Intersection micro-modeling simulation method and device and electronic equipment

Abstract

The invention provides an intersection microscopic modeling simulation method and device and electronic equipment, and relates to the technical field of traffic modeling simulation, and the method comprises the steps: constructing a task saturation and scene awareness relation model and a traffic element risk quantification model, based on the task saturation and scene awareness relation model and the traffic element risk quantification model, constructing a driver intersection approaching behavior model; and performing intersection analogue simulation by using the driver intersection approaching behavior model in a variable control mode. The intersection microscopic modeling simulation method and device and the electronic equipment provided by the invention not only can internally simulate and predict individual driver behaviors, but also have interpretable subjective human factors and subjective risk quantitative mathematical logic, and are sufficiently simplified.

Description

Technical Field

[0001] The present application relates to the technical field of traffic modeling and simulation, and in particular to a method, device and electronic equipment for microscopic modeling and simulation of intersections. Background Art

[0002] Microscopic traffic models primarily recreate the dynamic characteristics of safe traffic by simulating the independent behavior of individuals and analyze corresponding emerging phenomena through mathematical modeling. Signalized intersections play a key role in urban traffic networks. They manage the flow of vehicles, pedestrians, and other traffic participants to ensure the smooth operation of the transportation system. Drivers need to interact with traffic signals and other traffic participants, making intersections key points where driver behavior is complex and diverse.

[0003] In the related art, although interpretable human factor mechanisms (e.g., prospect theory and risk homeostasis theory) can be used to intrinsically predict potentially unsafe traffic maneuvers, the development and validation of current models are very complex in terms of both behavioral and computational burden.

[0004] Based on this, there is an urgent need for a microscopic modeling method for intersections that can not only intrinsically simulate and predict individual driver behavior, but also has interpretable subjective human factors and mathematical logic for quantifying subjective risks so that large-scale simulations can be performed. Summary of the Invention

[0005] The purpose of this application is to provide a microscopic modeling and simulation method, device and electronic equipment for intersections, which can not only intrinsically simulate and predict individual driver behavior, but also have explainable subjective human factors and subjective risk quantification mathematical logic, and are sufficiently simplified.

[0006] This application provides a microscopic modeling and simulation method for an intersection, comprising: A task saturation and situational awareness relationship model and a traffic element risk quantification model are constructed, and based on the task saturation and situational awareness relationship model and the traffic element risk quantification model, a driver's intersection approach behavior model is constructed; the task saturation and situational awareness relationship model is used to characterize the relationship between the driver's task saturation and situational awareness; the situational awareness is used to characterize the driver's understanding of the current situation and foresight of the future state based on the perception of internal and external environmental information; the traffic element risk quantification model is used to characterize the risk expression of each traffic element and the driver's subjective perception of risk at the intersection; the intersection approach behavior model combines the driver's subjective The expected risk of a trip after quantifying the risks of human factors and various traffic elements; the intersection approach behavior model is used to characterize the driver's behavioral expression at an intersection; by controlling variables, the driver's intersection approach behavior model is used to simulate an intersection; wherein the variables controlled during the simulation process include at least one of the following: the driver's perceptual bias, the driver's reaction time, the driver's expected risk, and the driver's heterogeneity; the perceptual bias is used to characterize the driver's ability to judge the relative distance and relative speed to the vehicle ahead; the expected risk is used to characterize the level of risk acceptable to the driver; and the heterogeneity is used to characterize the differences in driving ability between drivers.

[0007] Optionally, constructing a task saturation and situational awareness relationship model includes: obtaining a task requirement set by accumulating each driver's task requirement basic graph, and determining the logical relationship between the driver's task saturation and situational awareness by analyzing the impact of the accumulated task requirements on the driver's preferences and situational awareness; wherein, when the driver's task saturation is less than the critical task saturation, the driver's situational awareness takes the maximum value; when the driver's task saturation is greater than or equal to the maximum task saturation, the driver's situational awareness takes the minimum value; when the driver's task saturation is greater than or equal to the critical task saturation and less than the maximum task saturation, the driver's situational awareness takes the value of: the difference between the maximum situational awareness value and the target calculation result; the target calculation result is: the product of the target ratio and the first difference; the first difference is: the difference between the maximum situational awareness value and the minimum situational awareness value; the target ratio is: the ratio of the second difference to the third difference; the second difference is: the difference between the driver's task saturation and the critical task saturation; the third difference is: the difference between the maximum situational awareness value and the critical task saturation.

[0008] Optionally, the traffic element risk quantification model includes: a traffic participant risk field, a traffic signal risk field and a traffic marking risk field; the construction of the traffic element risk quantification model includes: constructing a traffic participant risk field, a traffic signal risk field and a traffic marking risk field, and obtaining an overall risk field of the intersection traffic environment based on the traffic participant risk field, the traffic signal risk field and the traffic marking risk field; based on the overall risk field, calculating the driver's subjective perceived risk value at the intersection; wherein, the traffic participant risk field is used to characterize the risk expression of traffic participants to any spatial position; the traffic signal risk field is used to characterize the risk expression of traffic signal periodic changes to any spatial position; the traffic marking risk field is used to characterize the risk expression of traffic markings to any spatial position; the overall risk field is used to characterize the risk expression of any spatial position in the intersection; the overall risk field is obtained by superimposing the risks of various traffic elements and taking the maximum value.

