An index screening method and device for studying low-correlation simulated driving indexes

Abstract

The application discloses an index screening method and device for researching low-correlation simulation driving indexes, and the method comprises the following steps: acquiring a simulation driving index sequence; constructing a matrix; starting to traverse the current matrix, taking out a to-be-screened matrix from the current matrix; determining whether the current to-be-screened matrix meets a correlation screening condition; if not, taking out a next to-be-screened matrix from the current matrix; if yes, making the value of m be 1 greater, determining whether the value of m multiplied by j is less than a second threshold value; when the current matrix is traversed, making the value of j be 1 greater, determining whether the value of m multiplied by j is less than the second threshold value; if the value of m multiplied by j is less than the second threshold value, returning to the step of constructing the matrix; if the value of m multiplied by j is not less than the second threshold value, constructing a matrix according to the values of the latest marked m and j; determining a target matrix in the matrix, and determining a target index in the target matrix.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to an index screening method and device for studying low-correlation simulation driving indexes. BACKGROUND

[0002] With the development of science and technology, the research in various disciplines is increasingly in-depth, and the correlation problem between research indexes gradually emerges. By deeply studying the correlation between indexes, the internal relationship between indexes can be better understood, and more accurate and comprehensive data support can be provided for scientific research.

[0003] In a simulation driving scene, the number of simulation driving indexes that can be used for research may be hundreds or thousands. However, the prior art cannot determine which indexes need to be selected from these indexes for research when facing such a number of indexes, which also leads to slow progress in subsequent research on indexes.

[0004] Therefore, how to design an index screening method that can quickly select the most meaningful indexes from a large number of indexes as the research object in the subsequent research has become a problem to be solved in the field. SUMMARY

[0005] Therefore, in a first aspect, the present application provides an index screening method for studying low-correlation simulation driving indexes, which comprises:

[0006] S100, acquiring a simulation driving index sequence;

[0007] S200, sequentially selecting simulation driving indexes in the simulation driving index sequence to construct m matrices; wherein m represents the order of the matrix, j represents a natural number starting from 1, and represents the combination of constructing a matrix by taking m simulation driving indexes from the first m times j simulation driving indexes in the simulation driving index sequence.

[0008] S300, starting to traverse the current m matrices, taking one to-be-screened matrix from the current m matrices; S400, determining whether the current to-be-screened matrix meets a correlation screening condition, the correlation screening condition being that the absolute value of the correlation coefficient between each index in the matrix is less than a first threshold;

[0009] If not, return to step S300, and take the next to-be-screened matrix from the current m matrices;

[0010]

[0011] ​​​​​If satisfied, then execute step S410, mark the current values ​​of m and j, increment the current value of m by 1, and then determine whether the current value of m multiplied by j is less than the second threshold.

[0012] After traversing the current... matrix, and currently If no matrix satisfying the correlation screening condition is found in the matrix, then

[0013] S420. After incrementing the current value of j by 1, determine whether the current value of m multiplied by j is less than the second threshold.

[0014] If the current value of m multiplied by j is less than the second threshold, return to step S200 and continue to construct based on the current values ​​of m and j. A matrix;

[0015] If the current value of m multiplied by j is not less than the second threshold, then proceed to step S500: construct based on the values ​​of the most recently marked m and j. A matrix;

[0016] S600, in the The target matrix is ​​determined from the matrix, and the target index is determined from the target matrix.

[0017] Preferably, step S100 includes:

[0018] S110, generate a simulated driving index matrix based on the simulated driving indexes and the correlation coefficients between the indexes;

[0019] S120, in the simulated driving index matrix, count the number of simulated driving indexes whose absolute values ​​of the correlation coefficients between the corresponding indices are less than the first threshold, and sort the simulated driving indexes in descending order according to the number to obtain a simulated driving index sequence.

[0020] More preferably, before step S110, the method further includes:

[0021] S111. Preprocess the simulated driving indicators and the correlation coefficients between the indicators.

