Method and device for testing the behavior of a driver in the presence of a hazard

Abstract

The application discloses a kind of driver evasive littering behavior characteristic test method and device.First, according to littering type, lane position, early warning measure and illumination condition design driving simulation experiment scheme;Second, using AutoCAD, latitude ground road traffic auxiliary design software, 3Dmax and SCANeR software complete experimental scene development work, including: littering model, vehicle and surrounding environment and road infrastructure etc.;Then, recruit qualified driver to complete driving simulation experiment, obtain basic index data and carry out data preprocessing, obtain effective data capable of representing vehicle space-time characteristics;Then, the index of driver transverse and longitudinal operation performance is described statistically analyzed, point by point significance analysis, difference analysis and comprehensive evaluation, analyze the driving behavior change characteristics of driver when evading littering and evaluate the danger degree of different types of littering and the effectiveness of different early warning measures.

Description

Technical Field

[0001] This invention belongs to the field of driving safety, and more specifically, relates to a method and apparatus for testing the characteristics of a driver's behavior in avoiding spilled materials. Background Technology

[0002] With the continuous increase in the number of motor vehicles and road freight transport volume, the frequency of traffic accidents caused by debris on highways is also increasing. The types of debris scattered on highways are diverse, seriously endangering the personal safety of drivers and passengers and having many adverse effects on society. Most debris is caused by improperly loaded cargo falling off vehicles during transit, or by vehicle malfunctions causing parts to scatter. It also includes garbage such as plastic bottles and food deliberately discarded by drivers and passengers. When drivers encounter large areas and large volumes of debris, they generally choose to detour, which slows down the vehicle. Changing lanes and sudden braking can easily cause collisions or rollovers. Conversely, small areas and small volumes of debris are less likely to be noticed by drivers, making it difficult for vehicles to avoid them in time, thus leading to collisions.

[0003] As a significant factor contributing to road traffic accidents, current technologies for dealing with spilled materials primarily focus on detection algorithms and inspection system design. However, methods for testing and analyzing changes in driver behavior when avoiding spilled materials using driving simulation experiments remain lacking. This fails to meet the theoretical research needs regarding how spilled materials affect driving behavior and hinders the development of early warning and control plans for spilled materials in practical engineering applications.

[0004] In real-world scenarios, it is quite difficult to obtain data on drivers' behavior when avoiding spilled objects. Driving simulators can accurately simulate vehicles and the surrounding road environment, and can collect and output fine-grained data on various indicators. Therefore, the test method and device for drivers' behavior characteristics in avoiding spilled objects proposed in this invention based on driving simulation experiments are feasible and can serve as the basis for theoretical research and early warning and prevention of spilled objects. This is of great significance for reducing traffic accidents and improving road traffic safety. Summary of the Invention

[0005] To address at least one of the aforementioned problems, this invention fills the gap in methods for testing the behavioral characteristics of drivers avoiding spilled materials and solves the problem of insufficient experimental support for the early warning and prevention of spilled materials. By utilizing driving simulation technology and designing relevant driving simulation experiments, it provides a method and apparatus for testing the behavioral characteristics of drivers avoiding spilled materials.

[0006] In a first aspect, embodiments of the present invention provide a method for testing the behavioral characteristics of drivers avoiding spilled materials. The method includes: designing a driving simulation experiment scheme based on the type of spilled material, lane location, warning measures, and lighting conditions; developing the experimental scenario using AutoCAD, Weidi software, 3Dmax, and ScaneR software, including a spilled material model, vehicle and surrounding environment, and road infrastructure; recruiting qualified drivers to complete the driving simulation experiment, acquiring basic indicator data, and preprocessing the data to obtain effective data that characterizes the spatiotemporal characteristics of the vehicle; performing descriptive statistical analysis, point-by-point significance analysis, difference analysis, and comprehensive evaluation on the indicators of the driver's lateral and longitudinal operational performance, analyzing the changing characteristics of the driver's driving behavior when avoiding spilled materials, and evaluating the degree of danger of different types of spilled materials and the effectiveness of different warning measures.

