Intelligent Monitoring System and Device for Motion Trajectory Based on Gyroscope Dynamic Feedback

Abstract

This application relates to the technical field of motion monitoring, and specifically relates to an intelligent motion trajectory monitoring system and device based on gyroscope dynamic feedback. The system includes: a motion data monitoring module that acquires the attitude angles of the gyroscopes on both the left and right sides at each acquisition moment within each motion period; a motion attitude analysis module that calculates the motion difference degree between any two motion periods of a single gyroscope; by constructing a weighted undirected graph, clustering all motion periods to obtain the left-right dissimilarity degree of each clustering cluster; distinguishing all clustering clusters into each consistent cluster and each dissimilar cluster; obtaining the motion fluctuation coefficient of a single gyroscope in each dissimilar cluster; a motion attitude evaluation module that evaluates the stability of the motion trajectory of the motion carrier. This application can evaluate the motion trajectory of fancy rope skipping, so as to accurately evaluate whether the motion trajectory of the rope skipping person is stable, enabling the rope skipping person to clarify the motion effects of each fancy rope skipping action.

Description

Technical Field

[0001] This application relates to the technical field of motion monitoring, and particularly to an intelligent motion trajectory monitoring system and device based on gyroscope dynamic feedback. Background Art

[0002] As a simple and efficient form of aerobic exercise, skipping rope has occupied a place in the national fitness activities due to its low cost, high efficiency and low requirements for venues. Although the traditional skipping rope method can achieve certain exercise purposes, there are obvious limitations in terms of scientificity and personalized guidance. In order to achieve more accurate exercise effect evaluation and personalized training plan design, the deep integration of intelligent technology and sports has become an emerging trend.

[0003] When evaluating the motion posture of skipping rope by traditional methods, it can only be applied to the scenario of uniform standing skipping, detecting whether the skipping rhythm is uniform. However, for fancy skipping rope movements, it is impossible to accurately evaluate whether the motion trajectory of the skipping rope person is stable, resulting in limitations in the evaluation and monitoring of the motion trajectory of fancy skipping rope. Summary of the Invention

[0004] In order to solve the above technical problems, an intelligent motion trajectory monitoring system and device based on gyroscope dynamic feedback are provided to solve the existing problems.

[0005] The solution of this application to solve the technical problem is to provide an intelligent motion trajectory monitoring system and device based on gyroscope dynamic feedback, including the following steps:

[0006] In the first aspect, an embodiment of this application provides an intelligent motion trajectory monitoring system based on gyroscope dynamic feedback, and the system includes:

[0007] A motion data monitoring module, which is used to record the time required for each revolution of the motion carrier as each motion period, collect the acceleration and angular velocity on both sides of the motion carrier through a gyroscope, perform attitude calculation on them, and obtain the attitude angles of the gyroscopes on both sides at each acquisition moment within each motion period, where the attitude angles include pitch angle, roll angle, and yaw angle;

[0008] A motion posture analysis module, which is used to obtain the left-right dissimilarity of each consistent cluster and the motion fluctuation coefficient of a single gyroscope in each dissimilar cluster, specifically including:

[0009] (1) Calculate the motion difference degree between any two motion periods of a single gyroscope through the difference situations of the pitch angle, roll angle, and yaw angle between any two motion periods of a single gyroscope, and the difference situation of the time lengths of any two motion periods; based on the motion difference degree, cluster all motion periods by constructing a weighted undirected graph;

[0010] (2) Analyze the change amounts of the pitch angle, roll angle, and yaw angle of the single-sided gyroscope at each acquisition moment in each motion period in each clustering cluster, as well as the discrete situation of the difference in the change amounts between the left and right gyroscopes at different acquisition moments and the extreme change situation of the change amounts, to obtain the left-right dissimilarity of each clustering cluster;

[0011] (3) Based on the left-right dissimilarity, distinguish all clustering clusters into each consistent cluster and each dissimilar cluster; through the difference change and fluctuation situation of the motion difference degree between different motion periods in each dissimilar cluster by the single-sided gyroscope, obtain the motion fluctuation coefficient of the single-sided gyroscope in each dissimilar cluster;

[0012] The motion attitude evaluation module evaluates the stability of the motion trajectory of the moving carrier based on the left-right dissimilarity of each consistent cluster and the motion fluctuation coefficients of the left and right gyroscopes corresponding to each dissimilar cluster.

[0013] Preferably, calculating the motion difference degree between any two motion periods by the single-sided gyroscope includes:

[0014] Calculate the difference between the pitch angles at all acquisition moments within any two motion periods by the single-sided gyroscope, and record it as the relative difference;

[0015] Calculate the difference in the time length between any two motion periods, and record it as the duration difference;

[0016] Take the product of the relative difference and the duration difference as the first difference coefficient between any two motion periods by the single-sided gyroscope;

[0017] For the roll angle and yaw angle at all acquisition moments within any two motion periods by the single-sided gyroscope, use the same method as the first difference coefficient to obtain the second difference coefficient and the third difference coefficient between any two motion periods by the single-sided gyroscope;

[0018] The motion difference degree is the mean value of the first difference coefficient, the second difference coefficient, and the third difference coefficient.