[0022] Preferably, if the current matrix to be screened does not meet the relevance screening criteria, the method further includes:

[0023] S421. Place the matrices that do not meet the correlation filtering conditions into the exclusion set.

[0024] More preferably, step S600 includes:

[0025] S610, in the above Filter out all first target matrices that meet the relevance filtering criteria from the matrices;

[0026] S620. Determine the second objective matrix among all the first objective matrices that has the smallest sum of the absolute values ​​of the correlation coefficients between the indicators;

[0027] S630. Obtain the target index from the second target matrix.

[0028] More preferably, prior to step S610, the method further includes:

[0029] S611, in the above The matrices in the exclusion set are excluded from the matrix.

[0030] A further preferred embodiment is characterized in that the initial value of m is 3.

[0031] Secondly, this application also provides an index screening device for studying low-correlation simulated driving indicators, the device comprising:

[0032] The acquisition module is configured to acquire a sequence of simulated driving indicators;

[0033] The processing module is configured as follows:

[0034] In the sequence of simulated driving indicators, simulated driving indicators are selected sequentially to construct... There are *m* matrices; where *m* represents the matrix order and *j* represents a natural number starting from 1. This represents the combination of m simulated driving indicators taken from the first m multiplied by j simulated driving indicators in the simulated driving indicator sequence to construct a matrix;

[0035] Start traversing the current... Matrix, from the current One matrix to be screened is taken from each matrix;

[0036] Determine whether the current matrix to be screened meets the correlation screening condition, wherein the correlation screening condition is: the absolute value of the correlation coefficient between each indicator in the matrix is ​​less than a first threshold.

[0037] If not satisfied, then from the current... Take the next matrix to be screened from the first matrix;

[0038] If satisfied, then mark the current values ​​of m and j, increment the current value of m by 1, and then determine whether the current value of m multiplied by j is less than the second threshold.

[0039] After traversing the current... matrix, and currently If there is no matrix to be screened that satisfies the correlation screening condition among the matrices, then the current value of j is incremented by 1, and it is determined whether the current value of m multiplied by j is less than the second threshold.

[0040] If the current value of m multiplied by j is less than the second threshold, then continue to construct based on the current values ​​of m and j. A matrix; if the current m multiplied by j is not less than the second threshold, then

[0041] Construct based on the most recently tagged values ​​of m and j A matrix;

[0042] The determination module is configured in the above. The target matrix is ​​determined from the matrix, and the target index is determined from the target matrix.

[0043] The index screening method provided in this application for studying low-correlation simulated driving indicators constructs and traverses an index matrix based on a simulated driving indicator sequence, and finally selects the target indicators from these matrices. This approach can quickly select the most meaningful simulated driving indicators from a large number of simulated driving indicators as the research objects for subsequent studies, providing a solid foundation for further research on the selected simulated driving indicators.

[0044] Other features and advantages of this application will be described in detail in the following detailed description section. Attached Figure Description

[0045] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application, and the illustrative embodiments and descriptions thereof are used to explain this application. In the drawings:

[0046] Figure 1 A flowchart illustrating a preferred embodiment of this application for a method of selecting indicators for studying low-correlation simulated driving indicators;

[0047] Figure 2 A schematic diagram of a portion of the simulated driving index matrix according to a preferred embodiment of this application;

[0048] Figure 3 This is a schematic diagram of the structure of an index screening device for studying low-correlation simulated driving indicators according to a preferred embodiment of this application. Detailed Implementation

[0049] The technical solution of this application will now be described in detail with reference to the accompanying drawings and embodiments.

[0050] The correlation between indicators refers to a certain relationship or dependency between two or more indicators. This relationship can be direct or indirect, and may manifest as positive correlation, negative correlation, or no correlation. The correlation between indicators can be analyzed and quantified using various methods, such as the Pearson correlation coefficient and the Spearman rank correlation coefficient. Studying low correlation between indicators helps reveal which indicators are statistically independent, while studying high correlation has the opposite effect. Both studies can be used to more accurately understand the individual roles and influencing factors of each indicator.