[0007] Preferably, the driving simulation experiment design includes: mainly considering four factors: type of spilled material, lane location, warning measures, and lighting conditions; the level classification under each factor is as follows: type of spilled material: plastic bottles, tarpaulins, tires, wooden crates; lane location: fast lane, slow lane; warning measures: voice prompts, traffic cone warnings, variable message signs; lighting conditions: daytime, nighttime; orthogonal design and factorial design can be used to determine the number of scenarios, each scenario contains 1-2 experiments, and the scenarios are divided into different control groups according to the elements to be studied.

[0008] Preferably, the experimental scenario development includes: basic road design information, surrounding road environment, experimental vehicle and surrounding vehicles, weather, lighting conditions, spill model, and early warning measures. The early warning measures include voice prompts, traffic cone warnings, and variable message sign prompts. The development and implementation method is as follows: a voice prompt file (.wav file) is created, and a trigger program is called in the driving simulator. When the driver is 800m away from the wooden crate, the voice prompt is triggered; red and white traffic cones are placed around the spill, but do not occupy adjacent lanes; when the variable message sign is placed 800m away from the wooden crate, it displays "Spilled material in the second lane 800m ahead, please drive carefully!" in traffic-specific font. Control codes for the experimental vehicle and surrounding vehicles need to be set in the SCANeR software, and the data acquisition frequency needs to be set to 20Hz.

[0009] Preferably, the data preprocessing includes the following steps: naming the data file according to the scene number and driver number, processing the data in columns, merging the scenes, spatial transformation, and averaging the data, dividing the data into 5m segments, and finally obtaining the distribution format of each indicator data as a function of mileage, so as to reflect the changes in driving behavior in detail.

[0010] Preferably, the analysis method for drivers' behavior in avoiding spilled materials includes the following steps: Descriptive statistical analysis: Descriptive analysis is performed on the mean and standard deviation of speed, speed standard deviation, acceleration, lane change point position, maximum steering wheel rotation angular velocity, maximum steering wheel rotation angle, lateral offset, and lateral offset standard deviation in various scenarios to describe the differences between scenarios as a whole; Point-by-point significance analysis: Comparing the differences in the changes of each indicator with mileage in different types of spilled material scenarios and different warning measure scenarios can reflect the driving behavior of drivers when dealing with spilled materials in various scenarios. At the same time, the influence range of spilled materials on drivers can be divided according to the point-by-point significance distribution characteristics of the indicators. This range is mainly the mileage before and after the driver is away from the spilled material; Difference analysis: Repeated measures ANOVA is used to explore the significant differences of indicators between different scenarios for speed performance and driving operation (lateral and longitudinal) indicators (P<0.05, indicating that the indicators have significant differences).

[0011] Preferably, the comprehensive assessment aims to evaluate the hazard level of different types of spills and the effectiveness of different early warning measures. Multiple indicators are selected from those showing significant differences in speed performance and driving operation (horizontal and longitudinal) to construct an assessment indicator system. The entropy method is a comprehensive and objective weighting method that determines the weight based on the amount of information conveyed by each indicator. It can better reflect the amount of information contained in various driving behavior characteristic indicators under the influence of different early warning measures, integrating various types of indicators to make the evaluation results more accurate. Based on the final assessment score, different types of spill scenarios are graded. The scenario with the lowest score indicates the highest hazard level of this type of spill, while the scenario with the highest score for early warning measures indicates the most effective early warning measures. These types of early warning measures should be prioritized in the early warning and control of spills.

[0012] Secondly, embodiments of the present invention provide a testing device for the characteristics of a driver's behavior in avoiding spilled materials, the device comprising:

[0013] The parameter setting module includes two functions: driving simulation test scheme parameter setting and basic index data parameter setting. The driving simulation test scheme parameter setting function is used to input the type of spilled material, lane location, warning measures, lighting conditions, number of test scenarios, and road parameter design information. The basic index data parameter setting function is used to input the names of the various basic indicators to be obtained.

