Tunnel driving behavior prediction method and device, equipment and storage medium

Abstract

The invention is suitable for the technical field of tunnel driving safety, and provides a tunnel driving behavior prediction method, device and equipment and a storage medium, and the method comprises the steps: obtaining the environment data of each position point in a tunnel, the physiological data of a driver at the current position point, and the driving data of a vehicle at the current position point; according to the physiological data, the cognitive load of the driver at the current position point is determined; predicting the cognitive load of each position point after the current position point according to the cognitive load and the environmental data; and according to the driving data and the cognitive load of each position point after the current position point, predicting the driving data of each position point after the current position point, and obtaining a driving behavior prediction result. The method can accurately predict the driving behavior change condition of the driver in the tunnel road section.

Description

Technical Field

[0001] The present invention belongs to the technical field of tunnel driving safety, and in particular relates to a tunnel driving behavior prediction method, device, equipment and storage medium. Background Art

[0002] The spiral tunnel of the highway adopts a small radius circular curve design in the design, and overcomes the difficulty of climbing the mountain line in the form of spiral line development, exchanging length for height. At the same time, it has the characteristics of closed internal space, long distance, continuous longitudinal slope, limited sight distance, and single braking. While the spiral tunnel greatly improves traffic accessibility and connectivity, the complexity of its driving environment also creates more uncontrollable factors and traffic safety risks. Therefore, accurately predicting the changes in the driver's driving behavior when driving in the tunnel section is of great significance to tunnel driving safety and tunnel operation management. Summary of the invention

[0003] In view of this, the embodiments of the present invention provide a tunnel driving behavior prediction method, device, equipment and storage medium to accurately predict the changes in the driver's driving behavior when driving in a tunnel section.

[0004] A first aspect of an embodiment of the present invention provides a method for predicting tunnel driving behavior, comprising:

[0005] Obtain environmental data of each location point in the tunnel, physiological data of the driver at the current location point, and driving data of the vehicle at the current location point;

[0006] Determining the cognitive load of the driver at the current location based on the physiological data;

[0007] Predicting the cognitive load of each location point after the current location point according to the cognitive load and the environmental data;

[0008] The driving data of each position point after the current position point is predicted according to the driving data and the cognitive load of each position point after the current position point to obtain a driving behavior prediction result.

[0009] In combination with the first aspect, in a possible implementation of the first aspect, the physiological data includes: a plurality of psychological factor data and a plurality of visual factor data;

[0010] Determining the cognitive load of the driver at the current location according to the physiological data includes:

[0011] Determine the psychological quantitative value based on various psychological factor data;

[0012] Determine the visual quantification value based on various visual factor data;

[0013] According to the preset weight values ​​of the psychological factors and the visual factors, the psychological quantization value and the visual quantization value are weightedly fused to obtain the cognitive load.

[0014] In combination with the first aspect, in a possible implementation of the first aspect, the environmental data includes: multiple environmental factor data;

[0015] The step of predicting the cognitive load of each location point after the current location point according to the cognitive load and the environmental data includes:

[0016] According to the cognitive load and any environmental factor data of each location point after the current location point, the cognitive load of each location point after the current location point corresponding to the environmental factor is predicted by a preset time series model of the environmental factor; wherein the time series model of the environmental factor is used to predict the cognitive load of the next location point according to the cognitive load of each location point and the environmental factor data of the next location point;

[0017] The cognitive load of each location point after the current location point is determined based on the cognitive load corresponding to various environmental factors of each location point after the current location point.

[0018] In combination with the first aspect, in a possible implementation of the first aspect, the driving data includes: a plurality of driving parameter data;

[0019] The predicting the driving data of each position point after the current position point according to the driving data and the cognitive load of each position point after the current position point includes:

[0020] According to any one of the driving parameter data in the driving data and the cognitive load of each position point after the current position point, the driving parameter data of each position point after the current position point is predicted by a preset time series model of the driving parameter;

[0021] The time series model of the driving parameter is used to predict the driving parameter data of the next location point according to the driving parameter data of each location point and the cognitive load of the next location point.

[0022] In combination with the first aspect, in a possible implementation manner of the first aspect, the environmental data includes multiple environmental factor data, and the driving data includes multiple driving parameter data;

[0023] After obtaining the driving behavior prediction result, the method further includes:

[0024] Obtain the influence of various environmental factors on various driving parameters;

[0025] Determine an adjustment amount of at least one environmental factor according to various driving parameter data of various location points after the current location point and the influence relationship of various environmental factors on various driving parameters;

[0026] Based on the adjustment amount of the at least one environmental factor, the corresponding environmental factor is adjusted.

[0027] In combination with the first aspect, in a possible implementation manner of the first aspect, obtaining the influence relationship of various environmental factors on various driving parameters includes:

[0028] Obtain driving data under different environmental data;

[0029] According to the driving data under the different environmental data, an LSTM prediction model is constructed with various environmental factors as input variables and any driving parameter as output variable;

[0030] The SHAP value of the LSTM prediction model is trained by the gradient interpreter;

[0031] The influence of various environmental factors on the driving parameters is determined according to the SHAP value.

[0032] In combination with the first aspect, in a possible implementation manner of the first aspect, the multiple environmental factor data include: at least one road condition factor data, at least one lighting environment factor data, and at least one traffic flow factor data;

[0033] The various driving parameter data include: driving speed data, acceleration data, lateral acceleration data and lateral offset data.