[0009] Optionally, the calculating of the driver's subjectively perceived risk value at the intersection based on the overall risk field includes: when the current vehicle driven by the driver is the leading vehicle at the intersection, calculating the risk value of the current vehicle based on the risk field of the current vehicle, and calculating the risk value of the traffic signal of the current road based on the traffic signal risk field, and normalizing the risk value of the current vehicle and the risk value of the traffic signal of the current road to obtain the driver's subjectively perceived risk value at the intersection; or, when the current vehicle driven by the driver is not the leading vehicle at the intersection, calculating the risk value of the current vehicle based on the risk field of the current vehicle, and calculating the risk value of the preceding vehicle based on the risk field of the preceding vehicle of the current vehicle, and normalizing the risk value of the current vehicle and the risk value of the preceding vehicle to obtain the driver's subjectively perceived risk value at the intersection.

[0010] Optionally, the driver intersection approach behavior model is constructed based on the task saturation and situational awareness relationship model and the traffic element risk quantification model, including: using risk homeostasis theory to determine the expected position of the vehicle after the preview time and the preview position after the preview time; the expected position and preview position of the vehicle are determined based on the driver's subjective perceived risk; based on the difference between the expected position of the vehicle after the preview time and the preview position after the preview time, a driver behavior model is constructed, and subjective human factors are embedded in the driver behavior model to obtain the driver intersection approach behavior model.

[0011] Optionally, the driver behavior model is constructed based on the difference between the expected position of the vehicle after the preview time and the preview position after the preview time, including: when the driver's current vehicle is the lead vehicle, based on the speed constraint and acceleration constraint of the current vehicle, calculating the final speed of the current vehicle after the preview time, and calculating the movement displacement of the current vehicle within the preview time based on the final speed; the movement displacement includes: the displacement of decelerating forward movement and the displacement of accelerating forward movement; based on the movement displacement of the current vehicle within the preview time, constructing a first behavior model of the vehicle in a braking state and a second behavior model of the vehicle in an accelerating state.

[0012] Optionally, the driver behavior model is constructed based on the difference between the expected position of the vehicle after the preview time and the preview position after the preview time, including: when the driver's current vehicle is not the lead vehicle, based on the difference between the preview position of the lead vehicle after the preview time and the preview position of the current vehicle after the preview time, calculating the relative distance between the lead vehicle and the current vehicle after the preview time; calculating the predicted risk of the current vehicle approaching the lead vehicle based on the ratio of the relative distance to the speed of the lead vehicle after the preview time, and calculating the expected position of the current vehicle after the preview time when the predicted risk is equal to the driver's expected risk; and constructing a third behavior model based on the difference between the expected position of the current vehicle after the preview time and the preview position of the current vehicle after the preview time.

[0013] Optionally, the subjective human factors are embedded into the driver behavior model to obtain the driver's intersection approach behavior model, including: determining the driver's perception expression and the driver's reaction time expression based on the situational perception error, determining the driver's expected risk expression based on the driver's task saturation deviation, and determining the driver's heterogeneity expression based on the maximum task capability, the minimum task capability and the driver's task capability; adding the perception expression, the reaction time expression, the expected risk expression and the heterogeneity expression to the driver behavior model to obtain the driver's intersection approach behavior model; wherein the situational perception error is: the difference between the situational perception maximum value and the driver's situational perception; the task saturation deviation is: the difference between the driver's task saturation and the critical task saturation; the perception expression includes: the driver's ability to judge the relative distance between the current vehicle and the preceding vehicle and the driver's ability to judge the relative speed between the current vehicle and the preceding vehicle; the relationship between the driver's situational perception and task saturation is determined based on the task saturation and situational awareness relationship model.

[0014] The present application also provides a microscopic modeling and simulation device for an intersection, comprising: The model construction module is used to construct a task saturation and situational awareness relationship model and a traffic element risk quantification model, and based on the task saturation and situational awareness relationship model and the traffic element risk quantification model, a driver's intersection approach behavior model is constructed; the task saturation and situational awareness relationship model is used to characterize the logical relationship between the driver's task saturation and situational awareness; the situational awareness is used to characterize the driver's understanding of the current situation and foresight of the future state based on the perception of internal and external environmental information; the traffic element risk quantification model is used to characterize the risk expression of each traffic element and the driver's subjective perception of risk at the intersection; the intersection approach behavior model combines the driver's The expected risk of a trip after quantification of subjective human factors and risks of various traffic elements; the intersection approach behavior model is used to characterize the driver's behavioral expression at an intersection; a simulation module is used to perform intersection simulation using the driver's intersection approach behavior model by controlling variables; wherein the variables controlled during the simulation process include at least one of the following: the driver's perceptual bias, the driver's reaction time, the driver's expected risk, and the driver's heterogeneity; the perceptual bias is used to characterize the driver's ability to judge the relative distance and relative speed to the vehicle ahead; the expected risk is used to characterize the level of risk acceptable to the driver; and the heterogeneity is used to characterize the differences in driving ability between drivers.

