Vestibular illusion strength training method

Abstract

The present invention relates to the technical field of vestibular illusion intensity training, and in particular to a vestibular illusion intensity training method. The training data to be tested is obtained, and the physiological parameters of the person to be tested for vestibular illusion test based on the training data to be tested are obtained; the physiological parameters are identified to determine the target vestibular illusion intensity corresponding to the person to be tested; the target vestibular illusion intensity is compared with the preset vestibular illusion intensity corresponding to the training data to be tested; and the training data to be tested is adjusted according to the comparison result. Thereby, it is ensured that the adjusted training data to be tested matches the person to be tested, the accuracy of testing the person to be tested based on the adjusted training data to be tested is achieved, and the training data to be tested can be adjusted according to individual differences, thereby ensuring the flexibility of vestibular illusion intensity training.

Description

Technical Field

[0001] The present invention relates to the technical field of vestibular illusion intensity training, and specifically to a method for vestibular illusion intensity training. Background Art

[0002] Spatial disorientation is one of the problems that almost all pilots have encountered during flight. Among them, vestibular illusions are caused by the physiological limitations of the human vestibular system and are also one of the types of spatial disorientation that pilots have difficulty coping with.

[0003] In the existing vestibular illusion training, generally, preset training data is used to conduct vestibular illusion training for all personnel to be tested, but it does not consider whether the vestibular illusion training intensity corresponding to the preset training data is applicable to each person to be tested. Therefore, the results of the vestibular illusion training intensity for each person to be tested are not accurate, and training cannot be carried out according to individual differences, resulting in poor training flexibility. Summary of the Invention

[0004] In view of this, the present invention provides a method for vestibular illusion intensity training to solve the problems that the results of the vestibular illusion training intensity for each person to be tested are not accurate, training cannot be carried out according to individual differences, and training flexibility is poor.

[0005] In a first aspect, the present invention provides a method for vestibular illusion intensity training, the method comprising:

[0006] Obtaining training data to be tested, and obtaining physiological parameters of a person to be tested during a vestibular illusion test based on the training data to be tested;

[0007] Identifying the physiological parameters to determine the target vestibular illusion intensity corresponding to the person to be tested;

[0008] Comparing the target vestibular illusion intensity with the preset vestibular illusion intensity corresponding to the training data to be tested;

[0009] Adjusting the training data to be tested according to the comparison result.

[0010] The vestibular illusion intensity training method provided by the embodiments of the present application obtains the training data to be tested and the physiological parameters of the person to be tested for the vestibular illusion test based on the training data to be tested, ensuring the accuracy of the physiological parameters for the vestibular illusion test of the training data to be tested. Identifying the physiological parameters to determine the target vestibular illusion intensity corresponding to the person to be tested ensures the accuracy of the determined target vestibular illusion intensity corresponding to the person to be tested. Comparing the target vestibular illusion intensity with the preset vestibular illusion intensity corresponding to the training data to be tested, and adjusting the training data to be tested according to the comparison result. Thus, it is ensured that the adjusted training data to be tested matches the person to be tested, the accuracy of testing the person to be tested based on the adjusted training data to be tested, and the adjustment of the training data to be tested can be realized according to individual differences, thereby ensuring the flexibility of the vestibular illusion intensity training.

[0011] In an alternative embodiment, identifying the physiological parameters to determine the target vestibular illusion intensity corresponding to the person to be tested includes:

[0012] Obtain an initial classification model;

[0013] Based on the initial classification model, identify the physiological parameters to determine the target vestibular illusion intensity corresponding to the person to be tested.

[0014] The vestibular illusion intensity training method provided by the embodiments of the present application obtains an initial classification model; based on the initial classification model, identifies the physiological parameters to determine the target vestibular illusion intensity corresponding to the person to be tested, ensuring the accuracy of the determined target vestibular illusion intensity corresponding to the person to be tested.

[0015] In an alternative embodiment, obtaining the initial classification model includes:

[0016] Obtain an initial training data set, where the initial training data set includes multiple pieces of initial training data, and each piece of initial training data includes training physiological parameters and the training vestibular illusion intensity corresponding to the training physiological parameters;

[0017] Input the initial training data set into an initial classification network, the initial classification network identifies each piece of initial training data, and according to the identification result, generates multiple initial micro-clusters and determines the training vestibular illusion intensity corresponding to each initial micro-cluster;

[0018] Generate an initial classification model according to each initial micro-cluster.

[0019] The vestibular illusion intensity training method provided by the embodiments of the present application obtains an initial training data set, inputs the initial training data set into an initial classification network, the initial classification network identifies each piece of initial training data, and generates a plurality of initial micro-clusters according to the identification results, and determines the training vestibular illusion intensity corresponding to each initial micro-cluster, ensuring the accuracy of the training vestibular illusion intensity corresponding to each determined initial micro-cluster. According to each initial micro-cluster, an initial classification model is generated, ensuring the accuracy of the generated initial classification model.

[0020] In an alternative embodiment, based on the initial classification model, physiological parameters are identified to determine the target vestibular illusion intensity corresponding to the person to be tested, including:

[0021] Obtain the number of preset classifiers and the number of preset micro-clusters corresponding to each preset classifier;

[0022] For each preset classifier, calculate the first similarity between the physiological parameters and each initial micro-cluster in the initial classification model;

[0023] Select the target micro-clusters with the number of preset micro-clusters from each initial micro-cluster according to the first similarity;

[0024] According to the training vestibular illusion intensity corresponding to each target micro-cluster, determine the training vestibular illusion intensity with the largest number as the standby vestibular illusion intensity corresponding to the preset classifier;

[0025] According to each standby vestibular illusion intensity, determine the target vestibular illusion intensity corresponding to the person to be tested.

[0026] The vestibular illusion intensity training method provided by the embodiments of the present application obtains the number of preset classifiers and the number of preset micro-clusters corresponding to each preset classifier; for each preset classifier, calculate the first similarity between the physiological parameters and each initial micro-cluster in the initial classification model, ensuring the accuracy of the calculated first similarity. Select the target micro-clusters with the number of preset micro-clusters from each initial micro-cluster according to the first similarity, ensuring the accuracy of the determined target micro-clusters. According to the training vestibular illusion intensity corresponding to each target micro-cluster, determine the training vestibular illusion intensity with the largest number as the standby vestibular illusion recognition intensity corresponding to the preset classifier, ensuring the accuracy of the standby vestibular illusion recognition intensity corresponding to each determined preset classifier. According to each standby vestibular illusion recognition intensity, determine the target vestibular illusion intensity corresponding to the person to be tested, ensuring the accuracy of the determined target vestibular illusion intensity and the robustness of the algorithm.

[0027] In an alternative embodiment, according to each standby vestibular illusion intensity, determine the target vestibular illusion intensity corresponding to the person to be tested, including:

[0028] Determine the accuracy corresponding to each preset classifier according to the backup vestibular illusion intensity corresponding to each preset classifier;

[0029] Determine the backup vestibular illusion intensity corresponding to the preset classifier with the maximum accuracy as the target vestibular illusion intensity.

[0030] The vestibular illusion intensity training method provided by the embodiments of the present application determines the accuracy corresponding to each preset classifier according to the backup vestibular illusion intensity corresponding to each preset classifier, ensuring the accuracy of determining the accuracy corresponding to each preset classifier. Determining the backup vestibular illusion intensity corresponding to the preset classifier with the maximum accuracy as the target vestibular illusion intensity ensures the accuracy of the determined target vestibular illusion intensity.

[0031] In an optional implementation manner, after determining the backup vestibular illusion intensity corresponding to the preset classifier with the maximum accuracy as the target vestibular illusion intensity, the method further includes:

[0032] Obtain each target microcluster corresponding to the preset classifier with the maximum accuracy;

[0033] Update each target microcluster according to the relationship between the training vestibular illusion intensity corresponding to each target microcluster and the target vestibular illusion intensity to generate a first updated microcluster;

[0034] Update the initial classification model according to each first updated microcluster to generate an updated classification model.