[0034] A second aspect of an embodiment of the present invention provides a tunnel driving behavior prediction device, comprising:

[0035] An acquisition module is used to acquire environmental data of various locations in the tunnel, physiological data of the driver at the current location, and driving data of the vehicle at the current location;

[0036] A first processing module, configured to determine the cognitive load of the driver at the current location according to the physiological data;

[0037] A second processing module, configured to predict the cognitive load of each location point after the current location point according to the cognitive load and the environmental data;

[0038] The third processing module is used to predict the driving data of each position point after the current position point according to the driving data and the cognitive load of each position point after the current position point to obtain a driving behavior prediction result.

[0039] In conjunction with the second aspect, in a possible implementation of the second aspect, the physiological data includes: a plurality of psychological factor data and a plurality of visual factor data;

[0040] The first processing module is used for:

[0041] Determine the psychological quantitative value based on various psychological factor data;

[0042] Determine the visual quantification value based on various visual factor data;

[0043] According to the preset weight values ​​of the psychological factors and the visual factors, the psychological quantization value and the visual quantization value are weightedly fused to obtain the cognitive load.

[0044] In conjunction with the second aspect, in a possible implementation of the second aspect, the environmental data includes: multiple environmental factor data;

[0045] The second processing module is used for:

[0046] According to the cognitive load and any environmental factor data of each location point after the current location point, the cognitive load of each location point after the current location point corresponding to the environmental factor is predicted by a preset time series model of the environmental factor; wherein the time series model of the environmental factor is used to predict the cognitive load of the next location point according to the cognitive load of each location point and the environmental factor data of the next location point;

[0047] The cognitive load of each location point after the current location point is determined based on the cognitive load corresponding to various environmental factors of each location point after the current location point.

[0048] In conjunction with the second aspect, in a possible implementation of the second aspect, the driving data includes: a plurality of driving parameter data;

[0049] The third processing module is used for:

[0050] According to any one of the driving parameter data in the driving data and the cognitive load of each position point after the current position point, the driving parameter data of each position point after the current position point is predicted by a preset time series model of the driving parameter;

[0051] The time series model of the driving parameter is used to predict the driving parameter data of the next location point according to the driving parameter data of each location point and the cognitive load of the next location point.

[0052] In combination with the second aspect, in a possible implementation manner of the second aspect, the environmental data includes multiple environmental factor data, and the driving data includes multiple driving parameter data;

[0053] After obtaining the driving behavior prediction result, the third processing module is further used for:

[0054] Obtain the influence of various environmental factors on various driving parameters;

[0055] Determine an adjustment amount of at least one environmental factor according to various driving parameter data of various location points after the current location point and the influence relationship of various environmental factors on various driving parameters;

[0056] Based on the adjustment amount of the at least one environmental factor, the corresponding environmental factor is adjusted.

[0057] In combination with the second aspect, in a possible implementation manner of the second aspect, the third processing module is used to:

[0058] Obtain driving data under different environmental data;

[0059] According to the driving data under the different environmental data, an LSTM prediction model is constructed with various environmental factors as input variables and any driving parameter as output variable;

[0060] The SHAP value of the LSTM prediction model is trained by the gradient interpreter;

[0061] The influence of various environmental factors on the driving parameters is determined according to the SHAP value.

[0062] In conjunction with the second aspect, in a possible implementation of the second aspect, the multiple environmental factor data include: at least one road condition factor data, at least one lighting environment factor data, and at least one traffic flow factor data;

[0063] The various driving parameter data include: driving speed data, acceleration data, lateral acceleration data and lateral offset data.

[0064] A third aspect of an embodiment of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the processor implements the steps of the method in the first aspect or any one of the implementations of the first aspect.

[0065] A fourth aspect of an embodiment of the present invention provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the steps of the method in the above-mentioned first aspect or any one of the implementation methods of the first aspect.

[0066] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0067] The embodiment of the present invention is based on the driving behavior feedback principle of the tunnel environment-people-vehicle. By acquiring the environmental data of each position point in the tunnel, the physiological data of the driver at the current position point and the driving data of the vehicle at the current position point, the cognitive load of the driver at the current position point is determined according to the physiological data, and the cognitive load of each position point after the current position point is predicted according to the cognitive load and environmental data. According to the driving data and the cognitive load of each position point after the current position point, the driving data of each position point after the current position point is predicted to obtain the driving behavior prediction result. The embodiment of the present invention can accurately predict the driver's cognitive load and vehicle operation status at each position in the tunnel under multi-dimensional complex factors, providing support for improving driving safety and enhancing tunnel traffic management. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0069] Figure 1 Schematic diagram of an application scenario of the tunnel driving behavior prediction method provided by an embodiment of the present invention;

[0070] Figure 2 It is a schematic diagram of the implementation flow of the tunnel driving behavior prediction method provided by an embodiment of the present invention;

[0071] Figure 3 is a schematic diagram of the structure of a tunnel driving behavior prediction device provided by an embodiment of the present invention;

[0072] Figure 4 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0073] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0074] In order to illustrate the technical solution of the present invention, a specific embodiment is provided below for illustration.

[0075] The multi-dimensional environmental factors of spiral tunnels include road conditions, lighting environment, and traffic flow status, and the different degrees of coupling of the multi-dimensional environment will have different effects on drivers and vehicles. Therefore, clarifying the mechanism of the effects of various dimensions of the spiral tunnel road environment on drivers and vehicles is of great significance to the driving safety of spiral tunnels and tunnel operation management.

[0076] See also Figure 1 As shown in FIG. 1 , in a spiral tunnel, there is a driving behavior feedback effect between the environment, the person, and the vehicle. The different traffic environments in the spiral tunnel cause the driver to have different cognitive loads, and then make corresponding driving decisions, changing the operating state of the vehicle. Based on this, the embodiment of the present invention proposes a tunnel driving behavior prediction method, such as Figure 2 As shown, including:

[0077] Step S201, obtaining environmental data of various positions in the tunnel, physiological data of the driver at the current position, and driving data of the vehicle at the current position.