[0019] Preferably, clustering all motion periods by constructing a weighted undirected graph includes:

[0020] Select a gyroscope on either side, denoted as the target gyroscope;

[0021] Take each motion period as a graph node, and there is an edge between any two graph nodes; take the reciprocal of the motion difference degree between any two motion periods by the target gyroscope as the edge weight between the corresponding two graph nodes, and construct a weighted undirected graph;

[0022] Cluster the weighted undirected graph to obtain multiple clustering clusters.

[0023] Preferably, obtaining the left - right dissimilarity of each clustering cluster includes:

[0024] Taking the sum of the accumulations of the differences in pitch angle, roll angle, and yaw angle between any acquisition moment of the unilateral gyroscope within each motion period of each clustering cluster and its adjacent acquisition moment as the attitude change amount at the any acquisition moment;

[0025] Calculating the motion dispersion of each motion period in each clustering cluster according to the discrete change situation of the differences in the attitude change amount between the left and right gyroscopes at different acquisition moments within each motion period in each clustering cluster;

[0026] Calculating the extreme difference of each motion period in each clustering cluster according to the difference situation between the maximum attitude change amount and the minimum attitude change amount of the left and right gyroscopes within each motion period in each clustering cluster;

[0027] The left - right dissimilarity is the mean value of the product of the motion dispersion and the extreme difference of all motion periods in each clustering cluster.

[0028] Preferably, calculating the motion dispersion of each motion period in each clustering cluster includes:

[0029] Taking the difference in the attitude change amount between the left and right gyroscopes at the same acquisition moment as the relative change difference;

[0030] The motion dispersion is the degree of dispersion of the relative change differences of all acquisition moments within each motion period in each clustering cluster.

[0031] Preferably, calculating the extreme difference of each motion period in each clustering cluster includes:

[0032] Calculating the difference in the maximum attitude change amount between the left and right gyroscopes at all acquisition moments within each motion period in each clustering cluster, and denoting it as the maximum difference;

[0033] Calculating the difference in the minimum attitude change amount between the left and right gyroscopes at all acquisition moments within each motion period in each clustering cluster, and denoting it as the minimum difference;

[0034] The extreme difference is the sum of the maximum difference and the minimum difference.

[0035] Preferably, distinguishing all clustering clusters into each consistent cluster and each dissimilar cluster includes:

[0036] Adopting a threshold segmentation algorithm to obtain the segmentation threshold of the left - right dissimilarity of all clustering clusters, and denoting the clustering clusters with the left - right dissimilarity less than or equal to the segmentation threshold as consistent clusters, and vice versa as dissimilar clusters.

[0037] Preferably, obtaining the motion fluctuation coefficients of the unilateral gyroscope in each different cluster includes:

[0038] Arrange all motion time periods in each different cluster in chronological order, and form a motion difference sequence of the unilateral gyroscope in each different cluster with the motion difference degree between the first motion time period and all the remaining motion time periods in each different cluster;

[0039] Using all the wave troughs in the motion difference sequence as segmentation points, divide the motion difference sequence into multiple subsequences;

[0040] Calculate the average value of the distances between any two subsequences of the unilateral gyroscope in each different cluster;

[0041] The motion fluctuation coefficient is the product of the degree of dispersion of the mean values of all subsequences of the unilateral gyroscope in each different cluster and the average value.

[0042] Preferably, evaluating the stability of the motion trajectory of the moving carrier includes:

[0043] Adopt an anomaly detection algorithm to respectively perform anomaly detection on the left-right difference degrees of all consistent clusters and the motion fluctuation coefficients of the left and right gyroscopes corresponding to all different clusters, and obtain the anomaly scores of each consistent cluster and the anomaly scores of each different cluster;

[0044] Normalize the anomaly scores of all consistent clusters and the anomaly scores of all different clusters respectively. If there are clusters with normalized anomaly scores greater than or equal to the preset threshold, the motion trajectories of the corresponding clusters are unstable.

[0045] In a second aspect, an embodiment of the present application further provides a motion trajectory intelligent monitoring device based on gyroscope dynamic feedback, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it adopts the motion trajectory intelligent monitoring system according to any one of the above.

[0046] The present application has at least the following beneficial effects:

[0047] This application analyzes the differences in pitch angle, roll angle, and yaw angle of a single-sided gyroscope during different motion periods, as well as the duration differences between different motion periods, and calculates the motion difference degree between any two motion periods of the single-sided gyroscope. Its beneficial effect is that it considers the consistency of the motion trajectories of the single-sided gyroscope during different motion periods, and based on the consistency of the motion trajectories of the left and right arms corresponding to the gyroscopes on the left and right sides, all motion periods are clustered, and the motion periods with similar motion trajectories are divided into one category; secondly, the left-right dissimilarity of each clustering cluster is calculated, and its beneficial effect is that it considers the inconsistent motion amplitudes of the left and right arms corresponding to the gyroscopes on the left and right sides to evaluate the same change degree of the left and right arms under various motion trajectories; based on the left-right dissimilarity, the fancy actions corresponding to different clustering clusters are distinguished, and the clustering clusters with similar motion trajectories of the left and right arms and the clustering clusters with inconsistent motion trajectories of the left and right arms are classified to obtain each consistent cluster and each dissimilar cluster. Finally, according to the discrete situation of the change in the motion difference degree between the first motion period and the remaining motion periods in each dissimilar cluster of the single-sided gyroscope, the motion fluctuation coefficient of the single-sided gyroscope in each dissimilar cluster is calculated. Its beneficial effect is that it considers the difference between the skipping rope motion in the first motion period and the remaining motion periods within each dissimilar cluster of the single-sided gyroscope to reflect the repetitive characteristics of the motion trajectory of the arm on the corresponding side of the single-sided gyroscope when the skipping rope person performs the skipping rope motion corresponding to the dissimilar cluster, thereby evaluating the stability of the motion trajectories of different fancy skipping rope motions; anomaly detection is performed on the left-right dissimilarity of each consistent cluster and the motion fluctuation coefficients of the left and right gyroscopes corresponding to each dissimilar cluster to evaluate the stability of the motion trajectory of the motion carrier. Its beneficial effect is that it can evaluate the motion trajectory of the fancy skipping rope motion so as to accurately evaluate whether the motion trajectory of the skipping rope person is stable, enabling the skipping rope person to clarify the motion effects of each fancy skipping rope action. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The following further elaborates on the intelligent motion trajectory monitoring system based on gyroscope dynamic feedback of this application with reference to the accompanying drawings.

[0049] Figure 1 It is a block diagram of the intelligent motion trajectory monitoring system based on gyroscope dynamic feedback provided by an embodiment of this application;

[0050] Figure 2 It is a step flowchart of the motion attitude analysis module provided by an embodiment of this application;

[0051] Figure 3 It is a step flowchart of the method for obtaining the left-right dissimilarity of each clustering cluster provided by an embodiment of this application;

[0052] Figure 4It is a flowchart of the steps for obtaining the motion fluctuation coefficients of the unilateral gyroscope provided in the embodiments of the present application in different clusters. Detailed implementation manners

[0053] In order to make the objectives, technical solutions and advantages of the present application clearer, the intelligent motion trajectory monitoring system and device based on gyroscope dynamic feedback proposed in the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0054] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present application belongs.

[0055] Please refer to Figure 1 , which shows a block diagram of an intelligent motion trajectory monitoring system based on gyroscope dynamic feedback provided in an embodiment of the present application. The system includes: a motion data monitoring module, a motion attitude analysis module, and a motion attitude evaluation module:

[0056] The motion data monitoring module is used to collect the acceleration and angular velocity on both sides of the motion carrier through the gyroscope, perform attitude calculation on them, and obtain the attitude angles of the gyroscopes on the left and right sides at each acquisition moment in each motion period after the motion carrier completes one motion cycle. Among them, the attitude angles include pitch angle, roll angle, and yaw angle.

[0057] By installing triaxial gyroscopes at the two handles of the skipping rope respectively, the gyroscope is a sensor that can measure angular velocity and acceleration, but its measurement results are based on its own coordinate system. Therefore, it is necessary to calibrate the gyroscope. When the skipping rope user stands still, the gyroscopes on the left and right sides are initially calibrated to align the coordinate system of the gyroscope with the carrier of the gyroscope, that is, the coordinate system of the skipping rope, and map the measurement results in the carrier coordinate system to the geographic coordinate system to analyze the trajectory and attitude of the skipping rope motion.

[0058] During the skipping rope motion of the skipping rope user, where the skipping rope is the motion carrier, the triaxial gyroscopes on the left and right sides collect the acceleration and angular velocity of the motion carrier where the gyroscopes are located, and perform attitude calculation on the data collected by the triaxial gyroscopes through the quaternion method of the Runge-Kutta method to respectively obtain the attitude angles of the gyroscopes on the left and right sides at each acquisition moment. The attitude angles include pitch angle, roll angle, and yaw angle.

[0059] It should be noted that the quaternion method of the Runge-Kutta method is a well-known technology and will not be elaborated here.

[0060] Among them, the attitude angle describes the attitude of an object through three consecutive rotations. It is usually represented by three angles, namely yaw angle, pitch angle, and roll angle; Yaw angle: The rotation angle around the vertical axis (usually the z-axis), which describes the orientation of the object on the horizontal plane; Pitch angle: The rotation angle around the horizontal axis (usually the y-axis), which describes the up-and-down inclination degree of the object; Roll angle: The rotation angle around the object's own axis (usually the x-axis), which describes the left-and-right inclination degree of the object.

[0061] Therefore, the attitude angle of the left gyroscope at the t-th acquisition moment Among them, is the pitch angle of the left gyroscope at the t-th acquisition moment; is the roll angle of the left gyroscope at the t-th acquisition moment, is the yaw angle of the left gyroscope at the t-th acquisition moment, and the attitude angle of the right gyroscope at the t-th acquisition moment Among them, is the pitch angle of the right gyroscope at the t-th acquisition moment; is the roll angle of the right gyroscope at the t-th acquisition moment, is the yaw angle of the right gyroscope at the t-th acquisition moment.