[0051] In the field of driving simulation, the evaluation metrics involved are complex, diverse, and numerous. For example, the Flamba driving simulator provides access to hundreds of driving simulation metrics. Researchers often need to precisely select the metrics that make key contributions to their research from this vast pool of data.

[0052] Based on this, this application first proposes a method for screening indicators to study low-correlation simulated driving indicators, such as... Figure 1 As shown, it includes S100-S600:

[0053] S100, obtain the simulated driving indicator sequence;

[0054] Specifically, before filtering the simulated driving indicators, a sequence of simulated driving indicators needs to be constructed. The indicators in this sequence are then sorted in descending order based on the number of indicators whose absolute correlation coefficients are less than a first threshold. An absolute value of the correlation coefficient less than the first threshold indicates that the two simulated driving indicators corresponding to that correlation coefficient have a low correlation. This first threshold can be set as needed.

[0055] Understandably, the correlation coefficients between simulated driving indicators can be expressed in various types, such as Pearson correlation coefficient, Spearman rank correlation coefficient, etc. The specific method to choose depends on the characteristics of the data and the purpose of the analysis, and this application does not limit it.

[0056] In a specific embodiment, the specific steps of S100 include S111-S120:

[0057] S111, preprocessing the simulated driving indicators and the correlation coefficients between the indicators;

[0058] Specifically, preprocessing includes steps such as data cleaning, missing value handling, outlier handling, dimensionless processing, and standardization.

[0059] S110, generate a simulated driving index matrix based on the simulated driving indexes and the correlation coefficients between the indexes;

[0060] Specifically, if N simulated driving indicators are obtained, an N×N simulated driving indicator matrix can be formed. The first row and first column of the matrix are the simulated driving indicators, and the element values ​​in the matrix are the correlation coefficients between the indicators.

[0061] In a like Figure 2 In the simulated driving indicator matrix example shown, the simulated driving indicators include "time, timestamp, trafficTime, scenarioTime, type, model, ID, customID, description, positionX, positionY, positionZ, yawAngle, pitchAngle, rollAngle, directionX, directionY, directionZ, bodyPitchAngle, bodyRollAngle, RPM, transmissionState, gearNumber," etc. The correlation coefficient between indicators ranges from -1 to 1, with smaller values ​​indicating weaker correlation. A positive value indicates a positive correlation between two simulated driving indicators, while a negative value indicates a negative correlation.

[0062] S120, In the simulated driving index matrix, count the number of simulated driving indexes whose absolute values ​​of the correlation coefficients between the corresponding indices are less than the first threshold, and sort the simulated driving indices in descending order according to the number of indices to obtain the simulated driving index sequence.

[0063] Specifically, count the number of simulated driving indicators whose absolute values ​​of the correlation coefficients between the indicators are less than the first threshold, and then sort all simulated driving indicators in descending order based on this number.

[0064] In a specific embodiment of the simulated driving index sequence, the first threshold is 0.4, and some values ​​in the simulated driving index sequence include: "('trailerWheelbase', 104), ('appliedThrottle', 103), ('lightState', 102), ('rawThrottle', 99), ('latestRoad', 96), ('clutch', 91), ('trailer', 91)"

[0065] ('dragForce', 90°), ('trailerAngle', 90°)

[0066] ('distanceAlongRoad', 90), ('model', 89), ('throttle', 89), ('rawBrake', 88)". In each parenthesis, the first string is the name of the simulated driving indicator (the content within the parentheses is the Chinese translation of the indicator), and the second is the number of correlation coefficients between that simulated driving indicator and other indicators that are within ±0.4. For example, the indicator 'trailerWheelbase' has 104 correlation coefficients between it and other indicators within ±0.4.