[0078] Here, the environmental data of each location point may include, but is not limited to, road condition factor data, lighting environment factor data, and traffic flow factor data. Among them, road condition factors may include turning radius, curve length, lane position, and turning direction, etc.; lighting environment factors may include daytime, night, cold color, warm, and reflective ring lights, etc.

[0079] The driver's physiological data may include but is not limited to psychological factor data and visual factor data. Psychological factors may include heart rate, which is collected by heart rate collection equipment. Visual factors may include eye angular velocity, pupil diameter, pupil area change rate, gaze, saccade, blinking status, etc. The spiral tunnel environment information can be captured by an eye tracker and the eye movement characteristics can be recorded for analysis.

[0080] The calculation formula of pupil area change rate is:

[0081]

[0082] In the formula, is the pupil area change rate, S0 is the pupil area of ​​the driver in normal state, S i is the pupil area at position i in the tunnel.

[0083] The driving data of the vehicle may include, but is not limited to, driving speed, acceleration, lateral acceleration, lateral offset, brake pedal pressure, steering wheel steering angular velocity, etc. The lateral offset may be represented by the offset of the center line of the vehicle from the center of the lane. In this embodiment, the driving data may be collected by various sensors installed on the vehicle.

[0084] Step S202: determining the cognitive load of the driver at the current location based on the physiological data.

[0085] As a possible implementation method, the physiological data includes: multiple psychological factor data and multiple visual factor data;

[0086] Based on physiological data, determine the driver's cognitive load at the current location, including:

[0087] Determine the psychological quantitative value based on various psychological factor data;

[0088] Determine the visual quantification value based on various visual factor data;

[0089] According to the preset weight values ​​of psychological factors and visual factors, the psychological quantification value and the visual quantification value are weighted and fused to obtain the cognitive load.

[0090] In this embodiment, the average values ​​of various factor indicators of different drivers when driving in the spiral tunnel can be obtained in advance, and the psychological factor and the visual factor are taken as common factors. According to the average values ​​of various factor indicators, the variance contribution rates of the psychological factor and the visual factor are calculated by factor analysis, and the weight values ​​of the psychological factor and the visual factor are obtained. Calculate the driver's cognitive load. In the formula, F is the driver's cognitive load, a1 is the weight value of the visual factor, a2 is the weight value of the psychological factor, F1 is the quantified value of the visual factor, and F2 is the quantified value of the psychological factor.

[0091] Factor analysis can reduce the dimensionality of multiple related indicators and extract the main indicators. For example, this embodiment uses eye angular velocity, pupil diameter, pupil area change rate and heart rate value to represent the driver's cognitive load state and quantitatively characterize the driver's cognitive ability to the spiral tunnel road environment information.

[0092] The following is a specific example to illustrate the factor analysis process:

[0093] Factor analysis requires that there is a certain correlation between the indicators. First, the Pearson correlation analysis method is used to test the correlation of the indicators. As shown in Table 1, under the 95% confidence interval, the correlation of the indicators is significant, which meets the conditions for factor analysis.

[0094] Table 1 Pearson correlation analysis

[0095]

[0096] The KMO test and Bartelett sphericity test were used to judge the applicability of factor analysis. The KMO value was 0.814>0.8, and the significance probability of Bartelett sphericity test was p<0.05, which was very suitable for factor analysis.

[0097] Table 2 is the total variance explanation table. According to the variance contribution rate of common factors, it can be seen that the indicators are reasonable. According to the variance explanation results of factor analysis, the number of common factors is determined to be 2. When the factors are orthogonally rotated using the varimax method, the cumulative variance contribution rate in the total variance explanation table reaches 98.560%, the variance contribution rate of component 1 is 68.946%, and the variance contribution rate of component 2 is 29.614%.

[0098] Table 2 Total variance explained

[0099]

[0100] Table 3 shows the component matrix and factor score coefficient matrix after rotation. In the first common factor, the load coefficients of the three indicators of eye angular velocity, pupil diameter and pupil area change rate are greater than 0.9, so these three indicators are classified as the driver's visual load. In the second common factor, the heart rate HR load is greater than 0.9, so this factor is classified as the driver's psychological load. The two common factors after rotation can well explain the driver's visual and psychological behaviors.

[0101] Table 3 Rotated component matrix and factor score coefficient matrix

[0102]

[0103]

[0104] The scores of the two common factors F1 and F2 can be calculated from the component score coefficient matrix:

[0105] F1=0.384x1+0.393x2-0.408x3-0.379x4

[0106] F2=-0.107x1-0.121x2+0.154x3+1.172x4

[0107] Among them, x1, x2, x3, and x4 are the dimensionless eye angular velocity, pupil diameter, pupil area change rate, and heart rate HR, respectively.

[0108] From the total variance explanation table, we can see that the variance contribution rates of the driver's visual factor and psychological factor are 68.946% and 29.614% respectively. Therefore, the driving cognitive load quantification model F is:

[0109] F=(0.6895F1+0.2961F2) / 0.9856

[0110] In the present embodiment, when the driver is driving in the multi-dimensional and complex scene of the spiral tunnel, the changing characteristics of the driver's visual and psychological behaviors in the spiral tunnel environment can be analyzed, and the two types of data can be weightedly fused to construct a driver's cognitive load model to obtain the driver's cognitive driving load at each position in the tunnel, which can accurately, truly and comprehensively reflect the driver's cognitive ability to environmental information.

[0111] Step S203: predicting the cognitive load of each location point after the current location point according to the cognitive load and the environmental data.

