A vehicle speed deviation analysis method based on a brain-like cognitive computing model

Abstract

The application discloses a kind of based on brain-like cognitive computing model's vehicle speed deviation analysis method, belong to vehicle speed deviation analysis technical field, including the following steps: S1, build low-grade highway simulation environment to carry out driving simulation experiment;S2, establish visual road environment model;S3, the electroencephalogram data of driver is preprocessed, and carries out spectrum analysis, and the relationship between environment and electroencephalogram signal intensity is established on this basis;S4, based on QN-MHP brain-like cognitive computing model, the LMM model is established using the signal intensity relationship of each brain area, and the deviation between the expected speed of driver and actual speed is analyzed by this.The application uses the above-mentioned based on brain-like cognitive computing model's vehicle speed deviation analysis method, quantifies the perception situation of driver to road environment condition, has important role to the perception, cognitive characteristics when determining the expected speed of driver, and its influence on the difference between expected speed and actual running speed, and has important significance to the improvement of road traffic safety.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle speed deviation analysis, and in particular to a vehicle speed deviation analysis method based on a brain-like cognitive computing model. Background Art

[0002] Low-grade highways play a crucial role in transportation in mountainous and rural areas. Due to their lower design requirements and complex driving environments, drivers often experience significant discrepancies between their desired and actual speeds on these roads, severely impacting driving safety. During this process, the driver's visual perception of the road environment influences their desired speed, which in turn affects their actual speed decisions.

[0003] Vehicle speed is a key factor influencing the occurrence and consequences of road traffic accidents, and the driver's desired speed determines their speed decisions. The desired speed is the speed a driver chooses to operate in free-flow conditions, unconstrained by linear constraints; or the maximum "safe" speed a driver desires to achieve when the vehicle is unconstrained or largely unconstrained by other vehicles. When drivers misjudge the safe operating speed permitted by road conditions, they subjectively generate an "optimal operating speed" that is inconsistent with road conditions, controlling their speed based on an inappropriate desired speed. This speed often exceeds the road's design speed, creating safety hazards and even causing traffic accidents.

[0004] Due to various factors, including the road environment, a driver's desired speed in certain road conditions can deviate from their actual speed, which can affect driving safety to a certain extent. Brain-inspired technology can provide a deeper understanding of the mechanisms of driver behavior, microscopically analyzing the process of "perceiving the road environment, making speed decisions, and implementing speed control." Currently, extensive research has been conducted to calibrate and predict a driver's desired speed, and quantitative models based on visual road environment diagrams have also been applied to evaluate road safety. However, existing research has not quantified how drivers perceive and understand the road environment using brain-inspired cognitive computing models, nor has it analyzed the deviation between the driver's desired speed and actual speed based on this. Summary of the Invention

[0005] The purpose of the present invention is to provide a vehicle speed deviation analysis method based on a brain-like cognitive computing model. Taking the deviation between the driver's expected speed and the actual speed as an indicator, the method quantifies the driver's perception of the road environment conditions. It plays an important role in analyzing the driver's perception and cognitive characteristics when determining the expected speed, and its impact on the difference between the expected speed and the actual operating speed, and is of great significance to improving road traffic safety.

[0006] To achieve the above objectives, the present invention provides a vehicle speed deviation analysis method based on a brain-inspired cognitive computing model, which is characterized by comprising the following steps:

[0007] S1. Build a low-grade highway simulation environment to conduct a driving simulation experiment, and collect the driver's expected speed, actual speed, and EEG data during the experiment;

[0008] S2. Based on schema theory, digitally analyze the driver's visual driving environment and establish a visual road environment model;

[0009] S3. Preprocess the driver's EEG data and perform spectrum analysis, based on which the relationship between the environment and EEG signal intensity is established;

[0010] S4. Based on the QN-MHP brain-like cognitive computing model, an LMM model is established using the relationship between the signal strength of each brain region to analyze the deviation between the driver's expected speed and actual speed.

[0011] Preferably, in step S1, a traffic scene simulation software is used to build a low-grade highway simulation environment, including three different scenes: towns, plains, and mountainous areas;

[0012] The EEG data were acquired by recording the EEG signals using Brain Vision Recorder software. The sampling rate was set to 1000 Hz, the maximum electrode resistance was set to 40 µΩ, and the electrode position numbering was based on the international standard system.

[0013] Speed ​​deviation It is obtained by subtracting the actual vehicle speed V from the expected vehicle speed Vp.