[0014] The experiment and data preprocessing module is used to complete the development, experimentation and data acquisition of the driving simulation experiment scheme in the parameter setting module, and to preprocess the acquired basic index data to obtain effective index data.

[0015] The driving behavior characteristic output module is used to perform descriptive statistical analysis, point-by-point significance analysis, and difference analysis on the basic effectiveness index data, and output relevant charts characterizing the driver's behavior of avoiding spilled materials.

[0016] The comprehensive assessment, processing, and dissemination module uses the entropy method to evaluate the hazard level of different types of spilled materials and the effectiveness of different early warning measures, and publishes the assessment results to provide traffic management personnel with the optimal early warning plan.

[0017] This invention, through designing driving simulation experiments, developing scenarios, recruiting drivers to complete experiments, and processing and analyzing data, can test and analyze the changing characteristics of drivers' driving behavior when avoiding debris. It can also assess the hazard level of different types of debris and the effectiveness of early warning measures. This provides a driving simulation-based testing method for theoretical research on debris and an analytical method for studying drivers' behavior in avoiding debris. It can effectively support highway management personnel in developing early warning and control plans for debris, reducing traffic accidents and improving traffic safety. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the test method for the driver's behavior characteristics in avoiding spilled materials according to the present invention;

[0019] Figure 2 This is a schematic diagram of the projectile model and experimental scenario developed in this invention;

[0020] Figure 3 This is a schematic diagram illustrating the experimental research scope of the present invention;

[0021] Figure 4 This is a schematic diagram illustrating the descriptive statistical analysis of the present invention (taking acceleration as an example).

[0022] Figure 5 This is a schematic diagram of the point-by-point saliency distribution of the present invention (taking lateral offset as an example);

[0023] Figure 6 A schematic diagram of the testing device for the driver's behavior characteristics in avoiding spilled materials according to the present invention;

[0024] Figure 7 A schematic diagram of the plastic bottle developed for this invention;

[0025] Figure 8 A schematic diagram of the tarpaulin developed for this invention;

[0026] Figure 9 A schematic diagram of the tire developed for this invention;

[0027] Figure 10 This is a schematic diagram of the wooden crate developed according to the present invention;

[0028] Figure 11 A schematic diagram showing the projectile developed for this invention located in the fast lane;

[0029] Figure 12 A schematic diagram showing the spread material developed for this invention located in the slow lane;

[0030] Figure 13 A schematic diagram of the variable message sign developed for this invention;

[0031] Figure 14 A schematic diagram of the traffic cone warning system developed in this invention;

[0032] Figure 15 A schematic diagram of a daytime scene developed for this invention;

[0033] Figure 16 A schematic diagram of a night scene developed for this invention;

[0034] Figure 17 A point-by-point significance distribution of acceleration indicators in scenarios without warning and with voice prompts;

[0035] Figure 18 The point-by-point significance distribution of acceleration indicators for scenarios with and without traffic cone warnings;

[0036] Figure 19 A point-by-point significance distribution map of acceleration indicators without warning or variable message sign prompts;

[0037] Figure 20 A mean distribution chart of lane change point location indicators in a daytime early warning measure comparison scenario;

[0038] Figure 21 A mean distribution chart of the maximum steering wheel rotation angle in a daytime warning measure comparison scenario;

[0039] Figure 22 A mean distribution chart of the maximum steering wheel rotation angular velocity in a daytime warning measure comparison scenario;

[0040] Figure 23 This is a mean distribution chart of the lateral offset index in a daytime early warning measure comparison scenario. Detailed Implementation

[0041] To make the design, experimental process, data analysis results, and evaluation results of the driving simulation scheme of the embodiments of the present invention clearer, the testing methods and apparatus of the embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0042] Figure 1 This is a schematic flowchart of a test method for assessing a driver's behavior in avoiding spilled materials, as provided in an embodiment of the present invention. Figure 1 As shown, this embodiment of the invention provides a method for testing the characteristics of a driver's behavior in avoiding spilled materials. The specific content of the method includes:

[0043] S101, Design a driving simulation test scheme based on the type of spilled material, lane location, warning measures and lighting conditions.