[0112] In this embodiment, the environmental variables can be controlled in advance through a simulated driving experiment design to obtain the impact and action mechanism of a certain dimensional environmental factor on the driving cognitive load under the condition of controlling other dimensional factors, and obtain the relative size of the coupling weight of each element on the driver's cognitive load. Based on this, a prediction model is constructed to predict the cognitive load of each position point after the current position point according to the cognitive load and environmental data.

[0113] Exemplarily, the process of building a prediction model is as follows:

[0114] Taking environmental data including road condition factor data, lighting environment factor data and traffic flow factor data as an example, multiple independent variables (feature variables) are environmental factor variables of three dimensions of spiral tunnel as input vectors, driver cognitive load F is used as dependent variable (target variable), and the training model is ARIMAX (autoregressive integral moving average model with exogenous variables) multi-feature variable time series model. This type of model can capture the relationship between dependent variables and multiple independent variables, making the prediction more accurate.

[0115] (1) By controlling the lighting environment and traffic flow as constants, the factors affecting the road conditions of the spiral tunnel are: turning radius, curve length, lane position and turning direction. Lane position and turning direction are 0-1 variables, where 0 represents the inner lane, 1 represents the outer lane, 0 represents left turn, and 1 represents right turn. The results of the multi-feature variable time series regression model of the road condition factors on the driver's cognitive load F are as follows:

[0116] F(t)=0.504+0.998F(t-1)-0.067+4.554E-6F(R1)-1.635E-5F(L1)-0.447(Lane1)+0.188F(Turn1)

[0117] Where F(t) is the driver's cognitive load at time t, F(t-1) is the driver's cognitive load at time t-1, F(R1) is the predicted value of the radius R first and F for univariate time series prediction, similarly, L1 is the length of the curve of the turning section, Lane1 is the lane, and Turn1 is the turning direction;

[0118] The driver's cognitive load has a negative impact on the lane (inner lane / outer lane), with a regression coefficient of -0.447, that is, during the long turn in the spiral tunnel, the driver's cognitive load is lower in the outer lane than in the inner lane. The driver's cognitive load has a positive impact on the turning direction (left turn / right turn), with a regression coefficient of 0.188, that is, the driver's cognitive load is higher when turning right in the spiral tunnel than when turning left.

[0119] (2) By controlling the traffic flow environment of the spiral tunnel as a constant, the lighting environment inside and outside the tunnel is regarded as a categorical variable, where 1 represents daytime driving outside the tunnel and 2 represents nighttime driving. Inside the tunnel, 1 represents cold color temperature lamps, 2 represents cold and warm color temperature superposition lamps, and 3 represents the setting of reflective rings. When the lighting environment is changed under the same spiral tunnel road conditions, the results of the multi-feature variable time series regression model of the factors of the lighting environment dimensions inside and outside the tunnel on the driver's cognitive load F are obtained as follows:

[0120] F(t)=3.547E-5+1.021F(t-1)+0.917ε(t)+9.596E-5F(R1)-4.026E-6F(L1)+0.089F(L-out-tunnel1)+0.086F(L-in-tunnel1)

[0121] Where, L-out-tunnel1 is the lighting condition outside the spiral tunnel, L-in-tunnel1 is the lighting environment inside the spiral tunnel, and the other variables are the same as above.

[0122] The degree to which the driver's cognitive load is affected by the lighting environment inside and outside the spiral tunnel is consistent, at around 0.09, and is much larger than the influence coefficient of the tunnel turning radius and curve length in the road alignment.

[0123] (3) By controlling the spiral tunnel road condition shape and lighting environment as constants, the traffic flow environment is represented by a 0-1 variable, where 0 represents no traffic flow and 1 represents traffic flow. The results of the multi-feature variable time series regression model of the spiral tunnel traffic flow environment factors on the driver's cognitive load F are as follows:

[0124] F(t)=0.004+1.254F(t-1)-1.229ε(t)+4.694E-5F(R1)-2.434E-5F(L1)+0.001F(L-tunnel1)-0.010F(Traffic-flow1)

[0125] Where L-tunnel1 is the lighting environment inside and outside the spiral tunnel, and Traffic-flow1 is the traffic flow environment.

[0126] The driver's cognitive load has a negative impact on the traffic flow state of the spiral tunnel section, with a regression coefficient of -0.01, that is, the driver's cognitive load is smaller when there is traffic flow (free flow) in the spiral tunnel. Since the lighting environment inside and outside the tunnel is consistent, after the traffic flow is increased, the driver's cognitive load is affected by the lighting environment in the tunnel from (0.086 / 0.089) to 0.001. Therefore, the impact of traffic flow environment factors on driver cognitive load is higher than that of lighting environment factors.

[0127] (4) By analyzing the model construction results, the driver's cognitive load is most affected by the time series. When other dimensional influencing factors are controlled as constants, the turning direction and lane have a greater impact on the driver's cognitive load, with an impact coefficient of >0.1. In addition, the degree to which the driver's cognitive load is affected by a single dimension is as follows: traffic flow environment > tunnel internal and external lighting environment > road line shape. Among them, the driver's cognitive load in a free flow environment is lower than that in a non-traffic flow environment. The lower the light intensity inside and outside the tunnel, the lower the driver's cognitive load. The impact of road line shape (turn radius and curve length) on the driver's cognitive load is relatively small, with an impact coefficient of <0.001.

[0128] (5) Based on the cognitive load and any environmental factor data of each location point after the current location point, the cognitive load of each location point after the current location point corresponding to the environmental factor is predicted by using a preset time series model of the environmental factor; wherein the time series model of the environmental factor is used to predict the cognitive load of the next location point based on the cognitive load of each location point and the environmental factor data of the next location point; based on the cognitive load corresponding to various environmental factors of each location point after the current location point, the cognitive load of each location point after the current location point is determined.