[0035] The vestibular illusion intensity training method provided by the embodiments of the present application obtains each target microcluster corresponding to the preset classifier with the maximum accuracy; updates each target microcluster according to the relationship between the training vestibular illusion intensity corresponding to each target microcluster and the target vestibular illusion intensity to generate a first updated microcluster, thereby ensuring the accuracy of the generated first updated microcluster. Update the initial classification model according to each first updated microcluster, thereby ensuring the accuracy of updating the initial classification model, so as to determine the target vestibular illusion intensity corresponding to the physiological parameters generated later based on the updated classification model.

[0036] In an optional implementation manner, the method further includes:

[0037] For each preset classifier, if the target microclusters of the preset microcluster quantity are not selected from each initial microcluster according to the first similarity, generate new microclusters based on the physiological parameters;

[0038] Calculate the second similarity between the new microclusters and each initial microcluster;

[0039] According to each second similarity, determine the training vestibular illusion intensity corresponding to the initial microcluster with the maximum second similarity as the candidate vestibular illusion intensity corresponding to the new microcluster;

[0040] Determine the target vestibular illusion intensity corresponding to the physiological parameter according to the candidate vestibular illusion intensity.

[0041] In the vestibular illusion intensity training method provided by the embodiments of the present application, for each preset classifier, if the target microclusters of the preset microcluster quantity are not selected from each initial microcluster according to the first similarity, new microclusters are generated based on the physiological parameter, ensuring that the new microclusters can represent the physiological parameter. Then, the second similarity between the new microclusters and each initial microcluster is calculated, ensuring the accuracy of the calculated similarity. According to each second similarity, the training vestibular illusion intensity corresponding to the initial microcluster with the largest second similarity is determined as the candidate vestibular illusion intensity corresponding to the new microcluster, ensuring the accuracy of the determined candidate vestibular illusion intensity corresponding to the new microcluster. Then, according to the candidate vestibular illusion intensity, the target vestibular illusion intensity corresponding to the physiological parameter is determined, and thus the accuracy of the target vestibular illusion intensity corresponding to the physiological parameter can be ensured.

[0042] In an alternative embodiment, to determine the target vestibular illusion intensity corresponding to the physiological parameter according to the candidate vestibular illusion intensity, the method includes:

[0043] Obtain the new quantity corresponding to the new microcluster;

[0044] If the new quantity is greater than the preset new quantity threshold, obtain a backup training data set; the backup training data set includes multiple pieces of backup training data, and each piece of backup training data includes a backup physiological parameter and a backup vestibular illusion intensity corresponding to the backup physiological parameter;

[0045] Calculate the third similarity between each piece of backup training data and each new microcluster and each initial microcluster;

[0046] Determine the initial microcluster or new microcluster corresponding to each piece of backup training data according to each third similarity;

[0047] Update each initial microcluster according to the backup training data corresponding to each initial microcluster to generate a second updated microcluster;

[0048] Compare the backup vestibular illusion intensity of the backup training data corresponding to the new microcluster with the candidate vestibular illusion intensity;

[0049] If the backup vestibular illusion intensity is consistent with the candidate vestibular illusion intensity, determine the candidate vestibular illusion intensity as the target vestibular illusion intensity;

[0050] If the backup vestibular illusion intensity is inconsistent with the candidate vestibular illusion intensity, replace the candidate vestibular illusion intensity with the backup vestibular illusion intensity and determine the backup vestibular illusion intensity as the target vestibular illusion intensity;

[0051] Update each newly added micro-cluster according to the corresponding backup training data of each newly added micro-cluster to generate a third updated micro-cluster;

[0052] Update the initial classification model according to each second updated micro-cluster and the third updated micro-cluster to generate a candidate classification model.

[0053] The vestibular illusion intensity training method provided by the embodiments of the present application obtains the newly added quantity corresponding to the newly added micro-clusters; if the newly added quantity is greater than the preset newly added quantity threshold, obtain the backup training data set; calculate the third similarity between each backup training data and each newly added micro-cluster and each initial micro-cluster, ensuring the accuracy of the third similarity calculated between each backup training data and each newly added micro-cluster and each initial micro-cluster. According to each third similarity, determine the initial micro-cluster or newly added micro-cluster corresponding to each backup training data, ensuring the accuracy of the third similarity between each determined backup training data and each newly added micro-cluster and each initial micro-cluster. Then, update each initial micro-cluster according to the backup training data corresponding to each initial micro-cluster to generate a second updated micro-cluster, ensuring the accuracy of the generated second updated micro-cluster. Compare the backup vestibular illusion intensity of the backup training data corresponding to the newly added micro-clusters with the candidate vestibular illusion intensity; if the backup vestibular illusion intensity is consistent with the candidate vestibular illusion intensity, determine the candidate vestibular illusion intensity as the target vestibular illusion intensity; if the backup vestibular illusion intensity is inconsistent with the candidate vestibular illusion intensity, replace the candidate vestibular illusion intensity with the backup vestibular illusion intensity and determine the backup vestibular illusion intensity as the target vestibular illusion intensity, thereby ensuring the accuracy of the determined target vestibular illusion intensity. Update each newly added micro-cluster according to the backup training data corresponding to each newly added micro-cluster to generate a third updated micro-cluster, ensuring the accuracy of the generated third updated micro-cluster. Update the initial classification model according to each second updated micro-cluster and the third updated micro-cluster to generate a candidate classification model, ensuring the accuracy of the generated candidate classification model.

[0054] In an alternative embodiment, updating the initial classification model according to each second updated micro-cluster and the third updated micro-cluster to generate a candidate classification model includes:

[0055] Obtain the current total number of the first micro-clusters corresponding to the candidate classification model; the total number of the first micro-clusters includes the number of the second updated micro-clusters and the number of the third updated micro-clusters;

[0056] Compare the total number of the first micro-clusters with the preset total number threshold of the micro-clusters;

[0057] If the total number of the first micro-clusters is greater than the preset total number threshold of the micro-clusters, calculate the fourth similarity between each third updated micro-cluster and each second updated micro-cluster;

[0058] According to each fourth similarity, merge each third updated micro-cluster with the second updated micro-cluster with the maximum fourth similarity to generate a first merged micro-cluster;

[0059] Based on the first merged micro-cluster, update the candidate classification model to generate a target classification model.

[0060] For the vestibular illusion intensity training method provided by the embodiments of this application, obtain the current total number of first micro-clusters corresponding to the candidate classification model, and compare the total number of first micro-clusters with a preset total micro-cluster number threshold; if the total number of first micro-clusters is greater than the preset total micro-cluster number threshold, calculate the fourth similarity between each third updated micro-cluster and each second updated micro-cluster, ensuring the accuracy of the calculated fourth similarity between each third updated micro-cluster and each second updated micro-cluster. According to each fourth similarity, merge each third updated micro-cluster with the second updated micro-cluster with the maximum fourth similarity to generate a first merged micro-cluster, ensuring the accuracy of the generated first merged micro-cluster, reducing the total number of first micro-clusters in the candidate classification model, and ensuring that the current number of micro-clusters corresponding to each training vestibular illusion intensity is not too small. Based on the first merged micro-cluster, update the candidate classification model to generate a target classification model, ensuring the accuracy of the generated target classification model.

[0061] In an alternative implementation manner, based on the first merged micro-cluster, updating the candidate classification model to generate a target classification model includes:

[0062] Obtain the current total number of second micro-clusters corresponding to the candidate classification model;

[0063] If the total number of second micro-clusters is greater than the preset total micro-cluster number threshold and there are no third updated micro-clusters in the candidate classification model, obtain the current number of micro-clusters corresponding to each vestibular illusion intensity in the candidate classification model;

[0064] Calculate the fifth similarity between each current micro-cluster corresponding to the vestibular illusion intensity with the largest current number of micro-clusters;

[0065] Merge the current micro-clusters with the closest fifth similarity to generate a second merged micro-cluster;

[0066] Update the candidate classification model according to the first merged micro-cluster and the second merged micro-cluster to generate a target classification model.