[0067] S200, in the sequence of simulated driving indicators, selects simulated driving indicators sequentially to construct... A matrix;

[0068] Specifically, this application sorts the indicators in the simulated driving indicator sequence, selects indicators sequentially from front to back, and constructs a set of correlation coefficient matrices for the simulated driving indicators step by step. Then, it searches for matrices that meet the requirements within this set of correlation coefficient matrices. The number of matrices in this set is... Where m represents the matrix order, and j represents a natural number starting from 1. This represents the combination of m simulated driving indicators taken from the first m multiplied by j simulated driving indicators in the simulated driving indicator sequence to construct a matrix.

[0069] Since a 2D matrix is ​​very simple and easy to find, the initial value of m in this application is "3", and the initial value of j is "1", meaning the analysis starts with a 3D matrix. When m=3 and j=1, the top m multiplied by j simulated driving indicators are selected. From these m multiplied by j simulated driving indicators, any m indicators are selected for combination. Each combination corresponds to a matrix, and all combinations form a matrix set. This matrix set includes... A matrix.

[0070] S300, begin iterating through the current... Matrix, from the current One matrix to be screened is taken from each matrix;

[0071] Specifically, from the current After retrieving one matrix to be screened from the given matrices, step S400 is executed to determine whether the current matrix to be screened meets the relevance screening criteria. If the current matrix meets the relevance screening criteria... If all matrices have been traversed and no matrix that meets the correlation screening criteria is found, then proceed to step S420.

[0072] Understandably, start iterating through the current... Having a matrix does not necessarily mean that the current matrix has been traversed. Matrix. If in the current Once a matrix that meets the relevance screening criteria is found in the matrix, the traversal ends, thus improving screening efficiency.

[0073] S400, determine whether the current matrix to be screened meets the relevance screening criteria;

[0074] Specifically, in this application, the correlation screening condition is: the absolute value of the correlation coefficient between each indicator in the matrix is ​​less than a first threshold, which corresponds to the sorting requirements of the indicators in the simulated driving indicator sequence. Determining whether the current matrix to be screened meets the correlation screening condition is equivalent to determining whether the absolute value of each element in the matrix is ​​less than the first threshold.

[0075] If the matrix to be screened meets the relevance screening criteria, then execute S410. If the matrix to be screened does not meet the relevance screening criteria, then the current matrix to be screened is placed in the exclusion set, and then... Take the next matrix to be screened from the previous matrix, and continue to check whether the absolute value of each element in the next matrix is ​​less than the first threshold, that is, return to step S300. If the current matrix has been traversed, proceed to step S300. If, after checking all matrices, no matrix that meets the correlation screening criteria is found, then proceed to step S420.

[0076] S410. If the current matrix to be screened meets the correlation screening condition, then increment the current value of m by 1 and determine whether the value of m multiplied by j is less than the second threshold.

[0077] Specifically, if the absolute value of each element in the current matrix is ​​less than the first threshold, then the current values ​​of m and j are marked. Next, the current value of m is incremented by 1, and it is determined whether the product of the incremented m and j is less than the second threshold. If it is less, the process returns to step S200 and continues to construct the matrix based on the current values ​​of m and j. A matrix. If it is not less than, then proceed to step S500.

[0078] Here, the purpose of determining whether the value of m multiplied by j is less than the second threshold is to establish a computational upper limit to conserve computing power. Therefore, the second threshold can be set based on factors such as computing power, data volume, and target requirements.

[0079] S420, When the current iteration is complete... matrix, and the current If there is no matrix to be screened that satisfies the correlation screening condition among the matrices, then the current value of j is incremented by 1, and it is determined whether the current value of m multiplied by j is less than the second threshold.

[0080] Specifically, after traversing the current... After multiplying the matrix by 1, increment the current value of j by 1 to determine if the current value of m multiplied by j is less than the second threshold. If it is less, it means the computational load is within the allowable range of computing power, so return to execute S200 and continue to construct based on the current values ​​of m and j. A matrix. If it is not less than, then proceed to step S500.

[0081] In other words, in steps S410 and S420 above, whenever the value of m or j is incremented by 1, a computing power threshold judgment is required to avoid a situation where the computing power cannot be sustained as the number of screening matrices increases during the index screening cycle.