[0014] Preferably, the visual road environment model in step S2 is divided into five different layers, including a visual road linear layer, a dynamic visual layer, a visual semantic layer, a visual depth layer, and a visual sensitivity layer;

[0015] The boundaries of the visual road alignment in the visual road alignment layer are fitted using Catmull-Rom spline curves. The dynamic visual layer represents the dynamic changes in the driver's field of view as a function of speed. The driver's field of view is negatively correlated with vehicle speed, and their visual attention environment is compressed to the center of the road as vehicle speed increases.

[0016] The visual semantic layer uses different colors to reflect different parts of the road environment perceived by the driver. The visual semantic layer includes the following different components: road, forest, sky, desert, traffic signs, protective measures, and cliffs. The area ratio of each component in the driver's visual perception of the road environment is used as a quantitative parameter.

[0017] The visual depth layer reflects the driver's perception of the depth information of the road environment. The average value of the driver's field of view is extracted from the visual depth layer and the dynamic visual layer as a quantitative parameter.

[0018] The visual sensitivity layer reflects the distribution of sensitive areas in each road environment area under the driver's visual perception. The calculation results are as follows:

[0019] ;

[0020] ;

[0021] in, Represents the heat map formed by the gradient-weighted class activation mapping algorithm; Representative feature maps; Representative The weight corresponding to each feature map; is the rectified linear unit; , is the size of the feature map; ; Indicates the In the feature map, Rank The pixel value of the column.

[0022] Preferably, the pre-processing method in step S3 is as follows:

[0023] Filtering, extracting useful information and removing unnecessary noise and interference bandpass filtering;

[0024] Resampling, in order to speed up the batch data processing, the sampling rate of the EEG signal is resampled from the original frequency of 1000Hz to 500Hz;

[0025] Independent component analysis, performed on EEG signals to separate the components of the mixed signal in order to analyze the activities of different neural sources;

[0026] Remove artifacts, remove electrooculogram (EOG), electromyography (EMG), electrode noise, and power supply interference, and improve the purity and accuracy of EEG signals for better analysis and interpretation of brain activity.

[0027] Remove extreme values ​​and remove extreme values ​​in EEG signals to ensure the accuracy and reliability of the data;

[0028] Rereferencing changes the reference base of the EEG signal to improve signal interpretation and analysis.

[0029] Preferably, the steps of spectrum analysis are as follows:

[0030] S301, time domain signal conversion, converting the EEG signal from the time domain to the frequency domain;

[0031] S302, calculating the power spectrum of the converted EEG signal, and determining the energy distribution at each frequency by calculating the power spectrum of the EEG signal, in decibels;

[0032] S303, using data windowing and averaging techniques to process the EEG signal to improve the accuracy and reliability of spectrum estimation, wherein the EEG signal of the driver during the entire driving process is located at selected points, and a 1-second time period is selected at each point for spectral band analysis;

[0033] S304, Spectral band analysis, analyze the power spectrum of different frequency bands, study the changes and characteristics of EEG signals in different frequency bands, and select Wave, Wave, Wave, The power spectrum intensity values ​​of these frequency bands are averaged to obtain the driver's brain activity.

[0034] Preferably, in step S4, a linear mixed effects regression model LMM is used for data analysis, and its expression is as follows:

[0035] ;

[0036] in, is the vector of dependent variables; is the fixed-effects score design matrix; is the parameter vector of the fixed effects; is the design matrix of random effects; is the parameter vector of the random effects; is the vector of random error terms; expresses the impact of fixed effects; represents the influence of random effects.

[0037] Preferably, in the LMM model, the linear mixed model is constructed with the help of the lme4 package in the R language. During the modeling process, the random error term is set to the subject number participating in the simulation experiment and the different visual road environment scene points, representing the random effects between different subjects and different road scenes, respectively;

[0038] Due to the differences in the size and distribution of each data, the maximum and minimum normalization method is used to scale the data to the range of [0,1]. The processing method is as follows:

[0039] ;

[0040] in, is the processed data point; is the data point in the original set; is the minimum value in the data set; is the maximum value in the data set;

[0041] For speed deviation Since there are both positive and negative values, the Z-Score standardization method is used to convert the data into a form with a mean of 0 and a standard deviation of 1. The Z-Score standardization method is as follows:

[0042] ;

[0043] in, is the processed data point; is the mean of the original set; is the standard deviation of the original set.