[0062] Install a Hall sensor on the skipping rope to collect the moments when the number of turns of the skipping rope changes. Whenever the Hall sensor detects that the skipping rope has completed one turn, a time point will be recorded. According to the recorded time points, all acquisition moments will be divided into multiple motion periods;

[0063] It should be noted that if the skipping rope user jumps N turns through the skipping rope, then from the start to the end of the skipping rope, the Hall sensor will record N moments of the number of turns change, and all the acquisition moments of the gyroscope collecting the entire skipping motion will be divided into N + 1 motion periods.

[0064] In this embodiment, the acquisition frequency of the three-axis gyroscope and the Hall sensor is 50 Hz, and the acquisition duration is from the start of the skipping rope until the end of the current skipping rope.

[0065] Thus, the attitude angles of the gyroscopes on the left and right sides at each acquisition moment within each motion period can be obtained respectively.

[0066] The motion attitude analysis module is used to obtain the left-right dissimilarity of each consistent cluster and the motion fluctuation coefficient of a single gyroscope in each dissimilar cluster.

[0067] Furthermore, the step flow chart of the motion attitude analysis module provided in this embodiment is as Figure 2 shown.

[0068] By analyzing the data collected by the gyroscope when a person is skipping rope, it is found that during normal standing still rope skipping, the degree of change in the data collected by the gyroscopes on the left and right sides at adjacent collection times has a certain similarity, and the movement trajectories collected by the left and right gyroscopes have a certain symmetry. When performing certain fancy rope skipping, although the movement trajectories of the left and right gyroscopes no longer have symmetry, under the same fancy rope skipping action, due to the repetition of the action, the data collected by the same gyroscope also has a certain similarity.

[0069] Step 1, calculate the movement difference degree between any two movement periods of a single-sided gyroscope based on the difference in pitch angle, roll angle, and yaw angle between any two movement periods of the single-sided gyroscope, as well as the difference in the time length of any two movement periods; based on the movement difference degree, cluster all movement periods by constructing a weighted undirected graph.

[0070] Since the time required to complete one circle for different fancy actions is different, classify the periods with similar movements by the difference in the time length of different movement periods corresponding to different fancy actions and the difference in the data collected within different movement periods, so as to distinguish different fancy actions, specifically including:

[0071] Calculate the difference between the pitch angles at all collection times within any two movement periods of a single-sided gyroscope, denoted as the relative difference;

[0072] In this embodiment, calculate the DTW distance between the pitch angles at all collection times within any two movement periods of a single-sided gyroscope. The calculation of the DTW distance is a well-known technology and will not be elaborated here. As other implementation manners, implementers can adopt other methods of the existing technology, such as Euclidean distance, etc. This embodiment does not make special restrictions on this.

[0073] Calculate the difference in the time length between any two movement periods of a single-sided gyroscope, denoted as the time length difference;

[0074] In this embodiment, calculate the absolute value of the difference in the time length between any two movement periods of a single-sided gyroscope, denoted as the time length difference.

[0075] Take the product of the relative difference and the time length difference as the first difference coefficient between any two movement periods of a single-sided gyroscope;

[0076] Similarly, for the roll angle and yaw angle at all collection times within any two movement periods of a single-sided gyroscope, use the same method as the first difference coefficient to obtain the second difference coefficient and the third difference coefficient between any two movement periods of a single-sided gyroscope.

[0077] It should be noted that the first difference coefficient can reflect the degree of difference in the pitch angle posture change of the arm on the corresponding side of the gyroscope during different motion periods; and considering that in the fancy rope skipping action, the posture angle changes in the same direction may also have a high degree of similarity. Therefore, the changes in the posture angles in all directions are measured by the first difference coefficient, the second difference coefficient, and the third difference coefficient to comprehensively reflect the consistency of the actions between two motion periods.

[0078] Take the mean value of the first difference coefficient, the second difference coefficient, and the third difference coefficient as the motion difference degree of the unilateral gyroscope between any two motion periods;

[0079] It should be noted that the motion difference degree can reflect the difference situation between the motion trajectories of the arm on the corresponding side of the rope skipper in different directions during the motion periods corresponding to two rope skipping turns.

[0080] Furthermore, based on the motion difference degree, construct a weighted undirected graph to perform clustering analysis on all motion periods, specifically as follows:

[0081] Select the gyroscope on either side, denoted as the target gyroscope;

[0082] Take each motion period as a graph node, and there is an edge between any two graph nodes; take the reciprocal of the motion difference degree between any two motion periods of the target gyroscope as the edge weight between the corresponding two graph nodes to construct a weighted undirected graph;

[0083] It should be noted that when calculating the reciprocal of the motion difference degree between any two motion periods of the target gyroscope, to avoid the motion difference degree being 0, a preset value greater than 0 is added to the denominator. In this embodiment, the preset value greater than 0 is taken as 0.1. As other implementation manners, the implementer can set it according to the actual situation.

