Traffic simulation scene complexity assessment method and device based on human-vehicle-road environment

Abstract

The invention relates to the technical field of traffic data analysis, and discloses a traffic simulation scene complexity evaluation method and device based on a human-vehicle-road environment. Firstly, based on a to-be-evaluated driving scene, environmental factor complexity is determined. And determining the vehicle factor complexity according to the driving speed of driving in the to-be-evaluated driving scene. And determining the driver factor complexity according to the eye movement data of the driver driving in the to-be-evaluated driving scene. And then, based on the environmental factor complexity, the vehicle factor complexity and the driver factor complexity, performing evaluation to obtain the scene complexity corresponding to the to-be-evaluated driving scene. By considering the influence of the scene on the human-vehicle-road system, the grading quantification of the scene is realized, so that the test efficiency of the test mode based on the scene is improved. According to the implementation mode, the actual complexity of different driving scenes can be simulated and evaluated more accurately, and then more efficient and reliable support is provided for an automatic driving system and a traffic safety test.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic data analysis, and in particular to a method and device for evaluating the complexity of a traffic simulation scenario based on a human-vehicle-road environment. Background Art

[0002] In the field of road traffic, road environment complexity is a description and evaluation of the environment surrounding traffic participants and is an important physical indicator for measuring the traffic environment.

[0003] In related technologies, the complexity assessment of simulation scenarios mainly focuses on traffic factors that affect vehicle driving. However, the interaction and comprehensive impact of various factors in complex traffic environments are not fully considered. Therefore, the assessment results of scenario complexity need to be improved. Summary of the invention

[0004] The embodiments of this specification are intended to solve at least one of the technical problems in the related art to a certain extent. To this end, the embodiments of this specification propose a method and device for evaluating the complexity of a traffic simulation scenario based on a human-vehicle-road environment.

[0005] The present specification provides a method for evaluating the complexity of a traffic simulation scenario based on a human-vehicle-road environment, the method comprising:

[0006] Determine the complexity of environmental factors based on the driving scenario to be evaluated;

[0007] determining a vehicle factor complexity according to a driving speed in the driving scenario to be evaluated;

[0008] determining the driver factor complexity according to the eye movement data of the driver driving in the driving scenario to be evaluated;

[0009] An evaluation is performed based on the environmental factor complexity, the vehicle factor complexity, and the driver factor complexity to obtain a scene complexity corresponding to the driving scene to be evaluated.

[0010] In one embodiment, determining the complexity of environmental factors based on the driving scenario to be evaluated includes:

[0011] Analyzing the driving scene to be evaluated, determining scene elements included in the driving scene to be evaluated and an element structure diagram corresponding to the scene elements;

[0012] The environmental factor complexity is determined based on the scene elements included in the driving scene to be evaluated, the element structure diagram and the scene element score table.

[0013] In one embodiment, the scene element score table includes first-level elements, the first-level elements include second-level elements, and determining the element structure diagram corresponding to the scene element includes:

[0014] The scene elements included in the driving scene to be evaluated are graded based on the first-level elements and the second-level elements to determine the element structure diagram.

[0015] In one embodiment, the scene element score table includes first-level elements and scores corresponding to the first-level elements, the first-level elements include second-level elements and scores corresponding to the second-level elements, and the determining of the environmental factor complexity based on the scene elements included in the driving scene to be evaluated, the element structure diagram and the scene element score table includes:

[0016] Determining environmental factor coefficients based on the element structure diagram;

[0017] Determining weights corresponding to the first-level elements included in the driving scene to be evaluated based on scores corresponding to the first-level elements in the scene elements included in the driving scene to be evaluated;

[0018] The environmental factor complexity is determined according to the environmental factor coefficient, the weight corresponding to the first-level element included in the driving scenario to be evaluated, and the score corresponding to the second-level element included in the driving scenario to be evaluated.

[0019] In one embodiment, determining the environmental factor coefficient based on the element structure diagram includes:

[0020] Grouping the nodes in the element structure graph according to the number of inputs and outputs to obtain a plurality of node groups, wherein each node group corresponds to the number of nodes in the same group;

[0021] Calculate the occurrence probability of each node group according to the number of nodes in the same group corresponding to each node group and the total number of nodes in the element structure graph;

[0022] The information entropy calculation method is used to calculate the occurrence probability corresponding to each node group and the total number of node groups to obtain the environmental factor coefficient.

[0023] In one embodiment, determining the driver factor complexity according to the eye movement data of the driver driving in the driving scenario to be evaluated includes:

[0024] Mapping the position coordinates of the visual gaze point of the driver driving in the driving scene to be evaluated according to a target coordinate system to determine the visual gaze point in each region of interest, wherein the target coordinate system is determined according to the target region, and the target region is divided into a plurality of regions of interest;

[0025] Calculating according to the number of visual fixation points in each region of interest and the total number of visual fixation points, determining the fixation probability corresponding to each region of interest;

[0026] The information entropy calculation method is used to calculate the total number of interest areas and the gaze probability corresponding to each interest area to determine the complexity of the driver factor.