[0044] Preferably, the expected speed Vp, actual speed V, speed deviation are calculated based on the fitting results of the linear mixed model. predictions.

[0045] Therefore, the present invention adopts the above-mentioned vehicle speed deviation analysis method based on the brain-like cognitive computing model to conduct an in-depth analysis of the driver's visual perception information presented by the visual road environment model, explore the impact of different road environment elements on the driver's perception of the road environment, and thus analyze its definition of the expected vehicle speed.

[0046] Based on the QN-MHP model, a linear mixed-effect model is constructed to model the driver's visual perception in different road environments to reflect the perception-cognition-response process. This is done to interpret the impact of various visual elements on the driver's cognitive behavior and analyze the relationship between the various parts of the brain-like cognitive computing model.

[0047] From a visual perspective, the relationship between the deviation between the driver's expected speed and the actual operating speed and the influencing factors is interpreted, thereby contributing to the optimization of road geometry and road environment.

[0048] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 1 is a schematic diagram of an overall method of an embodiment of a vehicle speed deviation analysis method based on a brain-inspired cognitive computing model of the present invention;

[0050] Figure 2 This is a visual road environment model diagram of an embodiment of a vehicle speed deviation analysis method based on a brain-inspired cognitive computing model of the present invention;

[0051] Figure 3 This is a construction diagram of a visual road linear layer in an embodiment of a vehicle speed deviation analysis method based on a brain-like cognitive computing model of the present invention;

[0052] Figure 4 It is the visual semantic layer and visual depth layer of an embodiment of a vehicle speed deviation analysis method based on a brain-like cognitive computing model of the present invention;

[0053] Figure 5 It is a visual sensitivity layer of an embodiment of a vehicle speed deviation analysis method based on a brain-like cognitive computing model of the present invention;

[0054] Figure 6 It is a visual processing architecture of the QN-MHP model of an embodiment of a vehicle speed deviation analysis method based on a brain-like cognitive computing model of the present invention;

[0055] Figure 7 This is a prediction result diagram of the expected vehicle speed Vp according to an embodiment of a vehicle speed deviation analysis method based on a brain-like cognitive computing model of the present invention; Figure 7 (a) is a scatter plot of the server F’s prediction results for the expected vehicle speed Vp; Figure 7 (b) is a scatter plot of the prediction results of the visual road environment model for the expected vehicle speed Vp; Figure 7 (c) is the histogram of the prediction results of the server F for the expected vehicle speed Vp; Figure 7 (d) is the histogram of the prediction results of the visual road environment model for the expected vehicle speed Vp;

[0056] Figure 8 This is a prediction result diagram of the actual vehicle speed V according to an embodiment of a vehicle speed deviation analysis method based on a brain-inspired cognitive computing model of the present invention; Figure 8 (a) is a scatter plot of the prediction results of server Z for the actual vehicle speed V; Figure 8 (b) is a scatter plot of the prediction results of the visual road environment model for the actual vehicle speed V; Figure 8 (c) is the histogram of the prediction results of the server Z for the actual vehicle speed V; Figure 8 (d) is the histogram of the prediction results of the visual road environment model for the actual vehicle speed V;

[0057] Figure 9 The vehicle speed deviation of the vehicle speed deviation analysis method embodiment of the present invention is based on the brain-like cognitive computing model The prediction result graph of Figure 9 (a) is the speed deviation of servers F and Z Scatter plot of prediction results; Figure 9 (b) in the figure is the speed deviation of server Z Scatter plot of prediction results; Figure 9(c) in the figure is the deviation of the visual road environment model from the vehicle speed. Scatter plot of prediction results; Figure 9 (d) is the speed deviation of servers F and Z Histogram of prediction results; Figure 9 (e) is the speed deviation of server Z Histogram of prediction results; Figure 9 (f) is the deviation of the visual road environment model from the vehicle speed Histogram of prediction results. DETAILED DESCRIPTION

[0058] The technical solution of the present invention is further described below with reference to the accompanying drawings and embodiments.

[0059] Unless otherwise defined, technical or scientific terms used in the present invention shall have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs.