[0084] Perform clustering on the weighted undirected graph to obtain multiple clustering clusters;

[0085] In this embodiment, the Markov clustering algorithm is used to cluster the weighted undirected graph to obtain the clustering result of each motion period. Among them, the Markov clustering algorithm is a well-known technology and will not be elaborated here.

[0086] It should be noted that each clustering cluster represents the motion periods with similar motion postures, that is, each clustering cluster corresponds to a kind of motion action, so as to distinguish the fancy actions of the arm on the corresponding side of the target gyroscope.

[0087] So far, multiple clustering clusters are obtained.

[0088] Step 2: Analyze the change amounts of the pitch angle, roll angle, and yaw angle of the unilateral gyroscope at each acquisition moment in each motion period of each clustering cluster, as well as the discrete situation of the difference in the change amounts between the left and right gyroscopes at different acquisition moments and the extreme change situation of the change amounts, to obtain the left-right dissimilarity of each clustering cluster.

[0089] Furthermore, when the rope skipping person is performing fancy rope skipping, the movement trajectories of the left and right arms of the rope skipping person are relatively corresponding, that is, the movements of the left and right arms are synchronized in time and amplitude at the same acquisition moment. Secondly, there are some opposite-direction movements of the arms in the rope skipping motion. For example, in cross skipping, the change amplitudes of the left and right arms are the same, but only the change directions in space are different, which leads to inconsistent attitude angles collected by the left and right gyroscopes. Therefore, analyze whether the change amplitudes of the movement trajectories of the left and right arms are the same when the rope skipping person is performing the fancy movement corresponding to each clustering cluster, calculate the left-right dissimilarity, so as to reflect whether the movements of the rope skipping person are standard. The step flowchart of the method for obtaining the left-right dissimilarity of each clustering cluster provided in the embodiments of the present application is as Figure 3 shown, and specifically includes:

[0090] Take the cumulative sum of the differences in the pitch angle, roll angle, and yaw angle between any acquisition moment of the unilateral gyroscope in each motion period within each clustering cluster and its previous acquisition moment as the attitude change amount at the any acquisition moment;

[0091] In this embodiment, taking the mth motion period in the kth clustering cluster as an example, the calculation formula for the attitude change amount of the right gyroscope at the tth acquisition moment is:

[0092]

[0093] where is the attitude change amount of the right gyroscope at the tth acquisition moment in the mth motion period in the kth clustering cluster; are respectively the pitch angle, roll angle, and yaw angle of the right gyroscope at the tth acquisition moment in the mth motion period in the kth clustering cluster, are respectively the pitch angle, roll angle, and yaw angle of the right gyroscope at the (t - 1)th acquisition moment in the mth motion period in the kth clustering cluster;

[0094] The calculation formula for the attitude change amount of the left gyroscope at the tth acquisition moment is:

[0095]

[0096] where is the attitude change amount of the left gyroscope at the tth acquisition moment in the mth motion period in the kth clustering cluster; They are respectively the pitch angle, roll angle, and yaw angle of the gyroscope on the left side at the t-th acquisition moment during the m-th motion period within the k-th clustering cluster. They are respectively the pitch angle, roll angle, and yaw angle of the gyroscope on the left side at the (t - 1)-th acquisition moment during the m-th motion period within the k-th clustering cluster.

[0097] Denote the difference in the attitude change amount between the gyroscopes on the left and right sides at any acquisition moment as the relative change difference.

[0098] In this embodiment, denote the absolute value of the difference in the attitude change amount between the gyroscopes on the left and right sides at any acquisition moment as the relative change difference.

[0099] Calculate the degree of dispersion of the relative change differences at all acquisition moments within each motion period in each clustering cluster as the motion dispersion degree of each motion period in each clustering cluster.

[0100] In this embodiment, the degree of dispersion is measured by calculating the variance of the relative change differences at all acquisition moments within each motion period in each clustering cluster. As other implementation manners, implementers can adopt other methods in the prior art, such as the coefficient of variation, etc. This embodiment does not make special restrictions on this.

[0101] It should be noted that the smaller the relative change difference, the closer the motion amplitudes of the left and right arms at the same acquisition moment. The smaller the motion dispersion degree, the smaller the fluctuation of the difference in the change amplitude of the motion trajectories of the left and right arms within the same motion period, and the more consistent the change of the motion trajectories of the left and right arms.

[0102] Calculate the difference in the maximum attitude change amount between the gyroscopes on the left and right sides at all acquisition moments within each motion period in each clustering cluster, and denote it as the maximum difference.

[0103] Calculate the difference in the minimum attitude change amount between the gyroscopes on the left and right sides at all acquisition moments within each motion period in each clustering cluster, and denote it as the minimum difference.

[0104] In this embodiment, calculate the absolute value of the difference in the maximum attitude change amount between the gyroscopes on the left and right sides at all acquisition moments within each motion period in each clustering cluster, and denote it as the maximum difference; calculate the absolute value of the difference in the minimum attitude change amount between the gyroscopes on the left and right sides at all acquisition moments within each motion period in each clustering cluster, and denote it as the minimum difference.