[0015] Optionally, the model building module is specifically used to obtain a set of task requirements by accumulating the basic graph of each driver's task requirements, and determine the logical relationship between the driver's task saturation and situational awareness by analyzing the impact of the accumulated task requirements on the driver's preferences and situational awareness; wherein, when the driver's task saturation is less than the critical task saturation, the driver's situational awareness takes the maximum value; when the driver's task saturation is greater than or equal to the maximum task saturation, the driver's situational awareness takes the minimum value; when the driver's task saturation is greater than or equal to the critical task saturation and less than the maximum task saturation, the driver's situational awareness takes the value of: the difference between the maximum situational awareness value and the target calculation result; the target calculation result is: the product of the target ratio and the first difference; the first difference is: the difference between the maximum situational awareness value and the minimum situational awareness value; the target ratio is: the ratio of the second difference to the third difference; the second difference is: the difference between the driver's task saturation and the critical task saturation; the third difference is: the difference between the maximum situational awareness value and the critical task saturation.

[0016] Optionally, the traffic element risk quantification model includes: a traffic participant risk field, a traffic signal risk field and a traffic marking risk field; the model construction module is specifically used to construct the traffic participant risk field, the traffic signal risk field and the traffic marking risk field, and obtain the overall risk field of the intersection traffic environment based on the traffic participant risk field, the traffic signal risk field and the traffic marking risk field; the model construction module is also specifically used to calculate the driver's subjective perceived risk value at the intersection based on the overall risk field; wherein, the traffic participant risk field is used to characterize the risk expression of traffic participants to any spatial position; the traffic signal risk field is used to characterize the risk expression of traffic signal periodic changes to any spatial position; the traffic marking risk field is used to characterize the risk expression of traffic markings to any spatial position; the overall risk field is used to characterize the risk expression of any spatial position in the intersection; the overall risk field is: obtained by superimposing the risks of various traffic elements and taking the maximum value.

[0017] Optionally, the model building module is specifically used to calculate the risk value of the current vehicle based on the risk field of the current vehicle, and calculate the risk value of the traffic signal of the current road based on the traffic signal risk field, when the current vehicle driven by the driver is the leading vehicle at the intersection, and normalize the risk value of the current vehicle and the risk value of the traffic signal of the current road to obtain the driver's subjective perceived risk value at the intersection; the model building module is also specifically used to calculate the risk value of the current vehicle based on the risk field of the current vehicle, and calculate the risk value of the preceding vehicle based on the risk field of the preceding vehicle of the current vehicle, and normalize the risk value of the current vehicle and the risk value of the preceding vehicle to obtain the driver's subjective perceived risk value at the intersection, when the current vehicle driven by the driver is not the leading vehicle at the intersection.

[0018] Optionally, the model construction module is specifically used to determine the expected position of the vehicle after the preview time and the preview position after the preview time by using the risk steady-state theory; the expected position and preview position of the vehicle are determined based on the driver's subjective perceived risk; the model construction module is also specifically used to construct a driver behavior model based on the difference between the expected position of the vehicle after the preview time and the preview position after the preview time, and embed subjective human factors into the driver behavior model to obtain the driver's intersection approach behavior model.

[0019] Optionally, the model building module is specifically used to calculate the final speed of the current vehicle after the preview time based on the speed constraint and acceleration constraint of the current vehicle when the driver's current vehicle is the lead vehicle, and calculate the movement displacement of the current vehicle within the preview time based on the final speed; the movement displacement includes: the displacement of decelerating forward movement and the displacement of accelerating forward movement; the model building module is also specifically used to construct a first behavior model of the vehicle in a braking state and a second behavior model of the vehicle in an accelerating state based on the movement displacement of the current vehicle within the preview time.

[0020] Optionally, the model building module is specifically used to calculate the relative distance between the front vehicle and the current vehicle after the preview time based on the difference between the preview position of the front vehicle after the preview time and the preview position of the current vehicle after the preview time when the driver's current vehicle is not the lead vehicle; the model building module is also specifically used to calculate the predicted risk of the current vehicle when approaching the front vehicle based on the ratio of the relative distance to the speed of the front vehicle after the preview time, and calculate the expected position of the current vehicle after the preview time when the predicted risk is equal to the driver's expected risk; the model building module is also specifically used to construct a third behavior model based on the difference between the expected position of the current vehicle after the preview time and the preview position of the current vehicle after the preview time.