[0060] Example 1

[0061] like Figure 1 As shown, the present invention provides a vehicle speed deviation analysis method based on a brain-like cognitive computing model, which is characterized by comprising the following steps:

[0062] S1. Build a low-grade highway simulation environment to conduct a driving simulation experiment, and collect the driver's expected speed, actual speed, and EEG data during the experiment;

[0063] In step S1, traffic scene simulation software is used to build low-level highway driving scenarios, including three different scenarios: rural, plain, and mountainous areas;

[0064] The EEG data were acquired by recording the EEG signals using Brain Vision Recorder software. The sampling rate was set to 1000 Hz, the maximum electrode resistance was set to 40 µΩ, and the electrode position numbering was based on the international standard system.

[0065] Speed ​​data includes the driver's actual speed V, the expected speed Vp, and the speed error obtained by subtracting the two. .

[0066] S2. Based on the schema theory, the driver's visual driving environment is digitally analyzed to establish a visual road environment model. The visual road environment model in step S2 is divided into five different layers, including the visual road linear layer, the dynamic visual layer, the visual semantic layer, the visual depth layer and the visual sensitivity layer. Figure 2 shown.

[0067] The boundaries of the visual road alignment in the visual road alignment layer are fitted by Catmull-Rom spline curves. The advantage of this method is that it can pass through all control points, is efficient and accurate, and the local change of one control point does not affect other parts, which has local controllability. Figure 3 As shown, the establishment of the visual road linear layer first requires the construction of a coordinate system. This coordinate system takes the lower left corner of the driver's field of view as the origin, and the left and right spline curves each have four control points, which are 、 These control points are used to fit the road line shape and divide the road into three regions: "near view", "mid view" and "far view". These three regions can quantify the driver's visual perception from the perspective of different visual features.

[0068] The dynamic visual layer represents the dynamic changes in the driver's field of view with speed. The size of the driver's field of view is negatively correlated with vehicle speed, and his visual attention environment will be compressed to the center area of ​​the road as the vehicle speed increases.

[0069] The visual semantic layer uses different colors to reflect different parts of the road environment perceived by the driver. The parts of the road environment will be grouped by semantic segmentation technology and represented by the same color, such as Figure 4 As shown in Figure 3, the visual semantic layer includes the following different components, including roads, forests, sky, Gobi, traffic signs, protective measures, and cliffs, with the area ratio of their appearance in the visual road environment as a quantitative parameter;

[0070] The visual depth layer reflects the driver’s perception of the depth information of the road environment, such as Figure 4 As shown in the figure, the depth estimation technology is used to estimate the depth value of each part of the image, and then the road environment components with different depth values ​​are represented by different colors in the visual depth layer. The average value of the driver's field of view is extracted from the visual depth layer and the dynamic vision layer as a quantitative parameter;

[0071] The visual sensitivity layer reflects the distribution of sensitive areas in each road environment area under the driver's visual perception. The visual sensitivity layer is based on visual sensitivity and divides the driver's visual sensitive areas based on the gradient weighted class activation mapping algorithm (Grad-CAM). The sensitivity layer is established, such as Figure 5 As shown, the calculation results are as follows:

[0072] ;

[0073] ;

[0074] in, Represents the heat map formed by the gradient-weighted class activation mapping algorithm; Representative feature maps; Representative The weight corresponding to each feature map; is the rectified linear unit; , is the size of the feature map; ; Indicates the In the feature map, Rank The pixel value of the column.

[0075] S3. Preprocess the driver's EEG data and perform spectrum analysis. On this basis, a relationship between the environment and EEG signal intensity is established. The preprocessing method in step S3 is as follows:

[0076] Filtering, extracting useful information and removing unnecessary noise and interference bandpass filtering;

[0077] Resampling, in order to speed up the batch data processing, the sampling rate of the EEG signal is resampled from the original frequency of 1000Hz to 500Hz;

[0078] Independent component analysis, performed on EEG signals to separate the components of the mixed signal in order to analyze the activities of different neural sources;

[0079] Remove artifacts, remove electrooculogram (EOG), electromyography (EMG), electrode noise, and power supply interference, and improve the purity and accuracy of EEG signals for better analysis and interpretation of brain activity.

[0080] Remove extreme values ​​and remove extreme values ​​in EEG signals to ensure the accuracy and reliability of the data;

[0081] Rereferencing changes the reference base of the EEG signal to improve signal interpretation and analysis.