[0082] Understandably, at this point, the values ​​of m and / or j are different from their initial values, and the constructed... The matrix is ​​not the initial set of matrices. Through repeated looping searches and marking, this application can find the highest-order simulated driving indicator matrix that meets the correlation screening conditions from a large number of simulated driving indicators within the limits of computing power. That is, it can find as many simulated driving indicators as possible that best meet the requirements.

[0083] S500. If the current value of m multiplied by j is not less than the second threshold, then construct based on the values ​​of the most recently labeled m and j. A matrix;

[0084] Specifically, if the current value of m multiplied by j is not less than the second threshold, it means that the current computational load is at the upper limit allowed by computing power, and the number of matrices in the matrix set should not be increased further. That is, there is no need to increase the value of m or j to construct a new matrix set. Therefore, the values ​​of M and J are determined based on the most recently marked values ​​of m and j, and the matrix set is constructed. A set of matrices is used as a basis for determining the final simulated driving indicators in subsequent steps.

[0085] In a specific embodiment, the specific steps of S500 include S510-S520:

[0086] S510, if the current value of m multiplied by j is not less than the second threshold, then confirm the value of the most recently marked m and the value of j;

[0087] Specifically, if the current value of m multiplied by j is not less than the second threshold, then the node is backtracked to determine the values ​​of m and j in the most recently marked node.

[0088] S520 is constructed based on the most recently tagged values ​​of m and j. A matrix;

[0089] Specifically, here the value of M is equal to the value of the most recently labeled m, and the value of J is equal to the value of the most recently labeled j. This represents the combination of M simulated driving indicators taken from the first M multiplied by J simulated driving indicators in the simulated driving indicator sequence to construct a matrix.

[0090] S600, in From the given matrices, determine the target matrix, and from the target matrix, determine the target indicators;

[0091] In a specific embodiment, the specific steps of S600 include S611-S630:

[0092] S611, in Exclude matrices from the exclusion set in the matrix;

[0093] Specifically, in the constructed Before filtering through the matrices, you can first remove the matrices from the already labeled exclusion set. The data in these matrices do not meet the relevance filtering criteria, so these matrices should be removed first. Eliminating matrices in a concentrated manner facilitates faster subsequent filtering.

[0094] S610, in Filter out all first target matrices that meet the relevance filtering criteria from the matrices;

[0095] Specifically, traversal A matrix is ​​selected, and the matrix with the absolute value of the correlation coefficient between indicators within the matrix being less than a first threshold is selected as the first target matrix.

[0096] S620, determine the second objective matrix that minimizes the sum of the absolute values ​​of the correlation coefficients between indicators among all the first objective matrices;

[0097] Specifically, the matrix with the smallest sum of absolute values ​​of correlation coefficients among all first objective matrices is determined as the second objective matrix. That is, the second objective matrix is ​​the matrix with the smallest average correlation coefficient among all first objective matrices.

[0098] S630, obtain the target indicators from the second target matrix;

[0099] Specifically, the simulated driving indicators in the second target matrix are used as target indicators to obtain all the simulated driving indicators that meet the correlation coefficient requirements in the final selection.

[0100] Understandably, if the initial number of simulated driving indicators is N, forming an N×N simulated driving indicator matrix, then the largest M×M submatrix that meets the requirements can be found in this N×N matrix. The M simulated driving indicators corresponding to this submatrix are the set of the largest range of simulated driving indicators that we want to study.

[0101] In a specific embodiment, the second threshold is "30". Currently, m = 8 and j = 4. At this time, m × j = 32 > 30. Assume that the most recently marked value of m is 7 and the value of j is 3. Then, M = 7 and J = 3. The top M × J = 21 simulated driving indicators are selected to construct the system. Assuming there are 7100 matrices in the exclusion set, after removing the 7100 matrices from the exclusion set, 17000 matrices remain. From these 17000 matrices, five matrices are selected as the first target matrices, where the absolute value of the correlation coefficient between indicators is less than a first threshold. Finally, among these five first target matrices, the second target matrix with the smallest sum of absolute values ​​of the correlation coefficients between indicators contains the following simulated driving indicators: "lightState (lighting status), rawThrottle (raw throttle), latestRoad (current road), trailer (trailer), dragForce (drag), model (model), centerOfGravityHeight (center of gravity height)". These seven simulated driving indicators are the final selected target indicators.