[0021] Optionally, the model construction module is specifically used to determine the driver's perception expression and the driver's reaction time expression based on the situational perception error, determine the driver's expected risk expression based on the driver's task saturation deviation, and determine the driver's heterogeneity expression based on the maximum task capability, the minimum task capability and the driver's task capability; the model construction module is also specifically used to add the perception expression, the reaction time expression, the expected risk expression and the heterogeneity expression to the driver behavior model to obtain the driver's intersection approach behavior model; wherein, the situational perception error is: the difference between the situational perception maximum value and the driver's situational perception; the task saturation deviation is: the difference between the driver's task saturation and the critical task saturation; the perception expression includes: the driver's ability to judge the relative distance between the current vehicle and the preceding vehicle and the driver's ability to judge the relative speed between the current vehicle and the preceding vehicle; the relationship between the driver's situational perception and task saturation is determined based on the task saturation and situational awareness relationship model.

[0022] The present application also provides a computer program product, comprising a computer program / instruction, which, when executed by a processor, implements the steps of any of the above-mentioned intersection micro-modeling and simulation methods.

[0023] The present application also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of any of the above-described intersection micro-modeling and simulation methods are implemented.

[0024] The present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of any of the above-mentioned intersection micro-modeling and simulation methods are implemented.

[0025] The present application provides an intersection micro-modeling and simulation method, device, and electronic device. First, a task saturation and situational awareness relationship model and a traffic element risk quantification model are constructed. Based on these models, a driver's intersection approach behavior model is constructed. The task saturation and situational awareness relationship model is used to characterize the relationship between the driver's task saturation and situational awareness. The situational awareness is used to characterize the driver's understanding of the current situation and foresight of future states based on the perception of internal and external environmental information. The traffic element risk quantification model is used to characterize the risk expression of each traffic element and the driver's subjective perceived risk at the intersection. The intersection approach behavior model combines the driver's subjective human factors and the expected risk of the trip after the risk quantification of each traffic element. The intersection approach behavior model is used to characterize the driver's behavioral expression at the intersection. Subsequently, the intersection simulation is performed using the driver's intersection approach behavior model by controlling variables. This method not only inherently simulates and predicts individual driver behavior, but also has explainable mathematical logic for subjective human factors and subjective risk quantification, and is sufficiently simplified. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions in the present application or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0027] Figure 1 It is a flowchart of the intersection micro-modeling and simulation method provided by this application; Figure 2 This is a diagram of the relationship between the information processing workload required for the driver to complete the task and the driver's situational awareness provided by this application; Figure 3 This is one of the schematic diagrams of the risk field quantification results of different traffic elements provided in this application; Figure 4This is the second schematic diagram of the risk field quantification results of different traffic elements provided by this application; FIG5( a ) is one of the schematic diagrams of the risk field quantification results of the overall traffic environment of the intersection provided by this application; FIG5( b ) is a second schematic diagram of the risk field quantification results of the overall traffic environment of the intersection provided by this application; Figure 6 This is a schematic diagram of the risk perception of the following vehicle to the preceding vehicle in the following vehicle scenario provided by this application; Figure 7 This is a schematic diagram of the changes in the driver's subjective risk perception at different times provided by this application; Figure 8 This is one of the simulation case diagrams provided by this application when there is a perception bias; Figure 9 This is the second schematic diagram of a simulation case in which perception bias exists, provided by this application; Figure 10 This is one of the schematic diagrams of a simulation case provided by this application when there is a perception bias and a reaction time difference; Figure 11 This is the second schematic diagram of a simulation case provided by this application when there is a perception bias and a reaction time difference; Figure 12 This is one of the simulation case diagrams provided by this application when there is perception bias, reaction dynamics, and expected risk; Figure 13 This is the second schematic diagram of a simulation case provided by this application when there is perception bias, reaction dynamics, and expected risk; Figure 14 This is one of the schematic diagrams of a simulation case when all subjective human factors provided in this application are different; Figure 15 This is the second schematic diagram of a simulation case in which all subjective human factors provided by this application are different; Figure 16 It is a structural schematic diagram of the intersection micro-modeling and simulation device provided by this application; Figure 17 It is a structural diagram of the electronic device provided in this application. DETAILED DESCRIPTION

[0028] To make the objectives, technical solutions, and advantages of this application more clear, the technical solutions of this application will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of this application.

[0029] The terms "first," "second," and the like in the specification and claims of this application are used to distinguish similar objects, and are not used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, so that the embodiments of this application can be implemented in an order other than that illustrated or described herein, and that the objects distinguished by "first," "second," and the like are generally of the same type, and do not limit the number of objects; for example, the first object can be one or more. In addition, the term "and / or" in the specification and claims refers to at least one of the connected objects, and the character " / " generally indicates that the objects connected are in an "or" relationship.

[0030] The following is a detailed explanation of the professional terms involved in the embodiments of this application: Time Headway (TH): It is an important indicator for evaluating driving safety. It is closely related to traffic flow composition and driving behavior. It is an important basis for reflecting road capacity and service level, and is of great significance for optimizing road design and management.