[0129] Here, the cognitive load corresponding to various environmental factors at each location point can be averaged to serve as the cognitive load at each location point.

[0130] Step S204, predicting the driving data of each position point after the current position point according to the driving data and the cognitive load of each position point after the current position point, and obtaining a driving behavior prediction result.

[0131] Similar to step S203, here, a multi-feature variable time series model can also be pre-constructed to obtain the mechanism and relationship model of the driver's cognitive load on the corresponding vehicle operating state, so as to predict the corresponding vehicle operating state when the driver is under different cognitive loads.

[0132] In the embodiment of the present invention, the driving data (driving speed v, acceleration a, lane center offset d) of the vehicle in the spiral tunnel obtained under different experimental scenarios are divided into a training set and a test set. This example uses the ARIMAX model to take the driver's cognitive load as the exogenous variable input data, and the vehicle running data speed, acceleration, and lane center offset as the output data, respectively, to perform time series learning and learn the mechanism of the driver's cognitive load F on the various data of the vehicle running status.

[0133] Exemplarily, the process of model building includes:

[0134] The results of the correlation analysis between the vehicle operation status indicators in the spiral tunnel and the driver's cognitive load at the corresponding position showed that the vehicle operation status indicators in the spiral tunnel were significantly correlated with the driver's cognitive load at a confidence interval of 99%. The correlation coefficient between the driver's cognitive load and speed was 0.098 (p<0.01), the correlation coefficient with acceleration was -0.204 (p<0.01), and the correlation coefficient with lane center offset was -0.115 (p<0.01). Among them, the driver's cognitive load had a negative impact on the vehicle's driving acceleration and lane center offset.

[0135] The basic structure of the ARIMAX model is:

[0136] y t =c+β1x 1,t +…+β i x i,t +φ1y t-1 …+φ p y t-p -θ1ε t-1 -…-θ q ε t-q +ε t

[0137] In the formula, y t is the value of the target variable at time t, c is a constant term, x i,t is the value of the i-th exogenous variable at time t, β i is the coefficient of the exogenous variable, φ p and θ q are the coefficients of the autoregressive (AR) and moving average (MA) parts, p and q are the orders of the autoregressive and moving average parts, ε t is the error term at time t.

[0138] The mechanism model of the driver's cognitive load F on the vehicle speed v is:

[0139] v t =-0.2341F t+0.9717v t-1 -0.9464ε t-1 +ε t

[0140] In the formula, v t is the speed at time t, F t is the exogenous variable driver cognitive load at time t, ε t is the residual term at time point t, with variance σ 2 8.67×10 -6 The time series influence coefficient of the driver's cognitive load F on the vehicle's driving speed v is -0.2341, which is a negative influence, that is, the higher the driver's cognitive load at time t, the lower the corresponding vehicle's driving speed v.

[0141] The mechanism model of the driver's cognitive load F on the vehicle's driving acceleration a is:

[0142] a t =0.6320F t +0.9994a t-1 +0.4669ε t-1 +0.1895ε t-2 +ε t

[0143] In the formula, a t is the acceleration at time t, F t is the exogenous variable driver cognitive load at time t, ε t is the residual term at time point t, with variance σ 2 is 1.858×10 -5 The time series influence coefficient of the driver's cognitive load F on the vehicle's driving acceleration a is 0.6320, which is a positive influence, that is, the higher the driver's cognitive load at time t, the greater the corresponding vehicle's driving acceleration a.

[0144] The mechanism model of the driver's cognitive load F on the lane center offset d is:

[0145] d t =-0.185F t +1.0d t-1 +0.515ε t-1 +0.093ε t-2 +ε t

[0146] Where, d t is the lane center offset at time t, F t is the exogenous variable driver cognitive load at time t, ε t is the residual term at time point t, with variance σ 2 9.75×10-5 The time series influence coefficient of driver cognitive load F on lane center deviation is -0.185, which is a negative influence, that is, the higher the driver's cognitive load at time t, the smaller the corresponding lane center deviation.

[0147] In this embodiment, according to the driving behavior feedback process, the final effect of the spiral tunnel environment is the driving state of the vehicle. The above-obtained action mechanism model is used to calculate the prediction results of the driver's cognitive load on the vehicle's speed, acceleration, lane center offset and other operating states under different environments. The driving speed v represents the longitudinal speed control of the vehicle, the lateral acceleration represents the lateral stability of the vehicle, and the lane center offset d represents the lateral offset of the vehicle's driving trajectory, reflecting the vehicle's lane keeping ability and linear following ability.

[0148] Optionally, the lateral acceleration is calculated as:

[0149] a σ =v 2 / R1,R1=R+offset

[0150] In the formula, a σ is the lateral acceleration, R is the curve radius of the center line of the lane where the vehicle is traveling to a certain point in the spiral tunnel, and offset is the offset of the center of the vehicle from the center of the lane.

[0151] The predicted values ​​of the above model on the test set are basically consistent with the actual values, and the model has good generalization ability. Furthermore, according to any driving parameter data in the driving data and the cognitive load of each location point after the current location point, the driving parameter data of each location point after the current location point can be predicted through the preset time series model of the driving parameter; wherein the time series model of the driving parameter is used to predict the driving parameter data of the next location point according to the driving parameter data of each location point and the cognitive load of the next location point.

[0152] It can be seen that the embodiment of the present invention is based on the driving behavior feedback principle of the environment-people-vehicle in the tunnel. By obtaining the environmental data of each position point in the tunnel, the physiological data of the driver at the current position point and the driving data of the vehicle at the current position point, the cognitive load of the driver at the current position point is determined according to the physiological data, and the cognitive load of each position point after the current position point is predicted according to the cognitive load and environmental data. According to the driving data and the cognitive load of each position point after the current position point, the driving data of each position point after the current position point is predicted to obtain the driving behavior prediction result. The embodiment of the present invention can accurately predict the driver's cognitive load and vehicle operation status at each position in the tunnel under multi-dimensional complex factors, providing support for improving driving safety and enhancing tunnel traffic management.