[0067] The vestibular illusion intensity training method provided by the embodiments of the present application obtains the current total number of second micro-clusters corresponding to the candidate classification model; if the total number of second micro-clusters is greater than the preset total micro-cluster number threshold and there is no third updated micro-cluster in the candidate classification model, then the current number of micro-clusters corresponding to each vestibular illusion intensity in the candidate classification model is obtained, ensuring the accuracy of the current number of micro-clusters corresponding to each vestibular illusion intensity in the obtained candidate classification model. Calculate the fifth similarity between the current micro-clusters corresponding to the vestibular illusion intensity with the largest current number of micro-clusters, ensuring the accuracy of the fifth similarity between the current micro-clusters corresponding to the vestibular illusion intensity with the largest current number of micro-clusters. Merge the current micro-clusters with the closest fifth similarity to generate a second merged micro-cluster, ensuring the accuracy of the generated second merged micro-cluster. Update the candidate classification model according to the first merged micro-cluster and the second merged micro-cluster to generate a target classification model. The total number of micro-clusters in the target classification model is reduced, and it is ensured that the current number of micro-clusters corresponding to each vestibular illusion intensity is not too small. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0069] Figure 1 is a flowchart of the vestibular illusion intensity training method according to the embodiments of the present invention;

[0070] Figure 2 is a flowchart of another vestibular illusion intensity training method according to the embodiments of the present invention;

[0071] Figure 3 is a flowchart of yet another vestibular illusion intensity training method according to the embodiments of the present invention;

[0072] Figure 4 is a structural block diagram of the vestibular illusion intensity training device according to the embodiments of the present invention;

[0073] Figure 5 is a hardware structure diagram of the electronic device according to the embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0074] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts fall within the protection scope of the present invention.

[0075] It should be noted that for the method for training the intensity of vestibular illusion provided in the embodiments of the present application, the execution subject may be a device for training the intensity of vestibular illusion. The device for training the intensity of vestibular illusion can be implemented as part or all of an electronic device through software, hardware, or a combination of software and hardware. Among them, the electronic device may be a server or a terminal. Among them, the server in the embodiments of the present application may be a single server or a server cluster composed of multiple servers. The terminal in the embodiments of the present application may be other intelligent hardware devices such as a smart phone, a personal computer, a tablet computer, a wearable device, and a smart robot. In the following method embodiments, the execution subject is taken as an electronic device for illustration.

[0076] According to an embodiment of the present invention, an embodiment of a method for training the intensity of vestibular illusion is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings may be executed in a computer system such as a set of computer-executable instructions. And although the logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than here.

[0077] In this embodiment, a method for training the intensity of vestibular illusion is provided, which can be used for the above-mentioned electronic device. Figure 1 is a flowchart of the method for training the intensity of vestibular illusion according to an embodiment of the present invention, as Figure 1 shown. The process includes the following steps:

[0078] Step S101, obtain the training data to be tested, and obtain the physiological parameters of the person to be tested for vestibular illusion testing based on the training data to be tested.

[0079] Specifically, the electronic device may receive the training data to be tested input by the user, or may receive the training data to be tested sent by other devices. The embodiments of the present application do not make specific limitations on the manner in which the electronic device obtains the training data to be tested.

[0080] Exemplarily, the training data to be tested may include {(axial), (steering), (rotation speed), (duration), (starting rotation time) (axial), (steering), (rotation speed), (duration)}. For example, the training data to be tested correspondingly includes {(zb), (clockwise), (60° / s), (60s), (30s) (xb), (clockwise), (45° / s), (1s)}.

[0081] Then, the electronic device can drive a six-degree-of-freedom motion platform to perform linear and rotational motions with the training data to be tested to conduct a vestibular illusion test on the person to be tested. The electronic device can acquire the corresponding physiological parameters of the person to be tested based on the physiological sensors worn on the person to be tested.

[0082] Among them, the six-degree-of-freedom motion platform can rotate and perform linear motions along the three axes of the six-degree-of-freedom motion platform coordinate system xp, yp, and zp, and can provide complex six-degree-of-freedom comprehensive motion acceleration stimuli. It is stipulated that the origin Op of the six-degree-of-freedom motion platform coordinate system is located at the center of mass of the six-degree-of-freedom motion platform. The xp axis of the six-degree-of-freedom motion platform coordinate system is within the symmetry plane of the six-degree-of-freedom motion platform, with the forward direction being positive; the yp axis of the six-degree-of-freedom motion platform coordinate system is perpendicular to the symmetry plane of the six-degree-of-freedom motion platform where the xp axis is located, with the direction pointing to the right being positive; the zp axis of the six-degree-of-freedom motion platform coordinate system is perpendicular to the xpOpyp plane, and the positive direction of the zp axis is determined to be downward according to the right-hand rule.

[0083] Among them, the physiological sensors may include an electrocardiogram sensor, a pulse sensor, a respiration sensor, etc. The physiological parameters include the electrocardiogram signal (ECG), pulse signal (PPG), and respiration signal (RESP) corresponding to the trainee.

[0084] Step S102: Identify the physiological parameters to determine the target vestibular illusion intensity corresponding to the person to be tested.

[0085] Specifically, the electronic device can input the physiological parameters into a preset recognition model to identify the physiological parameters and determine the target vestibular illusion intensity corresponding to the person to be tested.

[0086] Among them, the preset recognition model can be any one of a radial basis function (RBF) network, a feedforward neural network (FFNN), a convolutional neural network (CNN), a deconvolutional network (DN), a deep convolutional inverse graphics network (DCIGN), a generative adversarial network (GAN), a recurrent neural network (RNN), a long short-term memory network (LSTM), a deep residual network (DRN), and an extreme learning machine (ELM). The embodiments of the present application do not limit the preset recognition model.

[0087] Step S103: Compare the target vestibular illusion intensity with the preset vestibular illusion intensity corresponding to the training data to be tested.

[0088] Specifically, the electronic device can compare the target vestibular illusion intensity with the preset vestibular illusion intensity.

[0089] Step S104: Adjust the training data to be tested according to the comparison result.

[0090] Specifically, if the target vestibular illusion intensity is inconsistent with the preset vestibular illusion intensity, the electronic device adjusts the training data to be tested.

[0091] If the target vestibular illusion intensity is consistent with the preset vestibular illusion intensity, the electronic device does not need to adjust the training data to be tested.

[0092] The vestibular illusion intensity training method provided by the embodiments of the present application obtains the training data to be tested and the physiological parameters of the person to be tested for vestibular illusion testing based on the training data to be tested, ensuring the accuracy of the physiological parameters for vestibular illusion testing of the obtained training data to be tested. Identifying the physiological parameters to determine the target vestibular illusion intensity corresponding to the person to be tested ensures the accuracy of the determined target vestibular illusion intensity corresponding to the person to be tested. Comparing the target vestibular illusion intensity with the preset vestibular illusion intensity corresponding to the training data to be tested, and adjusting the training data to be tested according to the comparison result. Thereby ensuring that the adjusted training data to be tested matches the person to be tested, the accuracy of testing the person to be tested based on the adjusted training data to be tested, and the adjustment of the training data to be tested can be realized according to individual differences, thus ensuring the flexibility of vestibular illusion intensity training.

[0093] In this embodiment, a vestibular illusion intensity training method is provided, which can be used in the above-mentioned electronic device. Figure 2 It is a flowchart of the vestibular illusion intensity training method according to the embodiments of the present invention, as Figure 2 shown, and the process includes the following steps:

[0094] Step S201, obtain the training data to be tested and the physiological parameters of the person to be tested for vestibular illusion testing based on the training data to be tested.

[0095] For this step, please refer to the introduction of step S101 above and will not be elaborated here.

[0096] Step S202, identify the physiological parameters to determine the target vestibular illusion intensity corresponding to the person to be tested.

[0097] Specifically, the above step S202 may include the following steps:

[0098] Step S2021, obtain the initial classification model.

[0099] Optionally, the electronic device can receive the initial classification model input by the user or receive the initial classification model sent by other devices.

[0100] Among them, the initial classification model can be a k-means clustering algorithm model or other network models, and the embodiments of the present application do not make specific limitations on the initial classification model.

[0101] Specifically, the above step S2021 may include the following steps:

[0102] Step a1, obtain the initial training dataset.

[0103] The initial training data set includes multiple pieces of initial training data, each piece of initial training data includes a training physiological parameter and a training vestibular illusion intensity corresponding to the training physiological parameter.

[0104] Specifically, the electronic device may receive an initial training data set input by a user, or may receive an initial training data set sent by other devices.

[0105] The present application embodiment does not specifically limit the manner in which the electronic device obtains the initial training data set.