[0031] Headway represents the time difference between the front ends of two vehicles passing the same location. It is usually calculated by dividing the headway between the front and rear vehicles by the speed of the rear vehicle. Headway represents the maximum reaction time that the driver of the rear vehicle has when the front vehicle brakes. Therefore, it does not fluctuate with changes in speed. Situational awareness: The driver's understanding of the current situation and foresight of future states based on the perception of internal and external environmental information.

[0032] With the advent of traffic engineering and autonomous driving technology, developing an intersection approach behavior model that accounts for driver heterogeneity is of great significance both at the micro level (involving individual vehicle dynamics) and the macro level (involving overall traffic flow). Such a model is crucial for the seamless coexistence and interaction between autonomous vehicles and human-driven vehicles. Autonomous vehicles should possess the cognitive ability to recognize and adapt to diverse driving behaviors, enabling them to harmoniously integrate into the traffic ecosystem. Simulating driver heterogeneity is crucial for enhancing the adaptability of autonomous driving systems in various driving scenarios, which is essential for their operational efficiency. Simulating diverse driver behaviors, on the other hand, facilitates the development of more nuanced and customized traffic management strategies. This approach allows traffic guidance systems and signal timing protocols to be tailored to the specific needs of different driver groups (e.g., conservative and aggressive drivers). This differentiated traffic management can improve road safety and efficiency.

[0033] However, incorporating driver heterogeneity into intersection approach models increases complexity and presents significant challenges. Such models require integrating a multidimensional array of parameters reflecting subjective measures (such as reaction time and risk preferences) and objective elements (such as quantified risk to other road users and traffic elements). Accurately representing these variables is crucial to developing robust and reliable intersection models capable of accurately predicting and managing traffic flow under diverse conditions. This requires a deep understanding of human behavior and the mathematical quantification of these human factors based on reasonable assumptions to develop models that are both comprehensive and applicable to real-world applications.

[0034] Microscopic traffic models primarily reproduce the dynamic characteristics of safe traffic by simulating the independent behavior of individuals and analyzing corresponding emerging phenomena through mathematical modeling. Although interpretable human factor mechanisms (such as prospect theory and risk homeostasis theory) can be used to intrinsically predict potentially unsafe traffic maneuvers, the development and validation of current models are very complex in terms of both behavioral and computational burden. Therefore, urban traffic microscopic simulation models need to develop a signalized intersection approach behavior model that meets the following requirements: 1. It can intrinsically simulate and predict individual driver behavior; 2. It has interpretable mathematical logic for subjective human factors and subjective risk quantification to enable large-scale simulations (e.g., 300 or more vehicles approaching an intersection); 3. It can still reproduce reliable vehicle trajectories and macroscopic traffic patterns; and 4. It is sufficiently simplified, scalable, and accessible to allow for easy extension by further enriching complex human factors.

[0035] Based on this, an embodiment of the present application provides a microscopic modeling and simulation method for intersections. This method uses a multi-level modeling approach, expresses the driving task through a task requirement basic diagram, and quantifies the driver's subjective risk perception of traffic signals and interaction with the preceding vehicle based on risk field theory. Specifically, it includes the following three points: 1. The relationship between driver ability, task requirements, and situational awareness is quantified using the task requirement basic diagram theory; 2. The driver's subjective risk perception of traffic signals and interaction with the preceding vehicle are quantified based on risk field theory; 3. The driver's dynamic decision (i.e., vehicle acceleration) is generated using a preview-and-follow theory that combines subjective human factors with objective risk quantification.

[0036] The following describes in detail the intersection micro-modeling and simulation method provided by the embodiment of the present application through specific embodiments and their application scenarios in conjunction with the accompanying drawings.

[0037] like Figure 1 As shown, an embodiment of the present application provides a microscopic modeling and simulation method for an intersection, which may include the following steps 101 and 102: Step 101: construct a task saturation and situational awareness relationship model and a traffic element risk quantification model, and construct a driver's intersection approach behavior model based on the task saturation and situational awareness relationship model and the traffic element risk quantification model.

[0038] Among them, the task saturation and situational awareness relationship model is used to characterize the relationship between the driver's task saturation and situational awareness; the situational awareness is used to characterize the driver's understanding of the current situation and foresight of the future state based on the perception of internal and external environmental information; the traffic element risk quantification model is used to characterize the risk expression of each traffic element and the driver's subjective perceived risk at the intersection; the intersection approach behavior model combines the driver's subjective human factors and the expected risk of the journey after the risk quantification of each traffic element; the intersection approach behavior model is used to characterize the driver's behavioral expression at the intersection.

[0039] For example, in the embodiments of the present application, a basic diagram of driver task demand (TD) is used to describe the cumulative workload accepted by the driver, and situational awareness (SA) is used to describe the driver's perception of the environment, especially the environmental stimuli required for the safe and effective execution of the driver.