[0153] The above content describes how to predict the driver's tunnel driving behavior. In the following embodiment, after obtaining the driving behavior prediction result in step S204, the following may also be included:

[0154] Step S205, obtaining the influence relationship of various environmental factors on various driving parameters, including: obtaining driving data under different environmental data; constructing an LSTM prediction model with various environmental factors as input variables and any driving parameter as output variable according to the driving data under different environmental data; training the SHAP value of the LSTM prediction model through a gradient interpreter; and determining the influence relationship of various environmental factors on the driving parameter according to the SHAP value.

[0155] Step S206, determining an adjustment amount of at least one environmental factor based on various driving parameter data of various location points after the current location point and the influence relationship of various environmental factors on various driving parameters;

[0156] Step S207: adjusting the corresponding environmental factor based on the adjustment amount of at least one environmental factor.

[0157] That is, in this embodiment, the vehicle driving data is also used as a prediction variable, and the spiral tunnel environment that causes the driver to have a corresponding cognitive load is used as an input variable to construct a time series model to obtain the real-time and highly dynamic mechanism of the impact of various dimensional factors of the environment on the vehicle's operating status.

[0158] In the example of the present invention, on the basis of comparing and analyzing the influence of single-dimensional factors on the vehicle motion state under different conditions of simulated driving experiments, a data-driven method of long short-term memory neural network LSTM is used to construct a time series prediction model with road environmental factors as characteristic variables and acceleration as output variables. The optimal hyperparameters of the model are obtained through training, in which the number of layers of LSTM is 2 layers, each layer has 50 neurons, the learning time step is set to 30, batch_size is 32, and epochs is 30. The mean square error MSE and R2 are used to evaluate the regression performance of the model, and R-squared is 0.928. Finally, the predicted "black box" model structure is explained to obtain the mechanism of action of environmental factors on the vehicle motion state.

[0159] The neural network LSTM model is combined with the SHAP interpretation method, and the gradient interpreter is used to train the SHAP value of the LSTM model. The post-explanation method based on game theory is used to measure the influence of each feature by calculating the marginal contribution of each feature in the model, and then the model structure of the neural network model LSTM is explained. The SHAP value represents the contribution of each input feature to the model decision process, and then quantifies the importance of each input feature. The calculation formula is:

[0160] Sample explanation:

[0161] Global explanation:

[0162] In the formula, Φ j (val) represents the Shapley value of sample feature j under the condition that the value function is val; M is the number of input features; S is the subset of features used in the model; |S| represents the number of elements in the subset S; val(S) is the contribution of the combination S to the predicted acceleration; F j (val) represents the global Shapley value of feature j in the sample set; N represents the number of sample sets, and i represents the sample index.

[0163] Exemplarily, the experimental results are as follows:

[0164] Because the LSTM model is a time series prediction, the acceleration at the previous moment has the greatest impact on the acceleration at the current moment. The importance of the other features on the vehicle's driving acceleration is ranked as follows: traffic flow > turning direction > lane > lighting environment > curve length > turning radius. The SHAP average of the traffic flow with the greatest impact is 0.00052, and the SHAP average of the turning radius with the least impact is 0.00001, which is consistent with the degree to which the driver's cognitive load is affected by environmental factors of different dimensions. Traffic flow features have a negative impact on acceleration, with a feature importance of -0.00052, that is, when there is a vehicle in front or on the side of the road, the acceleration of the vehicle will decrease. The SHAP average of the characteristic importance of the turning direction on the acceleration is 0.00027. Since 0 represents a spiral left turn and 1 represents a spiral right turn, the acceleration of a right turn in a spiral tunnel is greater than that of a left turn. The characteristic importance of the lane on the vehicle's driving acceleration is 0.00009, which is relatively small. The tunnel lighting has a greater impact on the acceleration at some locations. The importance of the turning radius and the length of the curve section in the spiral tunnel on the vehicle's driving acceleration are both less than 0.00010, and they have a negative impact on the radius. The larger the radius, the smaller the change in acceleration. In summary, from the perspective of global explanation, the vehicle operation status in the spiral tunnel is most affected by the traffic flow. The operation management department can adjust the traffic flow status to free flow by detecting the traffic flow status, making the vehicle operation smoother. When driving up and down (left turn / right turn) in the spiral tunnel, the vehicle's operation status will be different, and the right turn status will be smoother.

[0165] Based on the real-time and highly dynamic mechanism of the impact of various dimensions of environmental factors on the operating status of the vehicle, this embodiment can achieve driver constraints and tunnel safety management by regulating environmental factors. For example, the real-time status of drivers and vehicles in different environmental locations in the spiral tunnel can be monitored and warned, and the operation and management of the tunnel can be achieved by adjusting the lighting environment, controlling traffic flow and other measures to ensure driving safety and smoothness in the spiral tunnel.