[0082] Spectral analysis describes the frequency characteristics of a signal by calculating its power spectrum over a range of frequencies. The power spectrum shows the power or energy distribution of a signal at different frequencies. This invention uses spectral analysis to process EEG signals, aiming to calculate the power spectral density of the EEG signal of a subject at a specific point in a short period of time, which represents the power or energy density of the EEG signal in different frequency ranges. The steps of spectral analysis are as follows:

[0083] S301, time domain signal conversion, converting the EEG signal from the time domain to the frequency domain;

[0084] S302, calculating the power spectrum of the converted EEG signal, and determining the energy distribution at each frequency by calculating the power spectrum of the EEG signal, in decibels;

[0085] S303, using data windowing and averaging techniques to process the EEG signal to improve the accuracy and reliability of spectrum estimation, wherein the EEG signal of the driver during the entire driving process is located at selected points, and a 1-second time period is selected at each point for spectral band analysis;

[0086] S304, Spectral band analysis, analyze the power spectrum of different frequency bands, study the changes and characteristics of EEG signals in different frequency bands, and select Wave, Wave, Wave, The power spectrum intensity values ​​of these frequency bands are averaged to obtain the driver's brain activity.

[0087] S4. Based on the QN-MHP brain-like cognitive computing model, an LMM model is established using the relationship between the signal strength of each brain region to analyze the deviation between the driver's expected speed and the actual speed. The visual processing architecture of the QN-MHP model is as follows: Figure 6 As shown, the servers involved are: visual servers 1, 2, 3, 4, cognitive servers A, C, F, H, and motion servers W, Y, Z.

[0088] The linear mixed effects regression model LMM was used for data analysis, and its expression is as follows:

[0089] ;

[0090] in, is the vector of dependent variables; is the fixed-effects score design matrix; is the parameter vector of the fixed effects; is the design matrix of random effects; is the parameter vector of the random effects; is the vector of random error terms; expresses the impact of fixed effects; represents the influence of random effects.

[0091] In the LMM model, the linear mixed model was constructed using the lme4 package in the R language. During the modeling process, the random error term was set as the subject number participating in the simulation experiment and the location of different visual road environment scenes, representing the random effects between different subjects and different road scenes, respectively.

[0092] Due to the differences in the size and distribution of each data, the maximum and minimum normalization method is used to scale the data to the range of [0,1]. The processing method is as follows:

[0093] ;

[0094] where, is the processed data point; is the data point in the original set; is the minimum value in the data set; is the maximum value in the data set;

[0095] For the speed deviation , since it has both positive and negative values, the Z-Score standardization method is used to convert the data as a whole into a form with a mean of 0 and a standard deviation of 1, and the Z-Score standardization method is as follows:

[0096] ;

[0097] where, is the processed data point; is the mean of the original set; is the standard deviation of the original set.

[0098] The results obtained after normalizing the power spectral density of each server electrical signal and the speed indicators are shown in the following table.

[0099] Table 1 Normalization results of model data

[0100]

[0101] Table 2 Normalization results of speed indicators

[0102]

[0103] According to the fitting results of the linear mixed model, the expected speed Vp, the actual speed V, and the speed deviation are predicted.

[0104] Prediction of expected speed Vp

[0105] Based on the linear mixed effect model, the relationship between server F and expected speed Vp is established with different points as random effects, and the relationship between visual road environment model significant elements and expected speed Vp is established with different drivers as random effects. After cross-validation by the bootstrap method, the distribution characteristics and prediction accuracy indicators are shown in Tables 3 and 4, and the scatter plot and prediction error histogram are shown in Figure 7 .

[0106] Table 3 Distribution characteristic indicators (unit: km / h)

[0107]

[0108] Table 4 Prediction accuracy indicators

[0109]

[0110] The data in Tables 3 and 4 show that compared to the visual road environment model, server F achieves a significantly lower root mean square error (RMSE), mean absolute error (MAE), mean square error (MSE), and mean absolute percentage error (MAPE) when predicting driver speed. The lower mean and median values ​​for driver speed prediction using server F indicate good overall prediction performance. However, the maximum and minimum prediction errors are quite extreme, at -41.99 and 52.56, respectively, with a standard deviation of 10.76, indicating significant limitations in the model's ability to predict extreme values.

[0111] Depend on Figure 7 The scatter plot shows that the results of driver speed prediction using server F are well distributed around the perfect prediction line, but there are still many extreme points that deviate from the perfect prediction line. The histogram shows that the prediction error is mainly concentrated in the range [-15, 15], indicating that the overall prediction error is small and relatively ideal. This shows that the overall prediction performance of the model is good, but the prediction accuracy for extreme data needs to be improved.