[0044] Specifically, based on on-site surveys of highways, four classic types of debris frequently encountered on highways were identified: plastic bottles, tarpaulins, tires, and wooden crates. Additionally, existing debris warning measures used in the actual operation and management of highways, including variable message signs and traffic cones, were incorporated. Navigation voice prompts and a blank control were added to form four factor levels for this experiment's design of debris warning measures. Furthermore, the experimental design also considered the differences in the impact of debris location and lighting conditions on driver behavior, designing different factor levels for each. The specific experimental scheme is shown in Table 1 below.

[0045] Table 1

[0046]

[0047] This experiment consists of 10 scenarios, each containing 1-2 experiments. Each experiment is 2km long, with a 1km buffer zone (where no data is collected) between experiments within the same scenario. Traffic flow along the entire experimental route is set to free flow. To avoid potential learning effects, a distraction task is added to scenarios containing two spill experiments. The spill warning measures are located 800m before the spill's location (significantly greater than the spill's visibility distance). The scenario design is summarized in Table 2 below.

[0048] Table 2

[0049]

[0050] S102 uses AutoCAD, Weidi software, 3Dmax and SCANeR software to complete the experimental scene development.

[0051] Specifically, the basic design parameters of the spill model and the road are as follows: Figure 2 As shown, the experimental study area is divided into 5 sections and 8 key points (AH), as follows: Figure 3 As shown.

[0052] The meanings of each section and point are as follows: Section 1: The driver confirms the destination information and adjusts the driving state; no data is collected. Section 2: Data is collected 1 km before and after the first spill. Section 3: The driver completes the distraction task and returns to the designated lane; no data is collected. Section 4: Data is collected 1 km before and after the second spill. Section 5: The driver pulls over to complete the experiment; no data is collected. Point A: The starting point of the experimental road, informing the test subject of the driving lane and destination. Point B: 1 km before the first spill. Point C: The location of the first spill. Point D: 1 km after the first spill. Point E: 1 km before the second spill. Point F: The location of the second spill. Point G: 1 km after the second spill. Point H: The end point of the experimental road.

[0053] S103: Recruit qualified drivers to complete driving simulation experiments, obtain basic indicator data, and preprocess the data to obtain effective data that can characterize the spatiotemporal characteristics of the vehicle.

[0054] Specifically, this experiment selected 36 drivers, including 24 male drivers and 12 female drivers.

[0055] After removing the three pre-experiment participants, the demographic characteristics of the subjects are shown in Table 3.

[0056] Table 3

[0057]

[0058] The specific steps of the driving simulation experiment are as follows:

[0059] (1) Preliminary experiment: Three drivers were randomly selected to conduct a preliminary experiment, mainly to test whether there were any stuttering or frame drops in the experiment, and to check whether there were any abnormalities in the collected data.

[0060] (2) Experimental preparation: The subjects signed the informed consent form and completed the basic information and psychological state before the driving simulation experiment.

[0061] (3) Adaptive driving: The subject needs to familiarize himself with the simulated driving vehicle and take a 3-5 minute test drive to confirm that the subject is familiar with the operation of the simulated driving vehicle and that no dizziness or other discomfort occurs.

[0062] (4) Formal Experiment: The experimenter read the experimental instructions and informed the participants of the driving task, speed limits, etc. The 33 participants were randomly assigned to 10 scenarios to minimize the learning effect of the scenarios and the randomness of the operation process. Participants were required to rest for 5-10 minutes after every 3 driving scenarios to prevent the driving time from affecting the participants and causing inaccurate data.

[0063] (5) Post-experiment questionnaire: After all experiments are completed, participants are required to complete a post-driving subjective questionnaire. The questionnaire includes the participants' subjective feelings about the realism of the scene and driving operation after driving, as well as their behavioral intentions to avoid litter.

[0064] The basic indicator data obtained are shown in Table 4.