[0105] Take the sum of the maximum difference and the minimum difference as the extreme difference of each motion period in each clustering cluster.

[0106] The mean of the product of the motion dispersions of all motion periods in each clustering cluster and the extreme difference is used as the left-right dissimilarity degree of each clustering cluster;

[0107] It should be noted that the extreme difference reflects the difference degree between the maximum and minimum change amplitudes of the left and right arms within the same motion period. The smaller the extreme difference, the smaller the obtained left-right dissimilarity degree, indicating that when the rope skipping person completes the motion actions corresponding to the clustering cluster, the change amplitudes of the motion trajectories of the left and right arms in each circle of rope skipping are relatively consistent, and the more stable the motion posture of the rope skipping person when completing the motion actions corresponding to the clustering cluster, and the greater the possibility of conforming to the standard rope skipping trajectory.

[0108] Thus, the left-right dissimilarity degrees of all clustering clusters are obtained.

[0109] Step 3: Based on the left-right dissimilarity degree, all clustering clusters are divided into consistent clusters and dissimilar clusters; through the difference change and fluctuation of the motion difference degree between different motion periods in each dissimilar cluster by the single-sided gyroscope, the motion fluctuation coefficient of the single-sided gyroscope in each dissimilar cluster is obtained.

[0110] In fancy rope skipping, the changes of the left and right arms are not all consistent in all actions. Therefore, it is necessary to distinguish the actions with inconsistent motion postures of the left and right arms. That is, based on the left-right dissimilarity degrees of all clustering clusters, the clustering clusters corresponding to the inconsistent motion postures of the left and right arms are distinguished, specifically:

[0111] The threshold segmentation algorithm is adopted to obtain the segmentation threshold of the left-right dissimilarity degrees of all clustering clusters. The clustering clusters with the left-right dissimilarity degree less than or equal to the segmentation threshold are recorded as consistent clusters, and vice versa, as dissimilar clusters;

[0112] In this embodiment, the Otsu threshold segmentation algorithm is adopted for threshold segmentation. Among them, the Otsu threshold segmentation algorithm is a well-known technology and will not be elaborated here.

[0113] Furthermore, in fancy rope skipping actions, although the trajectory changes of the two arms may not be consistent under the same group of fancy actions, the motion trajectory of a single arm will show a certain repeatability. Therefore, by analyzing whether the motion trajectories of a single arm in different circles within the same group of fancy actions are repeatable and calculating the motion fluctuation coefficient, the stability of the rope skipping person when performing fancy actions can be evaluated. The step flow chart of the method for obtaining the motion fluctuation coefficient of the single-sided gyroscope in each dissimilar cluster provided by the embodiment of the present application is as Figure 4 shown, specifically including:

[0114] Arrange all the motion periods in each different cluster in chronological order, and form a motion difference sequence of the unilateral gyroscope in each different cluster with the motion difference degree between the first motion period and all the remaining motion periods of the unilateral gyroscope in each different cluster.

[0115] It should be noted that for the convenience of understanding, assume that all the motion periods in the q-th different cluster are I3, I5, I7, and I8 respectively, and the motion difference degree between the right gyroscope in the I3-th motion period and the I5-th motion period is α 3.5 , and the motion difference degree between the right gyroscope in the I3-th motion period and the I7-th motion period is α 3.7 , and the motion difference degree between the right gyroscope in the I3-th motion period and the I8-th motion period is α 3.8 , then the motion difference sequence of the right gyroscope in the q-th different cluster is [α 3.5 , α 3.7 , α 3.8 .

[0116] It should be noted that the motion difference sequence reflects the change difference degree of the motion trajectory of the arm on the corresponding side of the unilateral gyroscope between the first jump circle and all the subsequent jump circles in the fancy actions corresponding to each different cluster.

[0117] Secondly, considering that although the motion trajectory of the unilateral arm is repetitive, it may repeat after several circles, that is, although the motion trajectory of the unilateral arm will repeat, there is a certain motion period interval. Since if the motion difference degree between two motion periods is small, the motion trajectories of the right arm in the two motion periods are relatively consistent, so by performing valley analysis on the motion difference sequence, the motion difference sequence can be divided into multiple subsequences to determine whether the motion change of the unilateral arm has stages and regularities.

[0118] Use all the valleys in the motion difference sequence as the segmentation points to divide the motion difference sequence into multiple subsequences;

[0119] In this embodiment, the difference method is used to obtain all the valleys in the motion difference sequence.

[0120] Calculate the average value of the distances between all arbitrary two subsequences of the unilateral gyroscope in each different cluster;

[0121] In this embodiment, calculate the average value of the DTW distances between all arbitrary two subsequences of the unilateral gyroscope in each different cluster. Among them, the calculation of the DTW distance is a well-known technology and will not be elaborated here. As other implementation manners, implementers can use other methods of existing technologies, such as Euclidean distance, etc. This embodiment does not make special restrictions on this.