[0027] In one embodiment, the evaluating based on the environmental factor complexity, the vehicle factor complexity and the driver factor complexity to obtain the scene complexity corresponding to the driving scene to be evaluated includes:

[0028] Calculating the complexity of the environmental factors, the complexity of the vehicle factors and the complexity of the driver factors by using the hierarchical analysis method, and determining the weights corresponding to the complexity of the environmental factors, the complexity of the vehicle factors and the complexity of the driver factors respectively;

[0029] The scene complexity corresponding to the driving scene to be evaluated is obtained by performing calculations based on the weights corresponding to the complexity of the environmental factor, the complexity of the vehicle factor, and the complexity of the driver factor and the complexity of the environmental factor, the complexity of the vehicle factor, and the complexity of the driver factor.

[0030] The embodiment of this specification provides a device for evaluating the complexity of a traffic simulation scenario based on a human-vehicle-road environment, the device comprising:

[0031] An environmental factor complexity determination module, used to determine the complexity of environmental factors based on the driving scenario to be evaluated;

[0032] A vehicle factor complexity determination module, used to determine the vehicle factor complexity according to a driving speed in the driving scenario to be evaluated;

[0033] A driver factor complexity determination module, used to determine the driver factor complexity according to the eye movement data of the driver driving in the driving scene to be evaluated;

[0034] The scenario complexity determination module is used to evaluate based on the complexity of the environmental factors, the complexity of the vehicle factors and the complexity of the driver factors to obtain the scenario complexity corresponding to the driving scenario to be evaluated.

[0035] An embodiment of the present specification provides a computer device, which includes: a memory, and one or more processors communicatively connected to the memory; the memory stores instructions executable by the one or more processors, and the instructions are executed by the one or more processors to enable the one or more processors to implement the steps of the method described in any of the above embodiments.

[0036] The embodiments of this specification provide a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the method described in any of the above embodiments are implemented.

[0037] An embodiment of the present specification provides a computer program product, wherein the computer program product includes instructions, and when the instructions are executed by a processor of a computer device, the computer device is enabled to perform the steps of the method described in any one of the above embodiments.

[0038] In the implementation manner of the above specification, first, the complexity of environmental factors is determined based on the driving scenario to be evaluated. The complexity of vehicle factors is determined according to the driving speed in the driving scenario to be evaluated. The complexity of driver factors is determined according to the eye movement data of the driver driving in the driving scenario to be evaluated. Then, an evaluation is performed based on the complexity of environmental factors, the complexity of vehicle factors and the complexity of driver factors to obtain the scene complexity corresponding to the driving scenario to be evaluated. By considering the impact of the scene on the human-vehicle-road system, the hierarchical quantification of the scene is achieved, thereby improving the test efficiency of the scene-based test method. The above implementation manner can more accurately simulate and evaluate the actual complexity of different driving scenarios, thereby providing more efficient and reliable support for autonomous driving systems and traffic safety testing. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 A schematic diagram of a scenario based on a human-vehicle-road environment traffic simulation scenario complexity assessment method provided in an embodiment of this specification;

[0040] Figure 2 A schematic diagram of a process for determining the complexity of environmental factors provided in an embodiment of this specification;

[0041] Figure 3a A schematic diagram of a scene element score table provided in an embodiment of this specification;

[0042] Figure 3b A schematic diagram of a driving scenario to be evaluated provided in an embodiment of this specification;

[0043] Figure 3c A schematic diagram of an element structure diagram provided for an embodiment of this specification;

[0044] Figure 4a A schematic diagram of a process for determining the complexity of environmental factors provided in an embodiment of this specification;

[0045] Figure 4b A schematic diagram of a scene element score table provided in an embodiment of this specification;

[0046] Figure 5 A schematic diagram of a process for determining an environmental factor coefficient provided in an embodiment of this specification;

[0047] Figure 6a A schematic diagram of a process for determining the complexity of driver factors provided in an embodiment of this specification;

[0048] Figure 6b A schematic diagram of a target area and an area of ​​interest provided for an embodiment of this specification;

[0049] Figure 7 A schematic diagram of a process for obtaining the scene complexity corresponding to the driving scene to be evaluated provided in an embodiment of this specification;

[0050] Figure 8 A schematic diagram of a traffic scene complexity assessment device based on a human-vehicle-road environment provided in an embodiment of this specification;

[0051] Fig. 9 An internal structural diagram of a computer device provided for an embodiment of this specification. DETAILED DESCRIPTION

[0052] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.

[0053] In the field of road traffic, road environment complexity is a description and evaluation of the environment surrounding traffic participants and is an important physical indicator for measuring the traffic environment.

[0054] Generally speaking, road complexity is determined by a variety of factors, including vehicle speed, road curves, traffic flow, road width, etc. These factors have a direct impact on traffic flow and safety. Therefore, the complexity assessment of the traffic environment usually requires a hierarchical analysis of these factors.

[0055] In related technologies, the Before-After Study method evaluates the impact of environmental changes on traffic safety by comparing changes in traffic parameters before and after environmental changes. This analysis can provide targeted strategies and suggestions for road design and safety assessment.