[0106] Step a2, input the initial training data set into the initial classification network, the initial classification network identifies each piece of initial training data, generates multiple initial micro-clusters based on the identification results, and determines the training vestibular illusion intensity corresponding to each initial micro-cluster.

[0107] Specifically, the electronic device can input the initial training data set into the initial classification network, and the initial classification network identifies each piece of initial training data, generates multiple initial micro-clusters according to the identification results, and determines the training vestibular illusion intensity corresponding to each initial micro-cluster.

[0108] Wherein, the initial classification network can be a k-means clustering algorithm or other networks. The embodiment of the present application does not specifically limit the initial classification network.

[0109] For example, the k-means clustering algorithm is used as an example. Specifically, the electronic device can input the initial training data set into the k-means clustering algorithm, and use the k-means clustering algorithm to divide the initial training data into k clusters. Among them:

[0110] The k-means clustering strategy selects the best partition function C* of the initial training data set by minimizing the loss function. The best partition function C* is as follows:

[0111] (1)

[0112] Wherein, the loss function is defined as the first similarity between a sample and the center of its category, as shown in formula (2).

[0113] (2)

[0114] In formula (2) is the mean or center of the lth cluster in the cluster, , is a classification function. Q describes the distance between the samples in the cluster and the cluster center. The smaller the Q value, the higher the similarity of the samples in the cluster.

[0115] Then, the electronic device calculates the feature information of each cluster and stores it in the initial microcluster in the form of cluster features. The cluster features include seven pieces of feature information as shown in formula (3).

[0116] (3)

[0117] In formula (3), and are the linear sum and the sum of squares of all the initial training data in the initial microcluster, which are and respectively; N represents the number of all the initial training data in the initial microcluster; Z is the reliability of the initial microcluster at the current time, and the initial value is set to 1; T is the update time of the initial microcluster, and the initial value is 0; ML is the label of the initial microcluster; R is the radius of the initial microcluster, and R is defined as the standard deviation of the distance between the initial training data points in the cluster and the centroid, and is calculated according to formula (4).

[0118] (4)

[0119] The electronic device obtains the training vestibular illusion intensity corresponding to each initial training data in the initial microcluster, and determines the training vestibular illusion intensity corresponding to each initial microcluster as the training vestibular illusion intensity corresponding to the initial training data with the largest quantity.

[0120] Step a3, generate an initial classification model according to each initial microcluster.

[0121] Specifically, the electronic device generates an initial classification model according to each initial microcluster.

[0122] Exemplarily, therefore, the same vestibular illusion intensity level can fall into different initial microclusters. Therefore, the present invention sets a certain number of initial microclusters for each vestibular illusion intensity level to avoid misclassification of a single instance of the vestibular illusion intensity level. The pseudo-code of this part of the algorithm is shown in Algorithm A.

[0123]

[0124] Step S2022, based on the initial classification model, identify the physiological parameters and determine the target vestibular illusion intensity corresponding to the person to be tested.

[0125] Optionally, the electronic device may calculate the similarity between the physiological parameter and each initial training data corresponding to each initial micro-cluster in the initial classification model. Then, for each initial micro-cluster, based on the similarity between each initial training data in each initial micro-cluster and the physiological parameter, calculate the average similarity between each initial training data and the physiological parameter. Compare the average similarities corresponding to each initial micro-cluster, and determine the initial micro-cluster with the largest average similarity as the target micro-cluster, and determine the training vestibular illusion intensity corresponding to the target micro-cluster as the target vestibular illusion intensity corresponding to the person to be tested.

[0126] Specifically, the above step S2022 may include the following steps:

[0127] Step b01, obtain the number of preset classifiers and the number of preset micro-clusters corresponding to each preset classifier.

[0128] Specifically, the electronic device may receive the number of preset classifiers input by the user, or may obtain the number of preset classifiers sent by other devices. In addition, the electronic device may also obtain the number of preset micro-clusters corresponding to each preset classifier, or may receive the number of preset micro-clusters corresponding to each preset classifier sent by other devices.

[0129] The embodiments of the present application do not specifically limit the manner in which the electronic device obtains the number of preset classifiers and the number of preset micro-clusters corresponding to each preset classifier.

[0130] Exemplarily, assume that the number of preset classifiers is θ, that is, θ × K-NN ( ) is used, and K represents the number of preset micro-clusters corresponding to the preset classifier. For example, when θ = 1, there is only one K-NN classifier (i.e., 1-NN, that is, the number of preset micro-clusters corresponding to the preset classifier is 1, or 3-NN, that is, the number of preset micro-clusters corresponding to the preset classifier is 3); θ = 2 represents an integration strategy of 2 × K-NN classifiers (i.e., 1-NN and 3-NN). And so on, when θ = 4, a 4 × K-NN classifier is adopted (i.e., the number of preset micro-clusters corresponding to the preset classifier is 1, 3, 5, 7 respectively).

[0131] Step b02, for each preset classifier, calculate the first similarity between the physiological parameter and each initial micro-cluster in the initial classification model.

[0132] Specifically, for each preset classifier, the electronic device can calculate the similarity between the physiological parameters and each initial training data corresponding to each initial microcluster in the initial classification model. Then, for each initial microcluster, based on the similarity between each initial training data in each initial microcluster and the physiological parameters, the average similarity between each initial training data and the physiological parameters is calculated, and the average similarity corresponding to each initial microcluster is determined as the first similarity between the physiological parameters and each initial microcluster in the initial classification model.

[0133] Step b03: Select a target microcluster with a preset number of microclusters from each initial microcluster according to the first similarity.

[0134] Specifically, the electronic device compares the first similarity between the physiological parameters and each initial microcluster in the initial classification model, arranges each initial microcluster in descending order according to the first similarity, and determines the initial microcluster with the preset number of microclusters ranked at the front as the target microcluster.

[0135] Exemplarily, if θ = 1 and there is only one K-NN classifier, that is, 1-NN, then the preset number of microclusters is determined to be 1, and the electronic device determines the initial microcluster ranked first as the target microcluster.

[0136] Exemplarily, if θ = 2 and there are two K-NN classifiers, that is, 1-NN and 3-NN. For 1-NN, the electronic device determines the preset number of microclusters to be 1, and the electronic device determines the initial microcluster ranked first as the target microcluster. For 3-NN, the electronic device determines the preset number of microclusters to be 3, and the electronic device determines the first three initial microclusters ranked as the target microclusters.

[0137] Step b04: Determine the training vestibular illusion intensity with the largest number as the standby vestibular illusion intensity corresponding to the preset classifier according to the training vestibular illusion intensity corresponding to each target microcluster.

[0138] Specifically, for each preset classifier, the electronic device determines the training vestibular illusion intensity corresponding to each target microcluster in the preset classifier, and then compares the training vestibular illusion intensity corresponding to each target microcluster to determine the training vestibular illusion intensity with the largest number as the standby vestibular illusion intensity corresponding to the preset classifier.

[0139] Exemplarily, for 1-NN, the electronic device determines the training vestibular illusion intensity corresponding to the target microcluster in 1-NN as the standby vestibular illusion intensity corresponding to 1-NN.

[0140] For 3-NN, the electronic device compares the training vestibular illusion intensities corresponding to the 3 target microclusters in 3-NN to determine the training vestibular illusion intensity with the largest number as the standby vestibular illusion intensity corresponding to 3-NN.

[0141] Exemplarily, the electronic device finds target micro-clusters of a preset number of micro-clusters closest to the physiological parameter χ in each initial micro-cluster of the initial classification model according to the Euclidean distance, and forms a set denoted as DK(χ); in DK(χ), the majority voting rule is used to determine the category y of χ, as shown in formula (5).

[0142] (5)

[0143] Among them, I in formula (5) is an indicator function. When , I is 1, otherwise I is 0. is the class of the physiological parameter, .

[0144] Step b05, determine the target vestibular illusion intensity corresponding to the person to be tested according to each spare vestibular illusion intensity.

[0145] Optionally, when the number of preset classifiers is 1, the electronic device may determine the spare vestibular illusion intensity corresponding to the preset classifier as the target vestibular illusion intensity corresponding to the person to be tested.

[0146] When the number of preset classifiers is greater than 1, the electronic device may compare the spare vestibular illusion intensities corresponding to the preset classifiers, and then determine the spare vestibular illusion intensity with the largest number as the target vestibular illusion intensity corresponding to the person to be tested.