[0040] Specifically, in the above step 101, the construction of the relationship model between task saturation and situational awareness may further include the following step 101a: Step 101a: Obtain a set of task requirements by accumulating each driver's task requirement basic graph, and determine the logical relationship between the driver's task saturation and situational awareness by analyzing the impact of the accumulated task requirements on the driver's preferences and situational awareness.

[0041] Among them, when the driver's task saturation is less than the critical task saturation, the driver's situational awareness takes the maximum value; when the driver's task saturation is greater than or equal to the maximum task saturation, the driver's situational awareness takes the minimum value; when the driver's task saturation is greater than or equal to the critical task saturation and less than the maximum task saturation, the driver's situational awareness is taken as: the difference between the maximum situational awareness value and the target calculation result; the target calculation result is: the product of the target ratio and the first difference; the first difference is: the difference between the maximum situational awareness value and the minimum situational awareness value; the target ratio is: the ratio of the second difference to the third difference; the second difference is: the difference between the driver's task saturation and the critical task saturation; the third difference is: the difference between the maximum situational awareness value and the critical task saturation.

[0042] For example, Figure 2The figure shows the logical relationship between the information processing workload (TD) required for the driver to complete the task and the driver's situational awareness. In this embodiment, all task requirements are first obtained by accumulating each individual task requirement basic graph. The impact of the accumulated task requirements on driver preferences (i.e., desired speed and headway) and situational awareness is analyzed. Situational awareness also dynamically affects perception error and reaction time. The calculation method for the information processing workload required to complete the task per unit time can be expressed by the following formula 1: (Formula 1) The information processing capacity (TC) of an unimpaired driver to effectively and safely perform a task can be calculated using the following formula: (Formula 2) in, is the total driving task requirement, The task requirements required to perform a specific driving task I, is the driver's task saturation, The driver's task capability.

[0043] For example, Endsley's dynamic model of situational awareness is widely used for situational awareness. It includes three levels: perception of environmental elements, comprehensive understanding of the current situation, and prediction of future states. Through the three-layer perception process, it can be clearly understood that there is a physical reaction time from the beginning of perception to the end of expectation. In addition, there is an attentional time lag as non-critical information competes for attention. Therefore, the total reaction time should be the sum of attention time lag and physical reaction time, which can be expressed by the following formula 3: (Formula 3) Among them, when the driver's task saturation is close to (or greater than) 1, the driver's driving performance will decline significantly, which may be manifested as changes in consciousness (perception error and slow reaction), response (reduced sensitivity) and driver state (increased stress). However, it should be pointed out that although high task saturation will reduce the state of consciousness, too low saturation may also cause cognitive distraction of the driver. Therefore, and The relationship is inverted U-shaped. As a function of time headway, it can be expressed by the following formulas 4 and 5: (Formula 4) (Formula 5) in, Indicates thei The headway between the vehicle and the preceding vehicle, Indicates the relative distance between them. It is i The speed of the car, and are the minimum TD value and the maximum TD value of the car-following task, which are set to 0.5 and 1 respectively. Set to 3. Minimum headway The function defined as the vehicle acceleration can be expressed by the following formula six: (Formula 6) in, is the maximum deceleration (which can be set to -8m / s in the embodiment of this application), is the comfortable deceleration (which can be set to -3m / s in the embodiment of the present application).

[0044] Based on the above formulas 1 to 6, we can know that situational awareness and task saturation The relationship between can be expressed by the following formula seven: (Formula 7) in, , is the maximum value of context perception (can be 1), It is the minimum value of situational awareness (can be 0.5). is the critical task saturation, which can be set to 0.8. is the maximum task saturation and can be set to 2.

[0045] Specifically, in the above step 101, the construction of the traffic factor risk quantification model may further include the following steps 101b and 101c: Step 101b: construct a traffic participant risk field, a traffic signal risk field, and a traffic marking risk field, and obtain an overall risk field of the intersection traffic environment based on the traffic participant risk field, the traffic signal risk field, and the traffic marking risk field.

[0046] Among them, the traffic participant risk field is used to characterize the risk expression of traffic participants to any spatial location; the traffic signal risk field is used to characterize the risk expression of periodic changes in traffic signals to any spatial location; and the traffic marking risk field is used to characterize the risk expression of traffic markings to any spatial location.

[0047] For example, in order to simultaneously model the risk characteristics of mobile and static traffic elements and simplify the risk field calculation, the embodiment of the present application designs an attenuation factor ,Using the area occupation and speed information of vehicles as the key to quantify the risk of traffic participants,,the traffic participant risk field is constructed, which can be expressed by the following formula,8: (Formula 8) in, Representation object n Risk expression (the object n It can be any traffic participant, such as other vehicles, motorcycles, bicycles, etc.). Representation object n The Euclidean norm of the risk source n At time t, spatial position The maximum risk attenuation degree of the risk at and denote the longitudinal and transverse attenuation factors, respectively, and denote the influence of speed on the longitudinal and lateral risk attenuation factors, respectively. and Represent the influence of relative distance on the vertical and horizontal risk attenuation coefficients, and Respectively represent the vertical and horizontal lengths of the object. The function can ensure the non-negativity of the attenuation factor on the vehicle occupation area.