[0056] Among the related technologies, the inductive summary method (causal analysis method) studies the impact of different factors on road traffic by establishing a model. Its research is usually based on a large amount of collected data, analyzing various influencing factors in the road traffic operation environment, evaluating the road environment, and thus providing theoretical support for traffic management, traffic equipment optimization and traffic safety.

[0057] Among the related technologies, models such as principal component analysis, neural network, grey theory, and information entropy are also commonly used to evaluate the road traffic environment. These models can more accurately evaluate the complexity of the traffic environment by establishing a set of comprehensive evaluation index systems.

[0058] In the testing of autonomous driving, relevant specifications (such as GB and ISO standards), traffic accident data, human driving experience, etc. are usually used to establish a test scenario library. After screening and combining the elements in the scenario library, specific simulation scenarios can be formed for vehicle testing. However, when evaluating the complexity of these simulation scenarios, the main consideration is the surrounding traffic factors that may affect the driving of the vehicle, while ignoring the complex giant system formed by the interaction and mutual influence of the three factors of the traffic environment: people, vehicles, and roads.

[0059] The driver's state and psychological load have a significant impact on driving behavior. In addition, drivers will have different perceptual responses when facing different stimuli. Therefore, when evaluating scene complexity, individual differences and responses of drivers must be taken into account.

[0060] Based on this, the implementation manner of this specification provides a method for evaluating the complexity of traffic simulation scenarios based on a human-vehicle-road environment. First, based on the driving scenario to be evaluated, the complexity of environmental factors is determined. According to the driving speed in the driving scenario to be evaluated, the complexity of vehicle factors is determined. According to the eye movement data of the driver driving in the driving scenario to be evaluated, the complexity of driver factors is determined. Then, based on the complexity of environmental factors, the complexity of vehicle factors and the complexity of driver factors, an evaluation is performed to obtain the scene complexity corresponding to the driving scenario to be evaluated. By considering the impact of the scene on the human-vehicle-road system, the hierarchical quantification of the scene is achieved, thereby improving the test efficiency of the scenario-based test method. The above implementation manner can more accurately simulate and evaluate the actual complexity of different driving scenarios, thereby providing more efficient and reliable support for autonomous driving systems and traffic safety testing.

[0061] This specification provides a method for evaluating the complexity of traffic simulation scenarios based on human-vehicle-road environments. Figure 1 , the method may include the following steps:

[0062] S110: Determine the complexity of environmental factors based on the driving scenario to be evaluated.

[0063] Specifically, when evaluating driving scenarios, we first need to analyze various environmental factors in the scenarios, such as road type, weather conditions, lighting conditions, traffic signs, etc. According to the different combinations and changes of these factors, the complexity of the environment is evaluated. For example, on urban roads, when the traffic volume is large, the traffic signals are frequent, and the road conditions are poor, the environmental complexity is high; on highways, when the traffic is smooth and the sight distance is long, the environmental complexity is low. Analyze the driving scenarios to be evaluated at will, determine the environmental influencing factors in the driving scenarios to be evaluated, and then quantify the degree of influence of these environmental influencing factors to obtain the complexity of environmental factors.

[0064] S120: Determine the complexity of vehicle factors according to a driving speed in the driving scenario to be evaluated.

[0065] Specifically, when evaluating driving scenarios, driving speed is a key vehicle factor that directly affects the complexity of driving. If the probability distribution of driving speed in the driving scenario to be evaluated is concentrated at a certain speed, the entropy value is low, which means that the range of driving speed variation is small, the uncertainty in the scenario is low, and therefore the complexity of the vehicle factor is low. If the probability distribution of driving speed in the driving scenario to be evaluated is relatively uniform in multiple speed intervals, the entropy value is high, indicating that there is a large volatility in driving speed, the uncertainty in the scenario is high, and therefore the complexity of the vehicle factor is high. Therefore, it is necessary to analyze the driving speed in the driving scenario to be evaluated, quantify the degree of influence of driving speed, and obtain the complexity of the vehicle factor.

[0066] For example, the vehicle attributes are fixed, and then the vehicle factor complexity is determined by the following formula:

[0067] C1=-∫P(v)log2P(v)dv

[0068] Among them, C1 is the complexity of vehicle factors, v is the driving speed, and P(v) is the probability distribution of driving speed.

[0069] S130: Determine the complexity of driver factors according to the eye movement data of the driver driving in the driving scenario to be evaluated.

[0070] Specifically, when evaluating driving scenarios, the driver's behavior and reaction ability are important factors affecting driving safety and efficiency. The driver's eye movement data can provide detailed information about key performances such as attention allocation, information processing, and reaction time. By analyzing the driver's eye movement data when driving in different driving scenarios, the driver's perception and judgment process of key factors in the environment can be revealed. Therefore, it is necessary to analyze the driver's eye movement data when driving in the driving scenario to be evaluated, quantify the degree of influence of the driver's eye movement data, and obtain the complexity of the driver factor.