[0147] Optionally, the above step b05 may include the following steps:

[0148] Step b051, determine the accuracy corresponding to each preset classifier according to the spare vestibular illusion intensity corresponding to each preset classifier.

[0149] Specifically, the electronic device may calculate the weight corresponding to each preset classifier according to the spare vestibular illusion recognition intensity corresponding to each preset classifier. Then, according to the weight corresponding to each preset classifier, determine the accuracy corresponding to the preset classifier.

[0150] Step b052, determine the spare vestibular illusion intensity corresponding to the preset classifier with the highest accuracy as the target vestibular illusion intensity.

[0151] Specifically, the electronic device may determine the spare vestibular illusion intensity corresponding to the preset classifier with the highest accuracy as the target vestibular illusion intensity.

[0152] Exemplarily, the weight calculation of each preset classifier at time ti is shown in formula (6).

[0153] (6)

[0154] In formula (6), ζ represents the weight matrix of each preset classifier, c represents the number of recently labeled instances, where an instance is used to represent a set of physiological parameters. represents the time step. for classification , is the true class label. Among them, is the time step of the instance's logical function.

[0155] If the recognition result is consistent with the actual result, the sum value is incremented by 1, otherwise by 0. Divide the result by c to obtain the average weight of the classifier. Therefore, the weight value represents the average effective classification accuracy of the corresponding classifier.

[0156] Finally, the best classifier is the one with the maximum weight, which is used to identify the class label of the current instance, as shown in formulas (7) and (8).

[0157] (7)

[0158] (8)

[0159] Among them, in formulas (7) and (8), K is an important parameter in the K-NN classification algorithm, which is related to the selection of the θ value in the ensemble strategy.

[0160] In an alternative embodiment of the present application, after the above step b52, the following steps may further be included:

[0161] Step b053, obtaining each target microcluster corresponding to the preset classifier with the highest accuracy.

[0162] Specifically, the electronic device may obtain each target microcluster corresponding to the preset classifier with the highest accuracy.

[0163] Step b054, updating each target microcluster according to the relationship between the training vestibular illusion intensity corresponding to each target microcluster and the target vestibular illusion intensity, to generate a first updated microcluster.

[0164] Specifically, the electronic device may update the features corresponding to each target microcluster according to the relationship between the training vestibular illusion intensity corresponding to each target microcluster and the target vestibular illusion recognition intensity, to generate a first updated microcluster.

[0165] Exemplarily, the electronic device may use an exponential decay function to calculate the change of the target microcluster weight over time, as shown in formula (9).

[0166] (9)

[0167] Among them, in formula (9), λ is the decay rate, and E is the time elapsed since the target micro-cluster was last updated.

[0168] Therefore, the reliability of the target micro-cluster gradually decreases over time. However, if the target micro-cluster is selected to make a correct classification, the reliability of the target micro-cluster increases by 1.

[0169] After eliminating the outdated target micro-clusters, the real-time incoming physiological parameter χ is incrementally added to the nearest target micro-cluster, and the update of the target micro-cluster is obtained according to formulas (10)-(13).

[0170] (10)

[0171] (11)

[0172] (12)

[0173] (13)

[0174] Among them, in formulas (10)-(13) is the current time.

[0175] Step b055, update the initial classification model according to each first updated micro-cluster to generate an updated classification model.

[0176] Specifically, the electronic device updates the initial classification model according to each first updated micro-cluster to generate an updated classification model.

[0177] In an alternative embodiment of the present application, the above method may further include the following steps:

[0178] Step b06, for each preset classifier, if the target micro-clusters of the preset micro-cluster quantity are not selected from each initial micro-cluster according to the first similarity, generate a new added micro-cluster based on the physiological parameter.

[0179] Specifically, for each preset classifier, if the first similarity between the physiological parameter and each initial micro-cluster in the initial classification model is less than the preset similarity threshold, that is, there is no target micro-cluster in each initial micro-cluster that is relatively similar to the physiological parameter, it is determined that there is no target micro-cluster in each initial micro-cluster. That is, if the electronic device does not select the target micro-clusters of the preset micro-cluster quantity from each initial micro-cluster, the electronic device may generate a new added micro-cluster based on the physiological parameter, and the new added micro-cluster includes the physiological parameter.

[0180] Step b07, calculate the second similarity between the new added micro-cluster and each initial micro-cluster.

[0181] The electronic device can calculate the second similarity between the new features corresponding to the new micro-clusters and the initial features corresponding to the initial micro-clusters, so as to obtain the second similarity between the new micro-clusters and each initial micro-cluster.

[0182] Among them, the second similarity can be the Euclidean distance, the cosine distance, or other parameters. The embodiments of the present application do not make specific limitations on the second similarity.

[0183] Step b08: According to each second similarity, determine the training vestibular illusion intensity corresponding to the initial micro-cluster with the largest second similarity as the candidate vestibular illusion intensity corresponding to the new micro-cluster.

[0184] Specifically, the electronic device can compare the second similarities between the new micro-clusters and each initial micro-cluster, and then determine the training vestibular illusion intensity corresponding to the initial micro-cluster with the largest second similarity as the candidate vestibular illusion intensity corresponding to the new micro-cluster.

[0185] Step b09: Determine the target vestibular illusion intensity corresponding to the physiological parameter according to the candidate vestibular illusion intensity.

[0186] In an alternative embodiment of the present application, the electronic device can determine the candidate vestibular illusion intensity corresponding to the new micro-cluster as the target vestibular illusion intensity corresponding to the physiological parameter.

[0187] In an alternative embodiment of the present application, the above step b09 may include the following steps:

[0188] Step b091: Obtain the new quantity corresponding to the new micro-cluster.

[0189] Specifically, after each generation of a new micro-cluster, the electronic device can count the new micro-clusters. Then, according to the counting situation of the new micro-clusters, determine the new quantity corresponding to the new micro-cluster.

[0190] Step b092: If the new quantity is greater than the preset new quantity threshold, obtain the backup training data set.

[0191] Among them, the backup training data set includes multiple pieces of backup training data, and each piece of backup training data includes a backup physiological parameter and the backup vestibular illusion intensity corresponding to the backup physiological parameter.

[0192] Specifically, the electronic device can receive the preset new quantity threshold input by the user, or can receive the preset new quantity threshold sent by other devices. The electronic device can also determine the preset new quantity threshold according to the storage situation and operation situation of the electronic device, or the total number of micro-clusters allowed in the initial classification model.

[0193] The embodiments of the present application do not specifically limit the manner in which the electronic device obtains the preset new quantity threshold.

[0194] Then, the electronic device compares the new quantity with the preset new quantity threshold. If the new quantity is greater than the preset new quantity threshold, it obtains a backup training data set.

[0195] Among them, the electronic device can receive the backup training data set input by the user, can also receive the backup training data set sent by other devices, and can also receive the backup training data set sent by other devices.

[0196] The embodiments of the present application do not specifically limit the manner in which the electronic device obtains the backup training data set.

[0197] Step b093: Calculate the third similarity between each piece of backup training data and each new microcluster and each initial microcluster.

[0198] Specifically, the electronic device can calculate the third similarity between each piece of backup training data and each new microcluster and each initial microcluster.

[0199] Among them, for the specific method of calculating the third similarity, reference can be made to the method of calculating the first similarity, which will not be elaborated here.

[0200] Step b094: Determine the initial microcluster or new microcluster corresponding to each piece of backup training data according to each third similarity.

[0201] Specifically, for each piece of backup training data, the electronic device can select the new microcluster or initial microcluster with the largest third similarity from among the third similarities between the backup training data and each new microcluster and each initial microcluster, so as to determine the initial microcluster or new microcluster corresponding to each piece of backup training data.

[0202] Step b095: Update each initial microcluster according to the backup training data corresponding to each initial microcluster to generate a second updated microcluster.

[0203] Specifically, the electronic device can update each initial microcluster according to the backup training data corresponding to each initial microcluster to generate a second updated microcluster.

[0204] Among them, for the process of updating each initial microcluster according to the backup training data corresponding to each initial microcluster to generate a second updated microcluster, reference can be made to the content of step b054 above, which will not be elaborated here.