[0166] In combination with the above two embodiments, the present invention first analyzes the changing characteristics of the driver's visual and psychological behaviors in the spiral tunnel environment, performs weighted fusion on the two types of data, and constructs a driver cognitive load model; based on the driving behavior feedback process of "tunnel environment information-driver-vehicle operating status", firstly compares and analyzes the action mechanism of the three-dimensional factors of the spiral tunnel "tunnel environment information-driver-vehicle operating status" on the driver's cognitive load, and obtains its coupling weight; secondly, by constructing a multi-feature variable time series model, the action mechanism and relationship model of the driver's cognitive load on the corresponding vehicle operating status are obtained, so as to obtain the corresponding vehicle operating status when the driver is under different cognitive loads; then, the vehicle operating status obtained by the driver's cognitive load is used as a prediction variable, and the spiral tunnel environment that causes the driver to have a corresponding cognitive load is used as an input variable to perform time series prediction; finally, the "black box" model structure is explained to obtain the degree of influence of environmental factors on the vehicle prediction results. The present invention can accurately obtain the high-dynamic action mechanism and coupling weight of the spiral tunnel environment-human-vehicle driving behavior feedback process under multi-dimensional complex factors, and can predict the cognitive load and vehicle operation status at each position in the tunnel, improve driving safety, enhance tunnel traffic management, and provide technical support for automatic driving trajectory planning in spiral tunnels.

[0167] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0168] Figure 3 : is a schematic diagram of the structure of a tunnel driving behavior prediction device 30 provided in an embodiment of the present invention, the device comprising:

[0169] The acquisition module 31 is used to acquire the environmental data of each position point in the tunnel, the physiological data of the driver at the current position point, and the driving data of the vehicle at the current position point.

[0170] The first processing module 32 is used to determine the cognitive load of the driver at the current location according to the physiological data.

[0171] The second processing module 33 is used to predict the cognitive load of each location point after the current location point according to the cognitive load and the environmental data.

[0172] The third processing module 34 is used to predict the driving data of each position point after the current position point according to the driving data and the cognitive load of each position point after the current position point to obtain a driving behavior prediction result.

[0173] As a possible implementation, the physiological data includes: multiple psychological factor data and multiple visual factor data; the first processing module 32 is used to:

[0174] Determine the psychological quantitative value based on various psychological factor data;

[0175] Determine the visual quantification value based on various visual factor data;

[0176] According to the preset weight values ​​of psychological factors and visual factors, the psychological quantification value and the visual quantification value are weighted and fused to obtain the cognitive load.

[0177] As a possible implementation, the environmental data includes: multiple environmental factor data; the second processing module 33 is used to:

[0178] According to the cognitive load and any environmental factor data of each location point after the current location point, the cognitive load of each location point after the current location point corresponding to the environmental factor is predicted by a preset time series model of the environmental factor; wherein the time series model of the environmental factor is used to predict the cognitive load of the next location point according to the cognitive load of each location point and the environmental factor data of the next location point;

[0179] The cognitive load of each location point after the current location point is determined based on the cognitive load corresponding to various environmental factors of each location point after the current location point.

[0180] As a possible implementation, the driving data includes: a plurality of driving parameter data; the third processing module 34 is used for:

[0181] According to any driving parameter data in the driving data and the cognitive load of each position point after the current position point, the driving parameter data of each position point after the current position point is predicted by a preset time series model of the driving parameter;

[0182] The time series model of the driving parameter is used to predict the driving parameter data of the next location point according to the driving parameter data of each location point and the cognitive load of the next location point.

[0183] As a possible implementation, the environmental data includes multiple environmental factor data, and the driving data includes multiple driving parameter data;

[0184] After obtaining the driving behavior prediction result, the third processing module 34 is further used to:

[0185] Obtain the influence of various environmental factors on various driving parameters;

[0186] Determine an adjustment amount of at least one environmental factor according to various driving parameter data of various location points after the current location point and the influence relationship of various environmental factors on various driving parameters;

[0187] Based on the adjustment amount of at least one environmental factor, the corresponding environmental factor is adjusted.

[0188] As a possible implementation manner, the third processing module 34 is used to:

[0189] Obtain driving data under different environmental data;

[0190] According to the driving data under different environmental data, an LSTM prediction model is constructed with various environmental factors as input variables and any driving parameter as output variable;

[0191] SHAP values ​​of LSTM prediction model trained by gradient interpreter;

[0192] The influence of various environmental factors on the driving parameters is determined based on the SHAP value.

[0193] As a possible implementation method, the multiple environmental factor data include: at least one road condition factor data, at least one lighting environment factor data and at least one traffic flow factor data; the multiple driving parameter data include: driving speed data, acceleration data, lateral acceleration data and lateral offset data.

[0194] The embodiment of the present invention is based on the driving behavior feedback principle of the tunnel environment-people-vehicle. By acquiring the environmental data of each position point in the tunnel, the physiological data of the driver at the current position point and the driving data of the vehicle at the current position point, the cognitive load of the driver at the current position point is determined according to the physiological data, and the cognitive load of each position point after the current position point is predicted according to the cognitive load and environmental data. According to the driving data and the cognitive load of each position point after the current position point, the driving data of each position point after the current position point is predicted to obtain the driving behavior prediction result. The embodiment of the present invention can accurately predict the driver's cognitive load and vehicle operation status at each position in the tunnel under multi-dimensional complex factors, providing support for improving driving safety and enhancing tunnel traffic management.

[0195] Figure 4 FIG. 4 is a schematic diagram of an electronic device 40 provided by an embodiment of the present invention. Figure 4As shown, the electronic device 40 of this embodiment includes: a processor 41, a memory 42, and a computer program 43 stored in the memory 42 and executable on the processor 41, such as a tunnel driving behavior prediction program. When the processor 41 executes the computer program 43, the steps in the above-mentioned tunnel driving behavior prediction method embodiments are implemented, such as Figure 2 Alternatively, when the processor 41 executes the computer program 43, the functions of each module in the above-mentioned device embodiments are realized, for example, Figure 3 The functions of the modules 31 to 34 are shown.