[0071] S140: Evaluate based on the complexity of environmental factors, the complexity of vehicle factors, and the complexity of driver factors to obtain a scenario complexity corresponding to the driving scenario to be evaluated.

[0072] Specifically, when evaluating driving scenarios, it is necessary to comprehensively consider multiple influencing factors to accurately reflect the complexity of the scenarios. Therefore, after obtaining the complexity of environmental factors, vehicle factors, and driver factors, it is necessary to comprehensively consider and obtain the overall complexity of the driving scenario to be evaluated. The complexity of environmental factors, vehicle factors, and driver factors are combined through weighted average or other mathematical models to quantify a comprehensive scenario complexity index, and the scenario complexity corresponding to the driving scenario to be evaluated is obtained, thereby more accurately evaluating the overall difficulty and potential risks of the driving scenario to be evaluated.

[0073] In the above implementation, first, the complexity of environmental factors is determined based on the driving scenario to be evaluated. The complexity of vehicle factors is determined according to the driving speed in the driving scenario to be evaluated. The complexity of driver factors is determined according to the eye movement data of the driver driving in the driving scenario to be evaluated. Then, an evaluation is performed based on the complexity of environmental factors, the complexity of vehicle factors and the complexity of driver factors to obtain the scene complexity corresponding to the driving scenario to be evaluated. By considering the impact of the scene on the human-vehicle-road system, the hierarchical quantification of the scene is achieved, thereby improving the test efficiency of the scene-based test method. The above implementation can more accurately simulate and evaluate the actual complexity of different driving scenarios, thereby providing more efficient and reliable support for autonomous driving systems and traffic safety testing.

[0074] In some embodiments, see Figure 2 , based on the driving scenario to be evaluated, determining the complexity of environmental factors may include the following steps:

[0075] S210: Analyze the driving scene to be evaluated, and determine the scene elements included in the driving scene to be evaluated and the element structure diagram corresponding to the scene elements.

[0076] S220: Determine the complexity of environmental factors based on the scene elements, the element structure diagram, and the scene element score table included in the driving scene to be evaluated.

[0077] Among them, scenario elements can be key factors that affect driving behavior, vehicle operation, road safety, etc. when evaluating driving scenarios.

[0078] Specifically, a driving scenario is usually composed of multiple elements, which determine the overall environmental complexity of the scenario. Common scenario elements include road types and road conditions, traffic signals and traffic signs, weather conditions, etc. First, the driving scenario to be evaluated needs to be analyzed to identify the scenario elements it contains. These scenario elements constitute the basic environmental framework of the driving scenario to be evaluated, and the degree of influence and characteristics of each element are identified to provide a basis for subsequent complexity evaluation. After clarifying the scenario elements in the driving scenario to be evaluated, the interrelationships and influences between the scenario elements are further analyzed and displayed to obtain the element structure diagram corresponding to the scenario elements. Through the element score table, specific scoring criteria can be formulated for each element, and the calculation method of the comprehensive score can be adjusted according to different combinations of actual scenarios. By analyzing the element structure diagram, and using the element score table and the scenario elements included in the driving scenario to be evaluated, the impact of each environmental factor can be comprehensively quantified, thereby evaluating the overall complexity of the driving scenario.

[0079] In the above implementation manner, the driving scene to be evaluated is analyzed, the scene elements included in the driving scene to be evaluated and the element structure diagram corresponding to the scene elements are determined, and the complexity of environmental factors is determined based on the scene elements, the element structure diagram and the scene element score table included in the driving scene to be evaluated, which helps to accurately describe the scene complexity and provide data support for subsequent comprehensive complexity evaluation.

[0080] In some embodiments, the scene element score table includes first-level elements, the first-level elements include second-level elements, and determining the element structure diagram corresponding to the scene elements may include the following steps: grading the scene elements included in the driving scene to be evaluated based on the first-level elements and the second-level elements, and determining the element structure diagram.

[0081] Specifically, the scene element score table clearly divides the elements into levels according to certain standards, mainly divided into first-level elements and second-level elements, among which the first-level elements include second-level elements. The first-level elements classify the most critical elements in the driving scene, such as road type, weather conditions, etc. The second-level elements are further refined and classified under the first-level elements, and analyzed for specific situations. After analyzing the driving scene to be evaluated and determining the scene elements included in the driving scene to be evaluated, the scene elements included in the driving scene to be evaluated are graded according to the first-level elements and second-level elements specified in the scene element score table, and the corresponding level of each scene element and its specific category in the level to which it belongs are determined. Finally, the complete element structure diagram is constructed in combination with the grading results of each scene element. Among them, the first-level elements are located above the second-level elements, and different second-level elements under the same first-level element are all sub-nodes of the first-level element.

[0082] For example, see Figure 3a , Figure 3a The scene element score table in includes first-level elements 302 and second-level elements 304. Figure 3b , in the right Figure 3b After analyzing the driving scenarios to be evaluated, determine Figure 3b The driving scenarios to be evaluated include the following scenario elements: road section, lane, traffic light, weather, four-way, two-lane, motor vehicle signal light, daytime and sunny. Figure 3a The scene element score table in the driving scene to be evaluated determines the corresponding levels and relationships of the scene elements included in the scene to be evaluated, and obtains the element structure diagram. Figure 3c , Figure 3c It is shown in Figure 3b The driving scenarios to be evaluated include scenario elements: road sections, lanes, traffic lights, weather, four directions, two lanes, motor vehicle lights, daytime and sunny days.