[0205] Step b096: Compare the backup vestibular illusion intensity of the backup training data corresponding to the new microcluster with the candidate vestibular illusion intensity.

[0206] Specifically, the electronic device may compare the spare vestibular illusion intensity of the spare training data corresponding to the newly added micro-cluster with the candidate vestibular illusion intensity.

[0207] Step b097, if the spare vestibular illusion intensity is consistent with the candidate vestibular illusion intensity, determine the candidate vestibular illusion intensity as the target vestibular illusion intensity.

[0208] Specifically, if the spare vestibular illusion intensity is consistent with the candidate vestibular illusion intensity, the electronic device determines the candidate vestibular illusion intensity as the target vestibular illusion intensity.

[0209] Step b098, if the spare vestibular illusion intensity is inconsistent with the candidate vestibular illusion intensity, replace the candidate vestibular illusion intensity with the spare vestibular illusion intensity, and determine the spare vestibular illusion intensity as the target vestibular illusion intensity.

[0210] Specifically, if the spare vestibular illusion intensity is inconsistent with the candidate vestibular illusion intensity, the electronic device may replace the candidate vestibular illusion intensity with the spare vestibular illusion intensity and determine the spare vestibular illusion intensity as the target vestibular illusion intensity.

[0211] Step b099, update each newly added micro-cluster according to the spare training data corresponding to each newly added micro-cluster to generate a third updated micro-cluster.

[0212] Specifically, the electronic device may update each newly added micro-cluster according to the spare training data corresponding to each newly added micro-cluster to generate a third updated micro-cluster.

[0213] Among them, for the process of updating each newly added micro-cluster according to the spare training data corresponding to each newly added micro-cluster to generate a third updated micro-cluster, reference can be made to the content of step b054 above, which will not be elaborated here.

[0214] Step b0910, update the initial classification model according to each second updated micro-cluster and the third updated micro-cluster to generate a candidate classification model.

[0215] Specifically, the electronic device may update the initial classification model according to each second updated micro-cluster and the third updated micro-cluster to generate a candidate classification model.

[0216] In an optional implementation manner of the present application, the above step b0910 may include the following steps:

[0217] Step c1, obtain the current total number of the first micro-clusters corresponding to the candidate classification model.

[0218] Among them, the total number of the first micro-clusters includes the number of the second updated micro-clusters and the number of the third updated micro-clusters.

[0219] Specifically, the electronic device counts the second updated micro-clusters and the third updated micro-clusters in the initial classification model to determine the current total number of the first micro-clusters corresponding to the candidate classification model.

[0220] Step c2: Compare the total number of the first micro-clusters with a preset total micro-cluster number threshold.

[0221] Specifically, the electronic device can receive the preset total micro-cluster number threshold input by the user, can also receive the preset total micro-cluster number threshold sent by other devices, or can also set the preset total micro-cluster number threshold according to the memory of the electronic device.

[0222] The embodiments of the present application do not specifically limit the manner in which the electronic device obtains the preset total micro-cluster number threshold.

[0223] Then, the electronic device compares the total number of the first micro-clusters with the preset total micro-cluster number threshold.

[0224] Step c3: If the total number of the first micro-clusters is greater than the preset total micro-cluster number threshold, calculate the fourth similarity between each third updated micro-cluster and each second updated micro-cluster.

[0225] Specifically, if the total number of the first micro-clusters is greater than the preset total micro-cluster number threshold, for each third updated micro-cluster, the electronic device can calculate the fourth similarity between the third updated micro-cluster and each second updated micro-cluster.

[0226] Step c4: Based on each fourth similarity, merge each third updated micro-cluster with the second updated micro-cluster with the maximum fourth similarity to generate a first merged micro-cluster.

[0227] Specifically, for each third updated micro-cluster, the electronic device can determine the second updated micro-cluster with the maximum fourth similarity to the third updated micro-cluster, and then merge the third updated micro-cluster with the second updated micro-cluster with the maximum fourth similarity to generate a first merged micro-cluster.

[0228] Step c5: Update the candidate classification model based on the first merged micro-cluster to generate a target classification model.

[0229] Specifically, the electronic device can update the candidate classification model based on the first merged micro-cluster to generate a target classification model.

[0230] In an alternative embodiment of the present application, the above step c5 may include the following steps:

[0231] Step c51: Obtain the current total number of the second micro-clusters corresponding to the candidate classification model.

[0232] Specifically, the electronic device can count the current micro-clusters in the candidate classification model to obtain the current total number of the second micro-clusters corresponding to the candidate classification model.

[0233] Step c52: If the number of second total micro - clusters is greater than the preset total micro - cluster number threshold and there is no third updated micro - cluster in the candidate classification model, obtain the current micro - cluster numbers corresponding to each vestibular illusion intensity in the candidate classification model.

[0234] Specifically, the electronic device can compare the number of second total micro - clusters with the preset total micro - cluster number threshold. If the number of second total micro - clusters is greater than the preset total micro - cluster number threshold, the electronic device can detect whether there is a third updated micro - cluster in the candidate classification model. If there is no third updated micro - cluster in the candidate classification model, the electronic device counts the vestibular illusion intensities corresponding to each current micro - cluster in the candidate classification model, so as to determine the current micro - cluster numbers corresponding to each vestibular illusion intensity.

[0235] Step c53: Calculate the fifth similarity between each of the current micro - clusters corresponding to the vestibular illusion intensity with the largest current micro - cluster number.

[0236] Specifically, the electronic device can calculate the fifth similarity between each of the current micro - clusters corresponding to the vestibular illusion intensity with the largest current micro - cluster number based on a preset similarity algorithm.

[0237] Among them, the preset similarity algorithm can be any one of algorithms such as the Euclidean distance algorithm, the cosine similarity algorithm, the Pearson correlation coefficient algorithm, etc. The embodiments of the present application do not make specific limitations on the preset similarity algorithm.

[0238] Step c54: Merge the current micro - clusters with the closest fifth similarity to generate a second merged micro - cluster.

[0239] Specifically, the electronic device can merge the current micro - clusters with the closest fifth similarity to generate a second merged micro - cluster.

[0240] Step c55: Update the candidate classification model according to the first merged micro - cluster and the second merged micro - cluster to generate a target classification model.

[0241] Specifically, the electronic device can update the candidate classification model according to the first merged micro - cluster and the second merged micro - cluster to generate a target classification model.

[0242] Exemplarily, the algorithm can be as follows:

[0243]

[0244] Specifically, to better introduce the vestibular illusion training method provided by the embodiments of the application, as Figure 3 shown, the embodiments of the present application provide a flow chart for identifying physiological parameters to determine the target vestibular illusion recognition intensity corresponding to the subject. As Figure 3As shown, the electronic device can acquire physiological parameters, which may include electrocardiogram signal (ECG), pulse signal (PPG), and respiration signal (RESP). Then the electronic device performs data preprocessing on the physiological parameters and divides the physiological parameters using a sliding time window to generate a sliding time window data stream. The electronic device determines the marker information corresponding to the physiological parameters based on a K-NN ensemble classifier. Among them, the K-NN ensemble classifier is generated based on an initial model and a preset classifier. Among them, the marker information is the target vestibular illusion recognition intensity corresponding to the physiological parameters. If the electronic device determines the marker information corresponding to the physiological parameters, it performs a representative learning method based on error driving to update the target microcluster in the initial classification model, thereby updating the initial classification model. If the electronic device fails to determine the marker information corresponding to the physiological parameters, it uses an exponential decay function to discard microclusters with negative confidence / low negative confidence. Then, it is judged whether the physiological parameters are within the radius R of the nearest microcluster in terms of distance. If so, the first updated microcluster; if not, a new microcluster is created, thereby updating the initial classification model. Among them, the initial classification model is obtained by training the k-means clustering algorithm based on an initial training set, and the initial classification model includes multiple initial microclusters.

[0245] Step S203, compare the target vestibular illusion intensity with the preset vestibular illusion intensity corresponding to the training data to be tested.

[0246] For the introduction of this step, please refer to the above introduction of step S103.

[0247] Step S204, adjust the training data to be tested according to the comparison result.

[0248] Specifically, the above step S204 may include the following steps:

[0249] Step S2041, if the target vestibular illusion intensity is greater than the preset vestibular illusion intensity, adjust the training data to be tested to reduce the preset vestibular illusion intensity corresponding to the training data to be tested.