[0065] Table 4

[0066] Serial Number Data Chinese Name Data Tags 1 speed nSpeed_X 2 speed standard deviation SD_nSpeed_X 3 acceleration nCoGAcceleration_X 4 Acceleration standard deviation SD_nCoGAcceleration_X 5 lateral offset nLaneGap 6 Lateral offset standard deviation SD_nLaneGap 7 Braking strength nBrakePedalForce 8 Steering wheel angle nSteeringWheelAngle 9 Throttle power nGasPedal 10 time time 11 mileage nTravelledDistance 12 Lane number nLaneId 13 Data partitioning flags nCluster_103_Empty

[0067] Data preprocessing includes: the data interception range is within 1km before and 1km after the dumped material, such as... Figure 3 Sections 2 and 5 were selected. Extreme and outlier values ​​were removed, and the data was divided into 5-meter segments to reflect changes in driving behavior in detail. MATLAB software was used to transform the data and verify its normality. Based on this, the Raida criterion was used to further remove data with values ​​greater than [a certain value]. or less Experimental data ( For mathematical expectation, The missing data (with the standard deviation as the standard deviation) was supplemented by interpolation, and finally the data cleaning was completed, yielding valid experimental data from 33 subjects in 10 scenarios.

[0068] S104 involves descriptive statistical analysis, point-by-point significance analysis, difference analysis, and comprehensive evaluation of the indicators of the driver's lateral and longitudinal operational performance.

[0069] Specifically, the study compares different daytime early warning scenarios:

[0070] (1) The results of the descriptive statistical analysis are shown in Table 5.

[0071] Table 5

[0072]

[0073] Note: The value outside the parentheses represents the mean, and the value inside the parentheses represents the standard deviation. The standard deviation refers to the standard deviation between subjects in the experiment.

[0074] (2) The results of the point-by-point significance analysis are shown in Table 6 (taking the acceleration index as an example). The point-by-point significance plot in Table 6 is consistent with... Figure 5Consistent.

[0075] Table 6

[0076] Comparison Scenes Comparison results No warning and voice prompt scenarios Figure 17 No warning and traffic cone warning scenarios Figure 18 No warnings or variable message signs Figure 19

[0077] (3) The results of the difference analysis are shown in Table 7. The descriptive statistics and... Figure 4 Consistent.

[0078] Table 7

[0079] index F Sig. Descriptive Statistical Chart speed 0.250 0.821 speed standard deviation 1.103 0.379 acceleration 1.240 0.044 Figure 4 Lane change point location 20.303 0.000 Figure 20 Maximum steering wheel rotation angle 8.840 0.000 Figure 21 Maximum angular velocity of steering wheel 1.275 0.029 Figure 22 lateral offset 0.547 0.046 Figure 23 Lateral offset standard deviation 2.541 0.629

[0080] Tables 5, 6, and 7 reveal that the driver's lane-changing point is largest in the voice prompt scenario, indicating that most drivers make a driving decision after hearing the voice prompt and change lanes before seeing the wooden crate. Simultaneously, the maximum steering wheel rotation angle and angular velocity are smallest in the voice prompt scenario compared to the other three scenarios, suggesting that drivers behave more stably and with less tension after hearing the voice prompt. The driver's behavior in the variable message sign scenario is similar to that in the voice prompt scenario. The maximum steering wheel rotation angle and absolute acceleration are largest in the traffic cone warning scenario compared to the other three scenarios. This may be because the driver sees the traffic cone first, rather than the wooden crate, and might mistakenly believe there is a traffic accident ahead, leading to greater psychological tension and thus larger steering inputs. Based on these indicators, it can be concluded that voice prompts have the best warning effect, followed by variable message sign prompts, and lastly, traffic cone warnings.

[0081] (4) Construct an evaluation index system

[0082] Based on the analysis of the effectiveness of different daytime warning measures in helping drivers avoid spilled materials, five indicators were selected to evaluate the effectiveness of different warning measures: acceleration, lane change point location, maximum steering wheel rotation angle, maximum steering wheel rotation angular velocity, and lateral deviation.