[0083] In the above implementation, the scene elements included in the driving scene to be evaluated are classified based on the first-level elements and the second-level elements, and the element structure diagram is determined to quickly understand the composition of the scene and the relationship between the elements, providing a basis for subsequent complexity calculation.

[0084] In some embodiments, see Figure 4a The scene element score table includes the first-level elements and the scores corresponding to the first-level elements, and the first-level elements include the second-level elements and the scores corresponding to the second-level elements. Determining the complexity of environmental factors based on the scene elements, the element structure diagram, and the scene element score table included in the driving scene to be evaluated may include the following steps:

[0085] S410. Determine the environmental factor coefficient based on the element structure diagram.

[0086] S420: Determine weights corresponding to the first-level elements included in the driving scene to be evaluated based on scores corresponding to the first-level elements in the scene elements included in the driving scene to be evaluated.

[0087] S430: Determine the complexity of the environmental factor according to the environmental factor coefficient, the weight corresponding to the first-level element included in the driving scenario to be evaluated, and the score corresponding to the second-level element included in the driving scenario to be evaluated.

[0088] Specifically, according to the number of inputs and outputs of each element in the driving scenario to be evaluated, it is first necessary to determine the environmental factor coefficient of each element in the overall evaluation. This coefficient reflects the contribution of each element to the environmental factor. Next, based on the scores corresponding to the first-level elements in the scene elements included in the driving scenario to be evaluated, the weights corresponding to the first-level elements included in the driving scenario to be evaluated are converted, and the weights corresponding to the first-level elements included in the driving scenario to be evaluated are determined. Among them, the higher the score, the higher the weight of the element. After determining the weight corresponding to the first-level element, the weight corresponding to the first-level element and the second-level element included in the first-level element are combined to obtain the influence of the first-level element. Then, the influence of the first-level element is combined with the environmental factor coefficient to obtain the complexity of the environmental factor.

[0089] Among them, the scenario element scoring table can be based on the expert scoring method, through which experts make subjective assessments of the importance, risk level or relevance of each scenario element, thereby assigning a score to each element.

[0090] For example, see Figure 3b , Figure 3b The first-level elements in the scene elements included in the driving scene to be evaluated are road segments, lanes, traffic lights, and weather. Figure 4b ,according to Figure 4b The scene element score table in Figure 3b The scores of the first-level elements of the driving scene to be evaluated, namely the road section, lane, traffic light and weather, are 3, 6, 7 and 4 respectively. The weight of the road section is calculated according to the score of the road section: 3 / (3+6+7+4)=0.15; the weight of the lane is calculated according to the score of the lane: 6 / (3+6+7+4)=0.30; the weight of the traffic light is calculated according to the score of the traffic light: 7 / (3+6+7+4)=0.35; the weight of the weather is calculated according to the score of the weather: 4 / (3+6+7+4)=0.20. Figure 4b The scene element score table in Figure 3b The scores of the second-level factors in the driving scenario to be evaluated, namely four-way, two-lane, motor vehicle signal lights, daytime and sunny days, are 7, 5, 6, 3, and 3 respectively. Then the complexity of environmental factors is calculated according to the following formula:

[0091]

[0092] Among them, C0 is the complexity of environmental factors, λ is the environmental factor coefficient, α n is the weight corresponding to the first-level element, It is the sum of the scores corresponding to the second-level elements included in the first-level elements.

[0093] In the above implementation, the environmental factor coefficient is determined based on the element structure diagram, the weight corresponding to the first-level element included in the driving scene to be evaluated is determined based on the score corresponding to the first-level element in the scene elements included in the driving scene to be evaluated, and the environmental factor complexity is determined according to the environmental factor coefficient, the weight corresponding to the first-level element included in the driving scene to be evaluated, and the score corresponding to the second-level element included in the driving scene to be evaluated, so as to provide a data basis for the subsequent determination of the scene complexity and improve the accuracy of the scene complexity.

[0094] In some embodiments, see Figure 5 , determining the environmental factor coefficient based on the element structure diagram may include the following steps:

[0095] S510 , grouping the nodes in the element structure graph according to the number of inputs and outputs to obtain a plurality of node groups.

[0096] S520 , calculating according to the number of nodes in the same group corresponding to each node group and the total number of nodes in the element structure graph, to obtain the occurrence probability corresponding to each node group.

[0097] S530. Calculate the occurrence probability corresponding to each node group and the total number of node groups using an information entropy calculation method to obtain an environmental factor coefficient.

[0098] Each node group corresponds to the number of nodes in the same group.