[0102] The index screening method provided in this application for studying low-correlation simulated driving indicators constructs and traverses an index matrix based on a simulated driving indicator sequence, and finally selects the target indicators from these matrices. This approach can quickly select the most meaningful simulated driving indicators from a large number of simulated driving indicators as the research objects for subsequent studies, providing a solid foundation for further research on the selected simulated driving indicators.

[0103] Secondly, this application also provides an index screening device for studying low-correlation simulated driving indicators. The device is used to implement the index screening method for studying low-correlation simulated driving indicators described in the first aspect above, such as... Figure 3 As shown, the device includes:

[0104] The acquisition module 310 is configured to acquire a sequence of simulated driving indicators.

[0105] Processing module 320 is configured to:

[0106] In the sequence of simulated driving indicators, simulated driving indicators are selected sequentially to construct... There are *m* matrices; where *m* represents the matrix order and *j* represents a natural number starting from 1. This represents the combination of m simulated driving indicators taken from the first m multiplied by j simulated driving indicators in the simulated driving indicator sequence to construct a matrix;

[0107] Start traversing the current... Matrix, from the current One matrix to be screened is taken from each matrix;

[0108] Determine whether the current matrix to be screened meets the correlation screening condition, wherein the absolute value of the correlation coefficient between each indicator in the matrix is ​​less than a first threshold.

[0109] If not satisfied, then from the current... Take the next matrix to be screened from the first matrix;

[0110] If the conditions are met, then after marking the current values ​​of m and j, increment the current value of m by 1, and then determine whether the current value of m multiplied by j is less than the second threshold.

[0111] After traversing the current... matrix, and currently If there is no matrix to be screened that satisfies the correlation screening condition among the matrices, then the current value of j is incremented by 1, and it is determined whether the current value of m multiplied by j is less than the second threshold.

[0112] If the current value of m multiplied by j is less than the second threshold, then continue to construct based on the current values ​​of m and j. A matrix; if the current m multiplied by j is not less than the second threshold, then

[0113] Construct based on the most recently tagged values ​​of m and j A matrix;

[0114] Determine module 330, configured in the The target matrix is ​​determined from the matrix, and the target index is determined from the target matrix.

[0115] Other preferred embodiments of the index screening device for studying low-correlation simulated driving indicators provided in this application, the technical problems they can solve, and the technical effects they can achieve are the same as the index screening method for studying low-correlation simulated driving indicators described above, and will not be repeated here.

[0116] The preferred embodiments of this application have been described in detail above. However, this application is not limited to the specific details of the above embodiments. Within the scope of the technical concept of this application, various simple modifications can be made to the technical solution of this application, and these simple modifications all fall within the protection scope of this application.

[0117] It should also be noted that the various specific technical features described in the above embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, this application will not describe the various possible combinations separately.

[0118] Furthermore, various different embodiments of this application can be combined in any way, as long as they do not violate the spirit of this application, they should also be regarded as the content disclosed by this invention.