[0122] Calculate the product of the degree of dispersion of the means of all subsequences of the unilateral gyroscope in each different cluster and the mean value as the motion fluctuation coefficient of the unilateral gyroscope in each different cluster;

[0123] It should be noted that the mean value reflects the differences in the motion trajectories of the unilateral arm at different stages under the fancy movements corresponding to different clusters. The smaller the mean value, the more repetitive the motion trajectory of the unilateral arm. The degree of dispersion of the means of all subsequences of the unilateral gyroscope in each different cluster reflects whether the motion trajectories of the unilateral arm at different stages are relatively stable. The smaller the obtained motion fluctuation coefficient, the more stable the actions of the rope skipping person under the fancy movements corresponding to different clusters. The motion fluctuation coefficient takes into account the possible intervals in the repetitive changes of the unilateral arm. By analyzing the motion difference sequences in stages, it can be ensured that regardless of whether there are intervals between the repetitive changes of the unilateral arm, it can accurately and comprehensively reflect whether the motion trajectory of the unilateral arm is stable, and thus can more accurately evaluate the stability of the rope skipping person during fancy movements.

[0124] Thus, the motion fluctuation coefficients of the unilateral gyroscope in each different cluster are obtained.

[0125] The motion posture evaluation module evaluates the stability of the motion trajectory of the moving carrier based on the left-right difference degrees of each consistent cluster and the motion fluctuation coefficients of the left and right gyroscopes corresponding to each different cluster.

[0126] Furthermore, based on the left-right difference degrees of all consistent clusters and the motion fluctuation coefficients of the left and right gyroscopes corresponding to all different clusters, evaluate the stability of the motion trajectory of the moving carrier, specifically:

[0127] Adopt an outlier detection algorithm to perform outlier detection on the left-right difference degrees of all consistent clusters to obtain the outlier scores of each consistent cluster;

[0128] Adopt an outlier detection algorithm to perform outlier detection on the motion fluctuation coefficients of the left and right gyroscopes corresponding to all different clusters to obtain the outlier scores of each different cluster;

[0129] In this embodiment, the LOF (Local Outliers Factor) outlier detection algorithm is used for outlier detection. Among them, the LOF outlier detection algorithm is a well-known technology and will not be elaborated here.

[0130] Normalize the outlier scores of all consistent clusters and the outlier scores of all different clusters respectively. If there are clusters with normalized outlier scores greater than or equal to the preset threshold, the motion trajectories of the corresponding clusters are unstable.

[0131] In this embodiment, the preset threshold value is 0.5. As for other implementation manners, the implementer can set it according to the actual situation.

[0132] It should be noted that if there is a cluster with a normalized abnormal score greater than or equal to the preset threshold, it indicates that when the rope skipping person performs the fancy movement corresponding to this cluster, the movement trajectory of the arm fluctuates greatly, and the rope skipping person has a poor movement effect on this rope skipping movement. Then the rope skipping person can repeat the practice of this rope skipping movement.

[0133] Based on the same inventive concept as the above system, the embodiment of the present application also provides a smart monitoring device for the movement trajectory based on gyroscope dynamic feedback, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it adopts the smart monitoring system for the movement trajectory based on gyroscope dynamic feedback described in any one of the above.

[0134] It should be understood that although Figure 1 the steps in the flowchart of Figure 1 are shown in sequence according to the indication of the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover,

[0135] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0136] The above-described embodiments only represent several implementation manners of the present application, and their descriptions are relatively specific and detailed, but they should not be construed as a limitation to the present application. It should be pointed out that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made. Therefore, any simple modification, equivalent change, and modification made to the above embodiments based on the technical essence of the present application without departing from the content of the technical solution of the present application all belong to the protection scope of the technical solution of the present application.

Claims

1. A smart motion trajectory monitoring system based on gyroscopic dynamic feedback, characterized in that, The system includes: A motion data monitoring module, which is used to record the time required for each revolution of the moving carrier as each motion period, collect the acceleration and angular velocity on both sides of the moving carrier through a gyroscope, perform attitude calculation on them, and obtain the attitude angles of the left and right gyroscopes at each acquisition moment within each motion period, where the attitude angles include pitch angle, roll angle, and yaw angle; A motion attitude analysis module, which is used to obtain the left-right dissimilarity of each consistent cluster and the motion fluctuation coefficient of a single-side gyroscope in each dissimilar cluster, specifically including: (1) Calculate the motion dissimilarity between any two motion periods of a single-side gyroscope based on the differences in pitch angle, roll angle, and yaw angle between any two motion periods of the single-side gyroscope, and the difference in the time lengths of any two motion periods; based on the motion dissimilarity, perform clustering on all motion periods by constructing a weighted undirected graph; (2) Take the cumulative sum of the differences in pitch angle, roll angle, and yaw angle between any acquisition moment and its adjacent acquisition moment within each motion period of a single-side gyroscope in each clustering cluster as the attitude change amount at the any acquisition moment; calculate the motion dispersion of each motion period in each clustering cluster according to the discrete change of the differences in the attitude change amounts between the left and right gyroscopes at different acquisition moments within each motion period in each clustering cluster; calculate the extreme difference of each motion period in each clustering cluster according to the difference between the maximum attitude change amount and the minimum attitude change amount of the left and right gyroscopes within each clustering cluster at each motion period; the left-right dissimilarity is the mean value of the product of the motion dispersion and the extreme difference of all motion periods in each clustering cluster; (3) Based on the left-right dissimilarity, distinguish all clustering clusters into consistent clusters and dissimilar clusters, adopt a threshold segmentation algorithm to obtain the segmentation threshold of the left-right dissimilarity of all clustering clusters, and mark the clustering clusters with the left-right dissimilarity less than or equal to the segmentation threshold as consistent clusters, and vice versa as dissimilar clusters; obtain the motion fluctuation coefficient of a single-side gyroscope in each dissimilar cluster through the difference change and fluctuation of the motion dissimilarity between different motion periods in each dissimilar cluster; A motion attitude evaluation module, which evaluates the stability of the motion trajectory of the moving carrier based on the left-right dissimilarity of each consistent cluster and the motion fluctuation coefficients of the left and right gyroscopes corresponding to each dissimilar cluster.