[0112] Prediction of actual vehicle speed V

[0113] Based on the linear mixed effect model, the relationship between server Z and actual vehicle speed V was established with different drivers as random effects; the relationship between the significant elements of the visual road environment model and actual vehicle speed V was established with different drivers as random effects. Its distribution characteristics and prediction accuracy indicators are shown in Tables 5 and 6, and the scatter plot and prediction error histogram are shown in Tables 5 and 6. Figure 8 shown.

[0114] Table 5 Distribution characteristic indicators (unit: km / h)

[0115]

[0116] Table 6 Prediction accuracy indicators

[0117]

[0118] The data in Tables 5 and 6 show that compared to the visual road environment model, the root mean square error (RMSE) for predicting driver speed using server Z is significantly lower, with similar values ​​for mean absolute error (MAE), mean square error (MSE), and mean absolute percentage error (MAPE). The mean and median values ​​for driver speed prediction using server Z are relatively low, indicating good overall prediction performance. However, the maximum and minimum prediction errors are quite extreme, at -50.60 and 40.46, respectively, with a standard deviation of 13.66, indicating that the model still has significant limitations when predicting extreme values.

[0119] Depend on Figure 8 The scatter plot shows that the results of predicting driver speed using server Z are well distributed around the perfect prediction line, but there are still many extreme points that deviate from the perfect prediction line. The histogram shows that the prediction error is mainly concentrated in the range [-15, 15], indicating that the overall prediction error is small and relatively ideal. This shows that the overall prediction performance of the model is good, but the accuracy of prediction for extreme data needs to be improved.

[0120] Speed ​​deviation Prediction

[0121] Based on the linear mixed effect model, the server Z and vehicle speed deviation were established with different drivers as random effects. The relationship between server F, Z and vehicle speed deviation The relationship between different drivers and the speed deviation of the visual road environment model is established with different drivers as random effects. Its distribution characteristics and prediction accuracy indicators are shown in Table 7 and Table 8, and the scatter plot and prediction error histogram are shown in Figure 9 shown.

[0122] Table 7 Distribution characteristic indicators (unit: km / h)

[0123]

[0124] Table 8 Prediction accuracy indicators

[0125]

[0126] The data in Tables 7 and 8 show that compared to the visual road environment model, the root mean square error (RMSE) for predicting driver speed using server Z is significantly lower, while the mean absolute error (MAE) and mean square error (MSE) are slightly higher. Because MAPE is very sensitive to small observations, its value is not very meaningful without comparison. The mean and median values ​​for predicting driver speed using server Z are relatively low, indicating good overall prediction performance. However, the maximum and minimum prediction errors are quite extreme, at -58.02 and 70.15, respectively, indicating that the model still has significant limitations when predicting extreme values.

[0127] Depend on Figure 9 The scatter plot shows that the results of predicting driver speed using server Z are well distributed around the perfect prediction line, but there are still many extreme points that deviate from the perfect prediction line. The histogram shows that the prediction error is mainly concentrated in the range [-15, 15], indicating that the overall prediction error is small and relatively ideal. This shows that the overall prediction performance of the model is good, but the accuracy of prediction for extreme data needs to be improved.

[0128] From the above results, it can be concluded that using a linear mixed-effects model to predict vehicle speed indicators through a server based on a brain-inspired cognitive computing model, while taking into account driver heterogeneity or driving space heterogeneity, is relatively more stable and accurate than predicting vehicle speed indicators through a visual road environment model. However, the prediction accuracy of extreme values ​​still needs to be improved.

[0129] Therefore, the present invention adopts the above-mentioned vehicle speed deviation analysis method based on the brain-like cognitive computing model to quantify the driver's perception of the road environment conditions. It plays an important role in analyzing the driver's perception and cognitive characteristics when determining the expected vehicle speed, and its impact on the difference between the expected vehicle speed and the actual operating speed, and is of great significance to improving road traffic safety.