[0099] Specifically, first, in the element structure diagram, each node represents a specific element, and these nodes are related to each other through input-output relationships. In order to better analyze the relationship between these elements, it is necessary to group the nodes according to the number of inputs and outputs of each node, that is, the nodes with the same number of inputs and outputs are regarded as nodes in the same group, and multiple node groups are obtained. Then, the number of nodes in the node group is calculated to obtain the number of nodes in the same group corresponding to the node group. The number of nodes in each node group will affect the probability of occurrence of the node group. Specifically, the probability of occurrence of the node group is determined by the ratio of the number of nodes in a node group, that is, the number of nodes in the same group, to the total number of nodes in the element structure diagram. On this basis, the entropy value of the occurrence probability corresponding to each node group and the total number of node groups is calculated using the information entropy calculation method to obtain the environmental factor coefficient.

[0100] For example, according to Figure 3c From the element structure diagram in , we can determine that the total number of nodes in the element structure diagram is 10. Figure 3cFrom the element structure diagram, we can determine that there is 1 node with 0 inputs and 4 outputs: scene elements; there are 3 nodes with 1 input and 1 output: road section, lane, traffic light; there is 1 node with 1 input and 2 outputs: weather; there are 5 nodes with 1 input and 0 outputs: four-way, two-lane, motor vehicle signal light, daytime, sunny.

[0101] The probability of occurrence of a node group with 0 inputs and 4 outputs is 1 / 10=0.1, the probability of occurrence of a node group with 1 input and 1 output is 3 / 10=0.3, the probability of occurrence of a node group with 1 input and 2 outputs is 1 / 10=0.1, and the probability of occurrence of a node group with 1 input and 0 output is 5 / 10=0.5.

[0102] Calculate the environmental factor coefficient according to the following formula:

[0103]

[0104] Among them, λ is the environmental factor coefficient, P i is the occurrence probability corresponding to each node group, and n is the number of node groups.

[0105] In the above implementation, nodes in the element structure diagram are grouped according to the number of inputs and outputs to obtain a plurality of node groups, and the occurrence probability corresponding to each node group is calculated based on the number of nodes in the same group corresponding to each node group and the total number of nodes in the element structure diagram. The information entropy calculation method is used to calculate the occurrence probability corresponding to each node group and the total number of node groups to obtain the environmental factor coefficient, thereby providing a data basis for the subsequent determination of the complexity of the environmental factor.

[0106] In some embodiments, see Figure 6a , determining the complexity of driver factors according to the eye movement data of the driver driving in the driving scenario to be evaluated may include the following steps:

[0107] S610: Mapping the position coordinates of the visual gaze point of the driver driving in the driving scene to be evaluated according to the target coordinate system to determine the visual gaze point in each region of interest.

[0108] S620: Calculate according to the number of visual fixation points in each region of interest and the total number of visual fixation points to determine the fixation probability corresponding to each region of interest.

[0109] The target coordinate system is determined according to the target area, and the target area is divided into a plurality of regions of interest.

[0110] Specifically, the position coordinates of the driver's visual gaze point are obtained by a sensor. Among them, the sensor can be an eye tracker. In the driving scene to be evaluated, the driver's visual gaze point is first spatially mapped according to the target coordinate system, and the coordinates of each visual gaze point are converted to the corresponding position in the target coordinate system. Since the target coordinate system is determined according to the target area, and the target area can be divided into multiple interest areas according to the characteristics of the driving scene. Therefore, based on the mapping results, the position of the driver's visual gaze point in different interest areas can be determined. Next, the number of visual gaze points in each interest area is counted, and the number of visual gaze points in all interest areas is calculated to obtain the total number of visual gaze points. The ratio of the number of visual gaze points in each interest area to the total number of visual gaze points is determined as the gaze probability corresponding to each interest area. The gaze probability of each interest area can represent the degree of attention paid by the driver to the area.

[0111] S630: Determine the complexity of the driver factor by calculating the total number of interest regions and the gaze probability corresponding to each interest region using an information entropy calculation method.

[0112] Specifically, after obtaining the gaze probability corresponding to each region of interest, the information entropy calculation method is used to further analyze the driver's gaze pattern to quantify the complexity of the driver's gaze distribution. The higher the information entropy value, the more uniform the driver's gaze distribution, and vice versa, it means that the driver's gaze is concentrated in a few areas. Finally, based on the gaze probability and information entropy value of all regions of interest, the complexity of the driver factor can be comprehensively calculated.

[0113] In some practical methods, the target area can be set as the front windshield of the vehicle, and a target coordinate system is established with the lower left corner of the front windshield as the origin. Then, the position coordinates of the driver's visual gaze point recorded by the eye tracker are mapped to the target coordinate system, that is, the position of each visual gaze point is mapped to the corresponding position of the front windshield.

[0114] In some implementations, the driver factor complexity is calculated using the following formula:

[0115]

[0116] Among them, C2 is the complexity of driver factors, N is the total number of interest areas during driving, PM i is the fixation probability corresponding to each region of interest.

[0117] For example, see Figure 6b , Figure 6b The target area 602 in FIG. 6 is divided into 9 interest areas. Visual fixation points 604, 606, and 608 are distributed in different interest areas.

