulative deviation of the recalibrated value is (calculated

[0099] Formula 17):

[0100]

[0101] Then, for all A subsets, there are A recalibrated cumulative deviations Value, the mean of the cumulative deviations of these A recalibrations A =(Equation 18):

[0102]

[0103] When n is gradually increased (i.e., the length of each subset is increased), the above y is calculated again k The mean of the cumulative deviations of the value and A recalibrated values A , we will eventually get the mean of the cumulative deviation recalibrated under different subset lengths n A For each n value and its corresponding mean AThe values ​​are all taken as logarithms with base 10, and two corresponding logarithmic series can be obtained. The linear regression relationship between these two logarithmic series is (Calculation Formula 19):

[0104] lgmean A =Slope×lgn+b

[0105] Where Slope is the slope and b is the intercept.

[0106] The Slope value removes the M when using the same training scheme. p The drift effect of the value is considered only when M p Therefore, the uniqueness of (Equation 13) is p The intuitive drift trend of the value and M p The difficulty of the subsequent training plan is set based on two aspects: the trend of value change and the autocorrelation.

[0107] Through the present invention, trainees can use the surface electromyography signals of the active muscles to drive the movement of virtual limbs in the screen animation when practicing with reference to the demonstration movements. By comparing the consistency of this movement with the demonstration movements, they can quantitatively judge their own movement control ability, improve the way of using force, and thus enhance motor function.

[0108] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any form. Although the present invention has been disclosed as above in terms of preferred embodiments, they are not intended to limit the present invention. Any person skilled in the art can make some changes or modifications to equivalent embodiments using the technical contents disclosed above without departing from the scope of the technical solution of the present invention. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A sports training system based on sports sensation, comprising an electromyographic signal acquisition unit, a data analysis unit, a storage unit and a display unit, characterized in that: The electromyographic signal acquisition unit is used to collect the electromyographic signals of the trainee during exercise and send them to the data analysis unit; The data analysis unit is used to receive and analyze the electromyographic signal; The storage unit is used to store electromyographic signals and analysis results; The data analysis unit is also used to generate the trainee's limb movement trajectory according to the analysis result and display it through the display unit; the sports training system, wherein: the data analysis unit is used to convert the original value R of the electromyographic signal into A: Where, R is the original value of the electromyographic signal; A is the converted electromyographic signal value; n is the number of original electromyographic signal values ​​sampled within a period of time; The sports training system, wherein: the speed and direction of the change of the limb movement trajectory are determined by the change amplitude of the electromyographic signal per unit time: otherwise Speed=0 This means that when A' meets the conditions in the first or second row, the speed is determined by the third row, and the direction is determined by the fourth row. If A' does not meet the conditions in the first or second row, the speed is 0. Among them, j is the number of consecutive sampling time periods in the relaxed state before exerting force, from 1 to j, each section corresponds to an A value, from A1 to A j ; A1 to A j The mean of the first and second rows A1 to A j ,as well as and R are the data values ​​in this relaxed state, while A' is the data value in the force state; J in the third row is the number of consecutive sampling time periods after the limb starts to exert force, from 1 to J; Δt is the time length of the electromyographic signal change; in the fourth row, θ is the built-in coefficient. When the value in the brackets is positive, the limb in the animation moves toward the end of the preset trajectory; when it is zero, the limb stops moving; when it is negative, the limb moves toward the starting point of the trajectory. i+1 and A i are the data values ​​of two adjacent periods under the force state.

2. The sports training system according to claim 1, wherein: The speed and direction of changes in limb movement trajectory are determined by the amplitude of changes in the electromyographic signal per unit time.

3. The sports training system according to claim 1, wherein: The system has preset multiple training plans. For a specified limb, the content of a training plan includes the following parameters: the axis of limb movement, that is, the static point; the active part other than the static point; the starting point, end point and trajectory of the active part of the limb movement, the pause point and pause time; the speed of movement; the difficulty of the training task can be changed by changing the position and number of pause points, pause time, and movement speed.

4. The sports training system according to claim 1, wherein: In the sports training system, after the trainee completes an action with reference to the demonstration trajectory, the degree of conformity M between the demonstration trajectory D and the limb movement trajectory L is used as a quantitative indicator of the movement control ability.

5. The sports training system according to claim 1, wherein: According to the quality of the user's completion of a training program, the difficulty C of the subsequent training program is selected.

Images