[0083] (5) Comprehensive evaluation

[0084] The principle of comprehensive evaluation is as follows: assuming n schemes are selected as samples, m evaluation indicators are designed. This represents the j-th evaluation index value of the i-th scheme. The comprehensive evaluation and application steps are as follows:

[0085] After dimensionless processing of the original data to eliminate the influence of physical quantities, the characteristic proportion or contribution of the nth scheme under the jth index is calculated.

[0086]

[0087] Entropy calculation. Calculate the entropy value of the j-th index;

[0088]

[0089] Calculation of the coefficient of difference;

[0090]

[0091] Determine the weights of the evaluation indicators Calculate the overall score of the proposed scheme.

[0092]

[0093] The evaluation results are shown in Table 8.

[0094] Table 8

[0095] Scene Overall score No warning 0.06 Voice prompts 0.59 Traffic cone warning 0.11 Variable message sign prompts 0.24

[0096] The evaluation results show that the effectiveness of early warning measures is ranked as follows: voice prompts > variable message sign prompts > traffic cone warnings > no early warnings.

[0097] Figure 6 This is a schematic diagram of the structure of a driver's behavior characteristic test device for avoiding spilled materials provided in an embodiment of the present invention. Figure 6 As shown, the device includes: a parameter setting module 601, an experimental and data preprocessing module 602, a driving behavior characteristic output module 603, and a comprehensive evaluation processing and release module 604. The driver avoidance of spilled material behavior characteristic testing device provided in this embodiment of the invention specifically executes the processes of the above-described method embodiments. For details, please refer to the content of the above method embodiments; further elaboration is not provided here.

[0098] Through the above description of the embodiments, those skilled in the art can clearly understand the implementation methods of each embodiment. Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for testing the characteristics of a driver's behavior in avoiding spilled materials, characterized in that, include: Different factors affecting driving are classified into different levels, and driving simulation experiments are designed accordingly. The experimental scenario was developed using drawing software, Weidi software, 3D modeling and driving simulation software. Qualified drivers are recruited to complete driving simulation experiments, basic indicator data are obtained, and the basic indicator data is preprocessed to obtain effective data that can characterize the spatiotemporal characteristics of the vehicle. Descriptive statistical analysis, difference analysis, point-by-point significance analysis, and comprehensive evaluation were conducted on the indicators of the driver's lateral and longitudinal operation performance. Descriptive statistical analysis: Descriptive analysis is performed on the mean and standard deviation of basic indicator data in each scenario to describe the differences between scenarios as a whole; Point-by-point significance analysis: Draw point-by-point significance distribution maps of various indicators between different comparison scenarios, divide the significance interval range of the data, and then divide the research scope before and after the spillage; Analysis of differences in indicators across different scenarios: Within the scope of the study, repeated measures ANOVA was conducted on indicators characterizing the driver's lateral and longitudinal operational performance across different comparison scenarios. Comprehensive assessment: An evaluation index system was constructed by selecting indicators of significance from repeated measures ANOVA, and the entropy method was used to evaluate the degree of danger of different types of spilled materials and the effectiveness of different early warning measures.

2. The test method according to claim 1, characterized in that, The factors and levels considered in the driving simulation experiment include: Four factors: type of spilled material, lane location, warning measures, and lighting conditions; The level classification under various factors is as follows: Type of spilled material: plastic bottles, tarpaulins, tires, wooden crates; Lane location: fast lane, slow lane; Warning measures: voice prompts, traffic cone warnings, variable message signs; Lighting conditions: daytime, nighttime.

3. The test method according to claim 1, characterized in that, The steps involved in developing the experimental scenario include: Use drawing software and Weidi software to develop the basic road sections, signs and markings, road shoulders and guardrails of the scene; The scene line skeleton generated by the Weidi software is imported into the 3D modeling software for scene correction, modeling, texturing and rendering. At the same time, the software is used to draw road signs, and finally a complete static virtual simulation scene is obtained. Four types of projectile models were developed using 3D modeling software. Import the rendered scene file and debris model into the driving simulation system. First, assign the 3D scene road parameter information to the driving simulation system and digitize the scene parameters in the driving simulation system, including setting the lane width, lane marking length and vehicle driving path in the driving simulation system. Add scene elements to the scene, including time, surrounding vehicles and weather, and perform scenario control based on the developed scene; connect the developed scene with the driving simulator to conduct scene joint debugging to lay the foundation for conducting experiments.