[0118] In the above implementation, the position coordinates of the visual fixation points of the driver driving in the driving scene to be evaluated are mapped according to the target coordinate system, the visual fixation points in each area of ​​interest are determined, and the fixation probability corresponding to each area of ​​interest is determined according to the number of visual fixation points in each area of ​​interest and the total number of visual fixation points. The information entropy calculation method is used to calculate the total number of areas of interest and the fixation probability corresponding to each area of ​​interest to determine the complexity of the driver factor, thereby providing a data basis for the subsequent determination of the scene complexity and improving the accuracy of the scene complexity.

[0119] In some embodiments, see Figure 7 , based on the complexity of environmental factors, the complexity of vehicle factors and the complexity of driver factors, obtaining the scene complexity corresponding to the driving scene to be evaluated may include the following steps:

[0120] S710. Calculate the complexity of environmental factors, the complexity of vehicle factors, and the complexity of driver factors by using the hierarchical analysis method, and determine the weights corresponding to the complexity of environmental factors, the complexity of vehicle factors, and the complexity of driver factors.

[0121] S720. Calculate the scene complexity corresponding to the driving scene to be evaluated based on the weights corresponding to the environment factor complexity, the vehicle factor complexity, and the driver factor complexity and the environment factor complexity, the vehicle factor complexity, and the driver factor complexity.

[0122] Specifically, the analytic hierarchy process (AHP) is a multi-level, multi-criteria decision analysis method that determines the weight of each factor by constructing a hierarchical model and comparing the factors at each level. The analytic hierarchy process is used to quantify the relative importance of the complexity of environmental factors, the complexity of vehicle factors, and the complexity of driver factors, and obtain the weights corresponding to the complexity of environmental factors, the complexity of vehicle factors, and the complexity of driver factors, thereby providing a basis for the evaluation of scene complexity. According to the weights corresponding to the complexity of environmental factors, the complexity of vehicle factors, and the complexity of driver factors, and the complexity of environmental factors, the complexity of vehicle factors, and the complexity of driver factors, a comprehensive calculation is performed to obtain the scene complexity corresponding to the driving scene to be evaluated. For example, the final scene complexity can be obtained by weighted average method or other weighted calculation methods.

[0123] Exemplarily, the scene complexity is calculated by the following formula:

[0124] C=α0C0+α1C1+α2C2

[0125] Among them, C is the scene complexity, α0 is the weight corresponding to the complexity of environmental factors, C0 is the complexity of environmental factors, α1 is the weight corresponding to the complexity of vehicle factors, C1 is the complexity of vehicle factors, α2 is the weight corresponding to the complexity of driver factors, and C2 is the complexity of driver factors.

[0126] In the above implementation, the complexity of environmental factors, vehicle factors and driver factors is calculated by using the hierarchical analysis method to determine the weights corresponding to the complexity of environmental factors, vehicle factors and driver factors, and the scene complexity corresponding to the driving scene to be evaluated is obtained according to the weights corresponding to the complexity of environmental factors, vehicle factors and driver factors and the complexity of environmental factors, vehicle factors and driver factors, and the scene complexity is calculated. By considering the impact of the scene on the human-vehicle-road system, the hierarchical quantification of the scene is achieved, thereby improving the test efficiency of the scene-based test method. The above implementation can more accurately simulate and evaluate the actual complexity of different driving scenes, thereby providing more efficient and reliable support for autonomous driving systems and traffic safety testing.

[0127] The embodiment of this specification provides a device 800 for evaluating the complexity of a traffic simulation scenario based on a human-vehicle-road environment. Figure 8 The scene complexity evaluation device 800 based on the human-vehicle-road environment traffic simulation includes: an environmental factor complexity determination module 810, a vehicle factor complexity determination module 820, a driver factor complexity determination module 830, and a scene complexity determination module 840.

[0128] An environmental factor complexity determination module 810, for determining the environmental factor complexity based on the driving scenario to be evaluated;

[0129] A vehicle factor complexity determination module 820, configured to determine the vehicle factor complexity according to a driving speed in the driving scenario to be evaluated;

[0130] A driver factor complexity determination module 830, configured to determine the driver factor complexity according to the eye movement data of the driver driving in the driving scene to be evaluated;

[0131] The scenario complexity determination module 840 is used to perform an evaluation based on the environmental factor complexity, the vehicle factor complexity and the driver factor complexity to obtain the scenario complexity corresponding to the driving scenario to be evaluated.

[0132] For a detailed description of the device for evaluating the complexity of a traffic simulation scenario based on a human-vehicle-road environment, please refer to the description of the method for evaluating the complexity of a traffic simulation scenario based on a human-vehicle-road environment above, which will not be repeated here.

[0133] An embodiment of the present specification provides a computer device, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the method steps in the above embodiment are implemented.

[0134] An embodiment of the present specification provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the method in any of the above embodiments are implemented.

[0135] One embodiment of the present specification provides a computer program product, which includes instructions. When the instructions are executed by a processor of a computer device, the computer device can perform the steps of the method of any of the above embodiments.