[0130] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the same. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solutions of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A vehicle speed deviation analysis method based on a brain-inspired cognitive computing model, characterized in that: The following steps are involved: S1. Build a low-grade highway simulation environment to conduct a driving simulation experiment. During the experiment, collect the driver's expected speed, actual speed, speed deviation, and EEG data; S2. Based on schema theory, digitally analyze the driver's visual driving environment and establish a visual road environment model; The visual road environment model is divided into five different layers, including the visual road linear layer, dynamic visual layer, visual semantic layer, visual depth layer and visual sensitivity layer; The boundaries of the visual road alignment in the visual road alignment layer are fitted using Catmull-Rom spline curves. The dynamic visual layer represents the dynamic changes in the driver's field of view as a function of speed. The driver's field of view is negatively correlated with vehicle speed, and their visual attention environment is compressed to the center of the road as vehicle speed increases. The visual semantic layer uses different colors to reflect different parts of the road environment perceived by the driver. The visual semantic layer includes the following different components: road, forest, sky, desert, traffic signs, protective measures, and cliffs. The area ratio of each component in the driver's visual perception of the road environment is used as a quantitative parameter. The visual depth layer reflects the driver's perception of the depth information of the road environment. The average value of the driver's field of view is extracted from the visual depth layer and the dynamic visual layer as a quantitative parameter. The visual sensitivity layer reflects the distribution of sensitive areas in each road environment area under the driver's visual perception. The calculation results are as follows: ； ； in, Represents the heat map formed by the gradient-weighted class activation mapping algorithm; Representative feature maps; Representative The weight corresponding to each feature map; is the rectified linear unit; , is the size of the feature map; ; Indicates the In the feature map, Rank The pixel value of the column; S3. Preprocessing the driver's EEG data and performing spectrum analysis, and establishing a relationship between the environment and EEG signal intensity based on the spectrum analysis results; S4. Based on the QN-MHP brain-like cognitive computing model, an LMM model is established using the relationship between the environment and EEG signal intensity to analyze the deviation between the driver's desired speed and actual speed; The linear mixed effects regression model LMM was used for data analysis, and its expression is as follows: ； in, is the vector of dependent variables; is the fixed-effects score design matrix; is the parameter vector of the fixed effects; is the design matrix of random effects; is the parameter vector of the random effects; is the vector of random error terms; expresses the impact of fixed effects; represents the influence of random effects.

2. The vehicle speed deviation analysis method based on a brain-inspired cognitive computing model according to claim 1 is characterized by: In step S1, traffic scene simulation software is used to build a low-grade highway simulation environment, including three different scenes: township, plain, and mountainous area; The EEG data were acquired by recording the EEG signals using Brain Vision Recorder software. The sampling rate was set to 1000 Hz, the maximum electrode resistance was set to 40 µΩ, and the electrode position numbering was based on the international standard system. Speed ​​deviation It is obtained by subtracting the actual vehicle speed V from the expected vehicle speed Vp.

3. The vehicle speed deviation analysis method based on the brain-inspired cognitive computing model according to claim 2 is characterized in that: The pre-processing method in step S3 is as follows: Filtering, resampling, independent component analysis, artifact removal, extreme value removal, and re-referencing.

4. The vehicle speed deviation analysis method based on the brain-inspired cognitive computing model according to claim 3 is characterized in that: In step S3, the steps of spectrum analysis are as follows: S301, time domain signal conversion, converting the EEG signal from the time domain to the frequency domain; S302, calculating the power spectrum of the converted EEG signal, and determining the energy distribution at each frequency by calculating the power spectrum of the EEG signal; S303, using data windowing and averaging techniques to process the EEG signals, wherein the EEG signals of the driver during the entire driving process are located at selected points, and a 1-second time period is selected at each point for spectral band analysis; S304, spectral band analysis, specifically to analyze the power spectrum of different frequency bands, study the changes and characteristics of EEG signals in different frequency bands, and select Wave, Wave, Wave, The power spectrum intensity values ​​of these frequency bands are averaged to obtain the driver's brain activity.

5. The vehicle speed deviation analysis method based on the brain-inspired cognitive computing model according to claim 4 is characterized in that: In the LMM model, the linear mixed model was constructed using the lme4 package in the R language. During the modeling process, the random error term was set as the subject number participating in the simulation experiment and the location of different visual road environment scenes, representing the random effects between different subjects and different road scenes, respectively. Due to the differences in the size and distribution of each data, the maximum and minimum normalization method is used to scale the data to the range of [0,1]. The processing method is as follows: ； in, is the processed data point; is the data point in the original set; is the minimum value in the data set; is the maximum value in the data set; For speed deviation Since there are both positive and negative values, the Z-Score standardization method is used to convert the data into a form with a mean of 0 and a standard deviation of 1. The Z-Score standardization method is as follows: ； in, is the processed data point; is the mean of the original set; is the standard deviation of the original set.

Images