4. The test method according to claim 1, characterized in that, The specific requirements for recruiting qualified drivers to complete the driving simulation experiment include: The number of recruited drivers needs to exceed 30, and the gender ratio needs to be consistent with the current gender ratio of drivers in society; all participants are required to be in good physical condition and have no color blindness or color weakness; they are not allowed to drink alcohol, coffee or tea within 24 hours before the experiment, and it must be ensured that the drivers are not fatigued before driving; the participants' gender, age, driving experience, occupation and weekly driving mileage are recorded. The specific steps of the driving simulation experiment are as follows: A. Preliminary experiment, B. Experiment preparation, C. Adaptive driving, D. Formal experiment, E. Post-experiment questionnaire.

5. The test method according to claim 1, characterized in that, The basic indicator data preprocessing method includes the following steps: The basic metrics that need to be extracted include: time, mileage, speed, acceleration, lateral deviation, braking intensity, steering wheel angle, throttle power, and lane number. The data collection range is from 1 km before to 1 km after the spill. At the same time, extreme and outlier values ​​are removed, and the data is divided into 5m segments to reflect changes in driving behavior in detail. The data was transformed using MATLAB software and its normality was verified. Then, data with values ​​greater than [a certain value] were removed. or less Experimental data, For mathematical expectation, The standard deviation was used, and missing data were supplemented by interpolation, thus completing the data cleaning and obtaining valid experimental data.

6. The test method according to claim 1, characterized in that, The comprehensive evaluation steps include: assuming n schemes are selected as samples, designing m evaluation indicators. This represents the j-th evaluation index value of the i-th scheme. After dimensionless processing of the raw data to eliminate the influence of physical quantities, the characteristic weight or contribution of the nth scheme under the jth index is calculated: Entropy calculation: Calculate the entropy value of the j-th index. Calculation of the coefficient of difference: Determine the weights of the evaluation indicators Calculate the overall score of the scheme: 。 7. A testing device for driver avoidance behavior characteristics of spilled materials, characterized in that, include: The parameter setting module includes two functions: setting parameters for driving simulation test schemes and setting parameters for basic index data. The driving simulation test scheme parameter setting function is used to input the type of spilled material, lane location, warning measures, lighting conditions, number of test scenarios, and road parameter design information; the basic index data parameter setting function is used to input the names of the various basic indicators to be obtained; The experiment and data preprocessing module is used to complete the development, experimentation and data acquisition of the driving simulation experiment scheme in the parameter setting module, and to preprocess the acquired basic index data to obtain effective index data. The driving behavior characteristic output module is used to perform descriptive statistical analysis, point-by-point significance analysis, and difference analysis on the basic effectiveness indicator data, and output relevant charts characterizing the driver's behavior in avoiding spilled materials. Descriptive statistical analysis: Descriptive analysis is performed on the mean and standard deviation of the basic indicator data in each scenario to describe the differences between scenarios as a whole. Point-by-point significance analysis: Point-by-point significance distribution maps of each indicator between different comparison scenarios are drawn, and the significance interval range of the data is defined, thereby defining the research scope before and after spilled materials. Difference analysis of indicators between different scenarios: Within the research scope, repeated measures ANOVA is further performed on indicators characterizing the driver's lateral and longitudinal operational performance between different comparison scenarios. The comprehensive assessment, processing, and dissemination module uses the entropy method to evaluate the hazard level of different types of spilled materials and the effectiveness of different early warning measures, and publishes the assessment results to provide traffic management personnel with the optimal early warning plan. Comprehensive assessment: The significance index of repeated measures ANOVA is selected to construct an assessment index system, and the entropy method is used to evaluate the hazard level of different types of spilled materials and the effectiveness of different early warning measures.

Images