[0048] like Figure 3 As shown, Figure 3 (a) to (c) show the risk field quantification results of moving vehicles at different speeds, where Figure 3 (a) represents a stationary vehicle, and Figure 3 (b) and Figure 3 (c) in the middle represents a moving vehicle with a longitudinal speed of 10m / s and 30m / s respectively.

[0049] For example, important static traffic elements, such as lane edges and lane lines, can also be considered as risk factors in space because they play a key role in restricting vehicle movement. Since the speed of static traffic elements is a constant 0, the risk attenuation factor of lane lines is Only its spatial impact should be considered. The traffic marking risk field can be expressed by the following formula 9 and formula 10: (Formula 9) (Formula 10) in, Indicates position in space The first iThe risk value of the lane line or edge line, is the risk attenuation factor for the lane line, and Represent the influence coefficients of the lane line in the longitudinal and lateral directions, and Respectively represent the longitudinal and lateral lengths of the lane line or road edge line.

[0050] It should be noted that lane edges or yellow lines that cannot be crossed are considered to have the highest risk (e.g., strictly prohibited), while solid lines that can be partially crossed are considered to have a certain risk, and dashed lane lines have the lowest risk. Figure 4 Middle (d), Figure 4 (e) and Figure 4 (f) shows a schematic diagram of the risk quantification of these three categories. Figure 4 (d) in the middle is the edge of the road that cannot be crossed. Figure 4 (e) is a solid line on the road. Figure 4 (f) in the middle is the dotted line on the road.

[0051] For example, when approaching an intersection, the phase of the traffic signal changes over time, which affects the driver's decision. Therefore, unlike static traffic elements (i.e., the time dimension is not considered), the risk field model of this information is also a time-periodic variation function at the stop line of the intersection. The driver can pass through the intersection during the green light phase. Therefore, it can be considered that there is no risk during the green light phase. When the signal light is red, it is prohibited to pass through the intersection, which represents the highest risk (set to 1 in this article). For yellow lights, the risk varies from person to person, and different drivers have different risk preferences. However, in general, with the start of the yellow light phase, the risk of the driver passing the intersection under traffic rules increases over time. Therefore, the risk quantification of traffic signals can be regarded as a risk field with time-periodic variation characteristics at the stop line of the intersection. The traffic signal risk field can be expressed by the following formula 11 and formula 12: (Formula 11) (Formula 12) in, represents the traffic signal risk field, represents the influence coefficient of traffic signal periodic changes on the risk field, Indicates the j traffic signal cycles, 、 and Represents the duration of green light, yellow light and red light respectively. Indicates the traffic signal cycle.

[0052] It should be noted that traffic signal risk only affects vehicles approaching from one direction. Therefore, when modeling its risk field, only spatial radiation in one direction is considered. Figure 4 Medium (g), Figure 4 Middle (h) and Figure 4 As shown in (i), it shows a schematic diagram of the risk field when the traffic signal is yellow at different times.

[0053] Step 101c: Calculate the driver's subjective perceived risk value at the intersection based on the overall risk field.

[0054] The overall risk field is used to represent the risk expression of any spatial position in the intersection; the overall risk field is obtained by superimposing the risks of various traffic elements and taking the maximum value.

[0055] For example, after obtaining the risk field models for all dynamic and static traffic elements, the risk field of the entire traffic environment can be modeled. However, unlike developing a risk field model for a single traffic element, the mutual influence between various elements in the entire traffic environment should be considered. The embodiment of the present application adopts a method of superimposing element risks and taking the maximum value to obtain the risk field of the entire traffic environment, which can be expressed by the following formula 13: (Formula 13) in, Represents the position at time t The risk value of the entire traffic environment.

[0056] For example, Figure 5 (a) shows a schematic diagram of the risk field at a signalized intersection. Based on Figure 5 (a), as shown in Figure 5 (b), the intersection includes many elements, such as a straight road, lane markings, the intersection, and vehicles moving in four directions. Currently, the intersection allows east-west traffic. A driver's subjective perception of risk may be influenced by the objective risk field of traffic elements. From the perspective of the following vehicle driver, the closer the forward object (moving or stationary) is to the front of the vehicle (on the x-axis), the higher the collision risk.

[0057] For example, in an embodiment of the present application, when a driver drives to an intersection, he or she may encounter two different situations, namely, the current vehicle driven by the driver is the lead vehicle or the non-lead vehicle. The driver's subjective perceived risk in different situations can be obtained through different calculation methods.