[0250] Step S2042, if the target vestibular illusion intensity is less than the preset vestibular illusion intensity, adjust the training data to be tested to increase the preset vestibular illusion intensity corresponding to the training data to be tested.

[0251] Specifically, if the target vestibular illusion intensity is greater than the preset vestibular illusion intensity, adjust the training data to be tested to reduce the preset vestibular illusion intensity corresponding to the training data to be tested.

[0252] If the target vestibular illusion intensity is less than the preset vestibular illusion intensity, adjust the training data to be tested to increase the preset vestibular illusion intensity corresponding to the training data to be tested.

[0253] Exemplarily, the electronic device can adjust at least one parameter in {(axial direction), (steering), (rotation speed), (duration), (starting rotation moment), (axial direction), (steering), (rotation speed), (duration) corresponding to the preset test control instruction.

[0254] For the vestibular illusion intensity training method provided by the embodiments of the present application, an initial training data set is obtained and input into an initial classification network. The initial classification network identifies each piece of initial training data, generates a plurality of initial micro-clusters according to the identification results, and determines the training vestibular illusion intensity corresponding to each initial micro-cluster, ensuring the accuracy of the training vestibular illusion intensity corresponding to each determined initial micro-cluster. According to each initial micro-cluster, an initial classification model is generated, ensuring the accuracy of the generated initial classification model.

[0255] The number of preset classifiers and the number of preset micro-clusters corresponding to each preset classifier are obtained; for each preset classifier, the first similarity between the physiological parameters and each initial micro-cluster in the initial classification model is calculated, ensuring the accuracy of the calculated first similarity. The target micro-clusters with the number of preset micro-clusters are selected from each initial micro-cluster according to the first similarity, ensuring the accuracy of the determined target micro-clusters. According to the training vestibular illusion intensity corresponding to each target micro-cluster, the training vestibular illusion intensity with the largest number is determined as the standby vestibular illusion recognition intensity corresponding to the preset classifier, ensuring the accuracy of the standby vestibular illusion recognition intensity corresponding to each determined preset classifier. According to the standby vestibular illusion intensity corresponding to each preset classifier, the accuracy of each preset classifier is determined, ensuring the accuracy of the determined accuracy of each preset classifier. The standby vestibular illusion intensity corresponding to the preset classifier with the largest accuracy is determined as the target vestibular illusion intensity, ensuring the accuracy of the determined target vestibular illusion intensity.

[0256] Then, each target micro-cluster corresponding to the preset classifier with the largest accuracy is obtained; according to the relationship between the training vestibular illusion intensity corresponding to each target micro-cluster and the target vestibular illusion intensity, each target micro-cluster is updated to generate the first updated micro-cluster, thereby ensuring the accuracy of the generated first updated micro-cluster. According to each first updated micro-cluster, the initial classification model is updated, thereby ensuring the accuracy of updating the initial classification model, so as to determine the target vestibular illusion intensity corresponding to the physiological parameters generated later based on the updated classification model in the later stage.

[0257] In another alternative implementation, for each preset classifier, if the target microclusters of the preset microcluster quantity are not selected from each initial microcluster according to the first similarity, new microclusters are generated based on physiological parameters, ensuring that the new microclusters can represent the physiological parameters. Then, the second similarity between each new microcluster and each initial microcluster is calculated, ensuring the accuracy of the calculated similarity. According to each second similarity, the training vestibular illusion intensity corresponding to the initial microcluster with the largest second similarity is determined as the candidate vestibular illusion intensity corresponding to the new microcluster, ensuring the accuracy of the determined candidate vestibular illusion intensity corresponding to the new microcluster.

[0258] Then, the new quantity corresponding to the new microcluster is obtained; if the new quantity is greater than the preset new quantity threshold, a backup training data set is obtained; the third similarity between each backup training data and each new microcluster and each initial microcluster is calculated, ensuring the accuracy of the calculated third similarity between each backup training data and each new microcluster and each initial microcluster. According to each third similarity, the initial microcluster or new microcluster corresponding to each backup training data is determined, ensuring the accuracy of the determined third similarity between each backup training data and each new microcluster and each initial microcluster. Then, according to the backup training data corresponding to each initial microcluster, each initial microcluster is updated to generate a second updated microcluster, ensuring the accuracy of the generated second updated microcluster. The backup vestibular illusion intensity of the backup training data corresponding to the new microcluster is compared with the candidate vestibular illusion intensity; if the backup vestibular illusion intensity is consistent with the candidate vestibular illusion intensity, the candidate vestibular illusion intensity is determined as the target vestibular illusion intensity; if the backup vestibular illusion intensity is inconsistent with the candidate vestibular illusion intensity, the candidate vestibular illusion intensity is replaced with the backup vestibular illusion intensity, and the backup vestibular illusion intensity is determined as the target vestibular illusion intensity, thereby ensuring the accuracy of the determined target vestibular illusion intensity. According to the backup training data corresponding to each new microcluster, each new microcluster is updated to generate a third updated microcluster, ensuring the accuracy of the generated third updated microcluster.

[0259] Next, the current first total microcluster quantity corresponding to the candidate classification model is obtained, and the first total microcluster quantity is compared with the preset total microcluster quantity threshold; if the first total microcluster quantity is greater than the preset total microcluster quantity threshold, the fourth similarity between each third updated microcluster and each second updated microcluster is calculated, ensuring the accuracy of the calculated fourth similarity between each third updated microcluster and each second updated microcluster. According to each fourth similarity, each third updated microcluster is merged with the second updated microcluster with the largest fourth similarity to generate a first merged microcluster, ensuring the accuracy of the generated first merged microcluster, reducing the first total microcluster quantity in the candidate classification model, and ensuring that the current microcluster quantity corresponding to each training vestibular illusion intensity is not too small. Based on the first merged microcluster, the candidate classification model is updated to generate a target classification model, ensuring the accuracy of the generated target classification model.

[0260] Then, obtain the current total number of second micro-clusters corresponding to the candidate classification model; if the total number of second micro-clusters is greater than the preset total micro-cluster number threshold and there is no third updated micro-cluster in the candidate classification model, obtain the current number of micro-clusters corresponding to each vestibular illusion intensity in the candidate classification model, ensuring the accuracy of the current number of micro-clusters corresponding to each vestibular illusion intensity obtained in the candidate classification model. Calculate the fifth similarity between the current micro-clusters corresponding to the vestibular illusion intensity with the largest current number of micro-clusters, ensuring the accuracy of the fifth similarity between the current micro-clusters corresponding to the vestibular illusion intensity with the largest current number of micro-clusters calculated. Merge the current micro-clusters with the closest fifth similarity to generate a second merged micro-cluster, ensuring the accuracy of the generated second merged micro-cluster. Update the candidate classification model according to the first merged micro-cluster and the second merged micro-cluster to generate a target classification model. This reduces the total number of micro-clusters in the target classification model and ensures that the current number of micro-clusters corresponding to each vestibular illusion intensity is not too small.

[0261] In this embodiment, a vestibular illusion intensity training device is also provided. This device is used to implement the above embodiments and preferred implementation manners, and those that have been described will not be repeated here. As used hereinafter, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0262] This embodiment provides a vestibular illusion intensity training device, as Figure 4 shown, including:

[0263] An acquisition module 301, configured to acquire test training data to be tested and acquire physiological parameters of a person to be tested for vestibular illusion testing based on the test training data to be tested;

[0264] An identification module 302, configured to identify the physiological parameters to determine the target vestibular illusion intensity corresponding to the person to be tested;

[0265] A comparison module 303, configured to compare the target vestibular illusion intensity with the preset vestibular illusion intensity corresponding to the test training data to be tested;

[0266] An adjustment module 304, configured to adjust the test training data to be tested according to the comparison result.

[0267] The further function descriptions of the above-mentioned various modules and units are the same as those in the corresponding above embodiments and will not be repeated here.

[0268] The vestibular illusion intensity training device in this embodiment is presented in the form of functional units. Here, the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and a memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0269] An embodiment of the present invention further provides an electronic device having the above Figure 4 vestibular illusion intensity training device shown.