Claims

1. A method for selecting indicators to study low-correlation simulated driving indicators, characterized in that, The method includes the following steps: S100, Obtain the simulated driving indicator sequence; S200. Select simulated driving indicators sequentially from the simulated driving indicator sequence to construct... There are *m* matrices; where *m* represents the matrix order and *j* represents a natural number starting from 1. This represents the combination of m simulated driving indicators taken from the first m multiplied by j simulated driving indicators in the simulated driving indicator sequence to construct a matrix; S300, Start iterating through the current... Matrix, from the current One matrix to be screened is taken from each matrix; S400. Determine whether the current matrix to be screened meets the correlation screening condition, wherein the correlation screening condition is: the absolute value of the correlation coefficient between each indicator in the matrix is ​​less than the first threshold. If the condition is not met, return to step S300, starting from the current state. Take the next matrix to be screened from the first matrix; If satisfied, then execute step S410, mark the current values ​​of m and j, increment the current value of m by 1, and then determine whether the current value of m multiplied by j is less than the second threshold. After traversing the current... matrix, and currently If no matrix among the matrices satisfies the aforementioned correlation screening criteria, then... S420. After incrementing the current value of j by 1, determine whether the current value of m multiplied by j is less than the second threshold. If the current value of m multiplied by j is less than the second threshold, return to step S200 and continue to construct based on the current values ​​of m and j. A matrix; If the current value of m multiplied by j is not less than the second threshold, then proceed to step S500: construct based on the values ​​of the most recently marked m and j. A matrix; where the values ​​of M and J are the values ​​of the most recently labeled m and j, respectively; S600, in the... The target matrix is ​​determined from the matrix, and the target index is determined from the target matrix.

2. The method according to claim 1, characterized in that, Step S100 includes: S110, generate a simulated driving index matrix based on the simulated driving indexes and the correlation coefficients between the indexes; S120, in the simulated driving index matrix, count the number of simulated driving indexes whose absolute values ​​of the correlation coefficients between the corresponding indices are less than the first threshold, and sort the simulated driving indexes in descending order according to the number to obtain a simulated driving index sequence.

3. The method according to claim 2, characterized in that, Prior to step S110, the method further includes: S111. Preprocess the simulated driving indicators and the correlation coefficients between the indicators.

4. The method according to claim 1, characterized in that, If the current matrix to be screened does not meet the relevance screening criteria, the method further includes: S421. Place the matrices that do not meet the correlation filtering conditions into the exclusion set.

5. The method according to claim 4, characterized in that, Step S600 includes: S610, in the above Filter out all first target matrices that meet the relevance filtering criteria from the matrices; S620. Determine the second objective matrix among all the first objective matrices that has the smallest sum of the absolute values ​​of the correlation coefficients between the indicators; S630. Obtain the target index from the second target matrix.

6. The method according to claim 5, characterized in that, Prior to step S610, the method further includes: S611, in the above The matrices in the exclusion set are excluded from the matrix.

7. The method according to any one of claims 1-6, characterized in that, The initial value of m is 3.

8. An index screening device for studying low-correlation simulated driving indicators, characterized in that, The device includes: The acquisition module is configured to acquire a sequence of simulated driving indicators; The processing module is configured as follows: In the sequence of simulated driving indicators, simulated driving indicators are selected sequentially to construct... There are *m* matrices; where *m* represents the matrix order and *j* represents a natural number starting from 1. This represents the combination of m simulated driving indicators taken from the first m multiplied by j simulated driving indicators in the simulated driving indicator sequence to construct a matrix; Start traversing the current... Matrix, from the current One matrix to be screened is taken from each matrix; Determine whether the current matrix to be screened meets the correlation screening condition, wherein the correlation screening condition is: the absolute value of the correlation coefficient between each indicator in the matrix is ​​less than a first threshold. If not satisfied, then from the current... Take the next matrix to be screened from the first matrix; If satisfied, then mark the current values ​​of m and j, increment the current value of m by 1, and then determine whether the current value of m multiplied by j is less than the second threshold. After traversing the current... matrix, and currently If there is no matrix to be screened that satisfies the correlation screening condition among the matrices, then the current value of j is incremented by 1, and it is determined whether the current value of m multiplied by j is less than the second threshold. If the current value of m multiplied by j is less than the second threshold, then continue to construct based on the current values ​​of m and j. A matrix; if the current m multiplied by j is not less than the second threshold, then Construct based on the most recently tagged values ​​of m and j There are several matrices; where the values ​​of M and J are the values ​​of the most recently labeled m and j, respectively. The determination module is configured in the above. The target matrix is ​​determined from the matrix, and the target index is determined from the target matrix.

Images