[0136] In some embodiments, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in FIG. Fig. 9 As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a method for evaluating the complexity of a traffic simulation scene based on a human-vehicle-road environment is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a button, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0137] Those skilled in the art will understand that Fig. 9 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution disclosed in this specification, and does not constitute a limitation on the computer device to which the solution disclosed in this specification is applied. Specifically, the computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0138] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or otherwise processing in a suitable manner if necessary, and then stored in a computer memory.

Claims

1. A method for evaluating the complexity of traffic simulation scenarios based on human-vehicle-road environment, characterized in that: The method comprises: Determine the complexity of environmental factors based on the driving scenario to be evaluated; determining a vehicle factor complexity according to a driving speed in the driving scenario to be evaluated; determining the driver factor complexity according to the eye movement data of the driver driving in the driving scenario to be evaluated; An evaluation is performed based on the environmental factor complexity, the vehicle factor complexity, and the driver factor complexity to obtain a scene complexity corresponding to the driving scene to be evaluated.

2. The method according to claim 1, characterized in that: The determining of the complexity of environmental factors based on the driving scenario to be evaluated includes: Analyzing the driving scene to be evaluated, determining scene elements included in the driving scene to be evaluated and an element structure diagram corresponding to the scene elements; The environmental factor complexity is determined based on the scene elements included in the driving scene to be evaluated, the element structure diagram and the scene element score table.

3. The method according to claim 2, characterized in that The scene element score table includes first-level elements, the first-level elements include second-level elements, and determining the element structure diagram corresponding to the scene element includes: The scene elements included in the driving scene to be evaluated are graded based on the first-level elements and the second-level elements to determine the element structure diagram.

4. The method according to claim 2, characterized in that: The scene element score table includes first-level elements and scores corresponding to the first-level elements, the first-level elements include second-level elements and scores corresponding to the second-level elements, and the determining of the environmental factor complexity based on the scene elements included in the driving scene to be evaluated, the element structure diagram, and the scene element score table includes: Determining environmental factor coefficients based on the element structure diagram; Determining weights corresponding to the first-level elements included in the driving scene to be evaluated based on scores corresponding to the first-level elements in the scene elements included in the driving scene to be evaluated; The environmental factor complexity is determined according to the environmental factor coefficient, the weight corresponding to the first-level element included in the driving scenario to be evaluated, and the score corresponding to the second-level element included in the driving scenario to be evaluated.

5. The method according to claim 4, characterized in that The determining of the environmental factor coefficient based on the element structure diagram includes: Grouping the nodes in the element structure graph according to the number of inputs and outputs to obtain a plurality of node groups, wherein each node group corresponds to the number of nodes in the same group; Calculate the occurrence probability of each node group according to the number of nodes in the same group corresponding to each node group and the total number of nodes in the element structure graph; The information entropy calculation method is used to calculate the occurrence probability corresponding to each node group and the total number of node groups to obtain the environmental factor coefficient.

6. The method according to claim 1, characterized in that The determining the driver factor complexity according to the eye movement data of the driver driving in the driving scenario to be evaluated includes: Mapping the position coordinates of the visual gaze point of the driver driving in the driving scene to be evaluated according to a target coordinate system to determine the visual gaze point in each region of interest, wherein the target coordinate system is determined according to the target region, and the target region is divided into a plurality of regions of interest; Calculating according to the number of visual fixation points in each region of interest and the total number of visual fixation points, determining the fixation probability corresponding to each region of interest; The information entropy calculation method is used to calculate the total number of interest areas and the gaze probability corresponding to each interest area to determine the complexity of the driver factor.

7. The method according to claim 1, characterized in that The evaluating based on the environmental factor complexity, the vehicle factor complexity and the driver factor complexity to obtain the scene complexity corresponding to the driving scene to be evaluated includes: Calculating the complexity of the environmental factors, the complexity of the vehicle factors and the complexity of the driver factors by using the hierarchical analysis method, and determining the weights corresponding to the complexity of the environmental factors, the complexity of the vehicle factors and the complexity of the driver factors respectively; The scene complexity corresponding to the driving scene to be evaluated is obtained by performing calculations based on the weights corresponding to the complexity of the environmental factor, the complexity of the vehicle factor, and the complexity of the driver factor and the complexity of the environmental factor, the complexity of the vehicle factor, and the complexity of the driver factor.

8. A device for evaluating the complexity of traffic simulation scenarios based on human-vehicle-road environment, characterized in that: The device comprises: An environmental factor complexity determination module, used to determine the complexity of environmental factors based on the driving scenario to be evaluated; A vehicle factor complexity determination module, used to determine the vehicle factor complexity according to a driving speed in the driving scenario to be evaluated; A driver factor complexity determination module, used to determine the driver factor complexity according to the eye movement data of the driver driving in the driving scene to be evaluated; The scenario complexity determination module is used to evaluate based on the complexity of the environmental factors, the complexity of the vehicle factors and the complexity of the driver factors to obtain the scenario complexity corresponding to the driving scenario to be evaluated.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: 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 having a computer program stored thereon, 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