[0270] Please refer to Figure 5 , Figure 5 which is a schematic structural diagram of an electronic device provided by an alternative embodiment of the present invention. As shown in Figure 5 , the electronic device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including a high-speed interface and a low-speed interface. Each component communicates with each other using different buses and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed within the electronic device, including instructions stored in the memory or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some alternative embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple electronic devices can be connected, and each device provides some necessary operations (such as an array of servers, a set of blade servers, or a multi-processor system). Figure 5 In

[0271]

[0272]

[0273] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0273] The memory 20 may include a program storage area and a data storage area. The program storage area may store an operating system and application programs required for at least one function. The data storage area may store data created according to the use of the electronic device and the like. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some alternative embodiments, the memory 20 may optionally include a memory remotely provided with respect to the processor 10, and these remote memories may be connected to the electronic device through a network. Examples of the above network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0274] The memory 20 may include a volatile memory, such as a random access memory. The memory may also include a non-volatile memory, such as a flash memory, a hard disk, or a solid-state drive. The memory 20 may further include a combination of the above types of memories.

[0275] The electronic device further includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30, and the output device 40 may be connected through a bus or other means. Figure 5 Taking connection through a bus as an example.

[0276] The input device 30 may receive input digital or character information, and generate key signal inputs related to user settings and function controls of the electronic device, such as a touch screen, a keypad, a mouse, a trackpad, a touchpad, a pointing stick, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (such as an LED), and a haptic feedback device (such as a vibration motor), etc. The above display device includes but is not limited to a liquid crystal display, a light-emitting diode, a display, and a plasma display. In some alternative embodiments, the display device may be a touch screen.

[0277] Embodiments of the present invention also provide a computer-readable storage medium. The method according to the embodiments of the present invention can be implemented in hardware, firmware, or be implemented as computer code that can be recorded on a storage medium, or be implemented as computer code that is originally stored in a remote storage medium or a non-transitory machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the method described herein can be stored as such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code, and when the software or computer code is accessed and executed by the computer, the processor, or the hardware, the method shown in the above embodiments is implemented.

[0278] A part of the present invention can be applied as a computer program product, for example, computer program instructions, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should be able to understand that the forms of existence of computer program instructions in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to the computer.

[0279] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A vestibular illusion intensity training method, characterized in that: The method comprises: Acquiring training data to be tested, and acquiring physiological parameters of the person to be tested for vestibular illusion test based on the training data to be tested; the training data to be tested is used to drive the six-degree-of-freedom motion platform; Acquire an initial training data set, wherein the initial training data set includes a plurality of initial training data, each of which includes a training physiological parameter and a training vestibular illusion intensity corresponding to the training physiological parameter; Inputting the initial training data set into an initial classification network, the initial classification network recognizes each piece of the initial training data, generates a plurality of initial micro-clusters according to the recognition result, and determines the training vestibular illusion intensity corresponding to each of the initial micro-clusters; generating an initial classification model according to each of the initial microclusters; Acquire the number of preset classifiers and the number of preset microclusters corresponding to each of the preset classifiers; the preset classifiers are components of the initial classification model; For each of the preset classifiers, calculating a first similarity between the physiological parameter and each of the initial micro-clusters in the initial classification model; Selecting the preset number of target micro-clusters from each of the initial micro-clusters according to the first similarity; According to the training vestibular illusion strengths corresponding to the target micro-clusters, determining the training vestibular illusion strength with the largest number as the backup vestibular illusion strength corresponding to the preset classifier; Determining the target vestibular illusion intensity corresponding to the person to be tested according to each of the standby vestibular illusion intensities; Comparing the target vestibular illusion intensity with a preset vestibular illusion intensity corresponding to the training data to be tested; According to the comparison result, the training data to be tested is adjusted.

2. The method according to claim 1, characterized in that Determining the target vestibular illusion intensity corresponding to the person to be tested according to each of the standby vestibular illusion intensities includes: Determining the accuracy corresponding to each of the preset classifiers according to the backup vestibular illusion strength corresponding to each of the preset classifiers; The backup vestibular illusion strength corresponding to the preset classifier with the greatest accuracy is determined as the target vestibular illusion strength.

3. The method according to claim 2, characterized in that After determining the backup vestibular illusion strength corresponding to the preset classifier with the greatest accuracy as the target vestibular illusion strength, the method further includes: Acquire each of the target micro-clusters corresponding to the preset classifier with the greatest accuracy; According to the relationship between the training vestibular illusion intensity and the target vestibular illusion intensity corresponding to each target micro-cluster, each target micro-cluster is updated to generate a first updated micro-cluster; The initial classification model is updated according to each of the first updated micro-clusters to generate an updated classification model.

4. The method according to claim 1, characterized in that: The method further comprises: For each of the preset classifiers, if the preset number of target microclusters is not selected from each of the initial microclusters according to the first similarity, generating new microclusters based on the physiological parameters; Calculating a second similarity between the newly added micro-cluster and each of the initial micro-clusters; According to each of the second similarities, the training vestibular illusion strength corresponding to the initial micro-cluster with the largest second similarity is determined as the candidate vestibular illusion strength corresponding to the newly added micro-cluster; The target vestibular illusion intensity corresponding to the physiological parameter is determined according to the candidate vestibular illusion intensity.

5. The method according to claim 4, characterized in that The method of determining the target vestibular illusion intensity corresponding to the physiological parameter according to the candidate vestibular illusion intensity comprises: Obtaining the newly added number corresponding to the newly added micro-cluster; If the newly added quantity is greater than a preset newly added quantity threshold, a spare training data set is obtained; the spare training data set includes a plurality of spare training data, each of which includes a spare physiological parameter and a spare vestibular illusion intensity corresponding to the spare physiological parameter; Calculating a third similarity between each of the spare training data and each of the newly added micro-clusters and each of the initial micro-clusters; Determining the initial micro-cluster or the newly added micro-cluster corresponding to each of the spare training data according to each of the third similarities; updating each of the initial micro-clusters according to the backup training data corresponding to each of the initial micro-clusters to generate a second updated micro-cluster; comparing the backup vestibular illusion strength of the backup training data corresponding to the newly added micro-cluster with the candidate vestibular illusion strength; If the backup vestibular illusion strength is consistent with the candidate vestibular illusion strength, determining the candidate vestibular illusion strength as the target vestibular illusion strength; If the backup vestibular illusion strength is inconsistent with the candidate vestibular illusion strength, replacing the candidate vestibular illusion strength with the backup vestibular illusion strength, and determining the backup vestibular illusion strength as the target vestibular illusion strength; updating each of the newly added micro-clusters according to the spare training data corresponding to each of the newly added micro-clusters to generate a third updated micro-cluster; The initial classification model is updated according to each of the second updated micro-clusters and the third updated micro-clusters to generate a candidate classification model.

6. The method according to claim 5, characterized in that The updating of the initial classification model according to each of the second updated micro-clusters and the third updated micro-clusters to generate a candidate classification model includes: Acquire a first total number of micro-clusters currently corresponding to the candidate classification model; the first total number of micro-clusters includes the number of the second updated micro-clusters and the number of the third updated micro-clusters; comparing the first total number of microclusters with a preset total number of microclusters threshold; If the first total micro-cluster quantity is greater than the preset total micro-cluster quantity threshold, calculating a fourth similarity between each of the third updated micro-clusters and each of the second updated micro-clusters; According to each of the fourth similarities, merging each of the third updated micro-clusters with the second updated micro-cluster having the largest fourth similarity to generate a first merged micro-cluster; Based on the first merged micro-cluster, the candidate classification model is updated to generate a target classification model.

7. The method according to claim 6, characterized in that The updating of the candidate classification model based on the first merged micro-cluster to generate a target classification model includes: Obtaining a second total number of micro-clusters currently corresponding to the candidate classification model; If the second total number of microclusters is greater than the preset total number of microclusters threshold, and the third updated microcluster does not exist in the candidate classification model, obtaining the current number of microclusters corresponding to each vestibular illusion intensity in the candidate classification model; Calculate the fifth similarity between each current micro-cluster corresponding to the vestibular illusion strength with the largest number of current micro-clusters; Merging the current micro-clusters with the closest fifth similarity to generate a second merged micro-cluster; The candidate classification model is updated according to the first merged micro-cluster and the second merged micro-cluster to generate the target classification model.

Images