[0196] Exemplarily, the computer program 43 may be divided into one or more modules / units, which are stored in the memory 42 and executed by the processor 41 to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program 43 in the electronic device 40. For example, the computer program 43 may be divided into an acquisition module 31, a first processing module 32, a second processing module 33, and a third processing module 34 (modules in a virtual device), and the specific functions of each module are as follows:

[0197] The acquisition module 31 is used to acquire the environmental data of each position point in the tunnel, the physiological data of the driver at the current position point, and the driving data of the vehicle at the current position point.

[0198] The first processing module 32 is used to determine the cognitive load of the driver at the current location according to the physiological data.

[0199] The second processing module 33 is used to predict the cognitive load of each location point after the current location point according to the cognitive load and the environmental data.

[0200] The third processing module 34 is used to predict the driving data of each position point after the current position point according to the driving data and the cognitive load of each position point after the current position point to obtain a driving behavior prediction result.

[0201] The electronic device 40 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The electronic device 40 may include, but is not limited to, a processor 41 and a memory 42. Those skilled in the art will appreciate that Figure 4 It is only an example of the electronic device 40 and does not constitute a limitation of the electronic device 40. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the electronic device 40 may also include input and output devices, network access devices, buses, etc.

[0202] The processor 41 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0203] The memory 42 may be an internal storage unit of the electronic device 40, such as a hard disk or memory of the electronic device 40. The memory 42 may also be an external storage device of the electronic device 40, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 40. Further, the memory 42 may also include both an internal storage unit of the electronic device 40 and an external storage device. The memory 42 is used to store the computer program and other programs and data required by the electronic device 40. The memory 42 may also be used to temporarily store data that has been output or is to be output.

[0204] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0205] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0206] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0207] In the embodiments provided by the present invention, it should be understood that the disclosed devices / electronic devices and methods can be implemented in other ways. For example, the device / electronic device embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0208] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0209] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0210] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0211] The embodiments described above 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 aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for predicting tunnel driving behavior, characterized in that: include: Obtain environmental data of each location point in the tunnel, physiological data of the driver at the current location point, and driving data of the vehicle at the current location point; Determining the cognitive load of the driver at the current location based on the physiological data; Predicting the cognitive load of each location point after the current location point according to the cognitive load and the environmental data; The driving data of each position point after the current position point is predicted according to the driving data and the cognitive load of each position point after the current position point to obtain a driving behavior prediction result.

2. The tunnel driving behavior prediction method according to claim 1, characterized in that: The physiological data includes: multiple psychological factor data and multiple visual factor data; Determining the cognitive load of the driver at the current location according to the physiological data includes: Determine the psychological quantitative value based on various psychological factor data; Determine the visual quantification value based on various visual factor data; According to the preset weight values ​​of the psychological factors and the visual factors, the psychological quantization value and the visual quantization value are weightedly fused to obtain the cognitive load.

3. The tunnel driving behavior prediction method according to claim 1, characterized in that: The environmental data includes: multiple environmental factor data; The step of predicting the cognitive load of each location point after the current location point according to the cognitive load and the environmental data includes: According to the cognitive load and any environmental factor data of each location point after the current location point, the cognitive load of each location point after the current location point corresponding to the environmental factor is predicted by a preset time series model of the environmental factor; wherein the time series model of the environmental factor is used to predict the cognitive load of the next location point according to the cognitive load of each location point and the environmental factor data of the next location point; The cognitive load of each location point after the current location point is determined based on the cognitive load corresponding to various environmental factors of each location point after the current location point.

4. The tunnel driving behavior prediction method according to claim 1, characterized in that: The driving data includes: a variety of driving parameter data; The predicting the driving data of each position point after the current position point according to the driving data and the cognitive load of each position point after the current position point includes: According to any one of the driving parameter data in the driving data and the cognitive load of each position point after the current position point, the driving parameter data of each position point after the current position point is predicted by a preset time series model of the driving parameter; The time series model of the driving parameter is used to predict the driving parameter data of the next location point according to the driving parameter data of each location point and the cognitive load of the next location point.

5. The tunnel driving behavior prediction method according to claim 1, characterized in that: The environmental data includes a variety of environmental factor data, and the driving data includes a variety of driving parameter data; After obtaining the driving behavior prediction result, the method further includes: Obtain the influence of various environmental factors on various driving parameters; Determine an adjustment amount of at least one environmental factor according to various driving parameter data of various location points after the current location point and the influence relationship of various environmental factors on various driving parameters; Based on the adjustment amount of the at least one environmental factor, the corresponding environmental factor is adjusted.

6. The tunnel driving behavior prediction method according to claim 5, characterized in that: The acquisition of the influence relationship between various environmental factors and various driving parameters includes: Obtain driving data under different environmental data; According to the driving data under the different environmental data, an LSTM prediction model is constructed with various environmental factors as input variables and any driving parameter as output variable; The SHAP value of the LSTM prediction model is trained by the gradient interpreter; The influence of various environmental factors on the driving parameters is determined according to the SHAP value.

7. The tunnel driving behavior prediction method according to claim 5, characterized in that: The multiple environmental factor data include: at least one road condition factor data, at least one lighting environment factor data and at least one traffic flow factor data; The various driving parameter data include: driving speed data, acceleration data, lateral acceleration data and lateral offset data.

8. A tunnel driving behavior prediction device, characterized in that: include: An acquisition module is used to acquire environmental data of various locations in the tunnel, physiological data of the driver at the current location, and driving data of the vehicle at the current location; A first processing module, configured to determine the cognitive load of the driver at the current location according to the physiological data; A second processing module, configured to predict the cognitive load of each location point after the current location point according to the cognitive load and the environmental data; The third processing module is used to predict the driving data of each position point after the current position point according to the driving data and the cognitive load of each position point after the current position point to obtain a driving behavior prediction result.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images