2. The intelligent motion trajectory monitoring system based on gyroscope dynamic feedback according to claim 1, characterized in that, The calculation of the motion dissimilarity between any two motion periods of a single-side gyroscope includes: Calculate the difference in pitch angles at all acquisition moments within any two motion periods of a single-side gyroscope, and record it as the relative difference; Calculate the difference in time lengths between any two motion periods, and record it as the duration difference; Take the product of the relative difference and the duration difference as the first difference coefficient between any two motion periods of a single-side gyroscope; For the roll angle and yaw angle at all acquisition moments within any two motion periods of a single-side gyroscope, use the same method as the first difference coefficient to obtain the second difference coefficient and the third difference coefficient between any two motion periods of a single-side gyroscope; The degree of motion difference is the mean of the first difference coefficient, the second difference coefficient, and the third difference coefficient.

3. The intelligent motion trajectory monitoring system based on gyroscope dynamic feedback according to claim 1, characterized in that, Clustering all motion periods by constructing a weighted undirected graph includes: Select a gyroscope on either side, denoted as the target gyroscope; Take each motion period as a graph node, and there is an edge between any two graph nodes; take the reciprocal of the degree of motion difference between the target gyroscope in any two motion periods as the edge weight between the corresponding two graph nodes, and construct a weighted undirected graph; Cluster the weighted undirected graph to obtain multiple clusters.

4. The intelligent motion trajectory monitoring system based on gyroscope dynamic feedback according to claim 1, characterized in that, Calculating the motion dispersion of each motion period in each cluster includes: Denote the difference in the amount of attitude change between the left and right gyroscopes at the same acquisition moment as the relative change difference; The motion dispersion is the degree of dispersion of the relative change differences at all acquisition moments within each motion period in each cluster.

5. The intelligent motion trajectory monitoring system based on gyroscope dynamic feedback according to claim 1, wherein Calculating the extreme difference of each motion period in each cluster includes: Calculate the difference in the maximum amount of attitude change between the left and right gyroscopes at all acquisition moments within each motion period in each cluster, denoted as the maximum difference; Calculate the difference in the minimum amount of attitude change between the left and right gyroscopes at all acquisition moments within each motion period in each cluster, denoted as the minimum difference; The extreme difference is the sum of the maximum difference and the minimum difference.

6. The intelligent motion trajectory monitoring system based on gyroscope dynamic feedback according to claim 1, characterized in that, Obtaining the motion fluctuation coefficient of the single-sided gyroscope in each different cluster includes: Arrange all motion periods in each different cluster in chronological order, and form a motion difference sequence of the single-sided gyroscope in each different cluster with the degree of motion difference between the first motion period and all other motion periods in each different cluster; Use all the wave troughs in the motion difference sequence as segmentation points to divide the motion difference sequence into multiple subsequences; Calculate the average value of the distances between any two subsequences of the single-sided gyroscope in each different cluster; The motion fluctuation coefficient is the product of the degree of dispersion of the mean values of all subsequences of the single-sided gyroscope in each different cluster and the average value.

7. The intelligent motion trajectory monitoring system based on gyroscope dynamic feedback according to claim 1, characterized in that, Evaluating the stability of the motion trajectory of the moving carrier includes: Adopt an anomaly detection algorithm to respectively perform anomaly detection on the left-right difference degree of all consistent clusters and the motion fluctuation coefficients of the left and right gyroscopes corresponding to all different clusters, and obtain the anomaly scores of each consistent cluster and the anomaly scores of each different cluster; Normalize the anomaly scores of all consistent clusters and all different clusters respectively. If there is a cluster with a normalized anomaly score greater than or equal to the preset threshold, the motion trajectory of the corresponding cluster is unstable.

8. An intelligent motion trajectory monitoring device based on gyroscopic dynamic feedback, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it adopts the intelligent motion trajectory monitoring system based on gyroscope dynamic feedback as described in any one of claims 1-7.

Images