[0058] Specifically, in the above step 101c, the step of calculating the driver's subjective perceived risk value at the intersection for two different situations may include the following step 101c1 or step 101c2: Step 101c1: When the current vehicle driven by the driver is not the first vehicle at the intersection, the risk value of the current vehicle is calculated based on the risk field of the current vehicle, and the risk value of the preceding vehicle is calculated based on the risk field of the preceding vehicle of the current vehicle. After normalizing the risk value of the current vehicle and the risk value of the preceding vehicle, the driver's subjective perceived risk value at the intersection is obtained.

[0059] For example, when the current vehicle is the lead vehicle, when the vehicle approaches an intersection, the driver pays attention to the risk of the object ahead. Therefore, the driver's subjective perceived risk can be expressed by the following formula 14: (Formula 14) in, represents the subjective perceived risk of the driver of the following vehicle (i.e. the current vehicle), and Represent the risk field quantification of the following vehicle and the leading vehicle respectively.

[0060] For example, Figure 6 As shown in Figure 2, the closer the relative distance ds between the rear vehicle and the front vehicle, the higher the subjective risk perceived by the rear vehicle driver. According to the steady-state risk theory, the driver can reduce the subjective risk by increasing the relative distance from the front vehicle by reducing speed.

[0061] Step 101c2: When the current vehicle driven by the driver is the leading vehicle at the intersection, the risk value of the current vehicle is calculated based on the risk field of the current vehicle, and the risk value of the traffic signal on the current road is calculated based on the traffic signal risk field. After normalizing the risk value of the current vehicle and the risk value of the traffic signal on the current road, the driver's subjective perceived risk value at the intersection is obtained.

[0062] For example, when the vehicle driven by the driver is the lead vehicle, the driver can drive freely and is only affected by human factors and traffic signals. At this time, the current vehicle should only consider the situation where there is a traffic signal ahead and can use the formula 14 Replace with To calculate the driver's subjective perceived risk. And, since only the longitudinal direction is considered, the traffic signal risk field after the preview time can be expressed by the following formula 17: (Formula 15) For example, the formula calculated based on the fifteenth formula Replace the formula 14 , we can get the driver’s subjective perceived risk when the current vehicle is the lead vehicle.

[0063] For example, after constructing the relationship model between task saturation and situational awareness and the traffic element risk quantification model, we can further construct the driver's intersection approach behavior model.

[0064] Specifically, in step 101 above, the step of constructing a driver's intersection approach behavior model may further include the following steps 101d and 101e: Step 101d: Using the risk steady-state theory, determine the expected position of the vehicle after the preview time and the preview position after the preview time.

[0065] Among them, the vehicle's desired position and preview position are determined based on the driver's subjective perceived risk.

[0066] Step 101e: Based on the difference between the expected position of the vehicle after the preview time and the preview position after the preview time, a driver behavior model is constructed, and subjective human factors are embedded in the driver behavior model to obtain the driver intersection approach behavior model.

[0067] The difference between the risk value corresponding to the vehicle at the expected position and the driver's expected risk is less than a preset risk threshold; the preview position is: the driver's predicted position after the preview time.

[0068] For example, the risk homeostasis theory states that individuals maintain their internal risk level through behavioral adjustments, and reduce the difference between their perceived risk and expected risk level by increasing (or decreasing) safe behaviors. The risk compensation theory also states that drivers have a fixed acceptable risk level (i.e., expected risk). Through behavioral adjustments, they will bring their subjective perceived risk closer to this expected risk level. The preview-follow theory states that the driver's behavioral goal is to minimize the error between the actual trajectory and the expected trajectory. It assumes that the driver can have a preview time (PT) and preview position within the visual range, and control the vehicle to approach the expected position by comparing the predicted position (i.e., preview position) with the expected position, thereby forming a decision-making process. In the embodiment of the present application, it is assumed that the vehicle moves toward the expected position with a constant acceleration. Therefore, based on the above theory, the driver's behavior model can be obtained by determining the expected position and the preview position, i.e., by the following formulas 16 and 17: (Formula 16) (Formula 17) in, represents the acceleration of the vehicle, Indicates the desired position, Indicates the preview position. and are the position and velocity of the vehicle at time t respectively.

[0069] For example, the expected position is related to the driver's subjective expected risk Therefore, we can search for the corresponding The position of the desired position is determined by Figure 7 As shown, the driver's subjective perceived risk and expected risk at time t, as well as the subjective perceived risk after the preview time (t+PT).

[0070] For example, after calculating the real risk of the traffic signal, the The position (or nearby position) of the vehicle is used as the target for trajectory planning. The preceding vehicle may accelerate or decelerate during the preview time. If the vehicle's speed and acceleration constraints are set to and , then there is a reach range within the preview time (the upper limit is , the lower limit is ). If the final speed after the preview time is greater than the speed limit, the vehicle will travel at a constant speed (i.e., the speed limit) after reaching the speed threshold. This can be specifically expressed by the following formulas 18, 19, 20, and 21: (Formula 18) (Formula 19) (Formula 20) (Formula 21) in, and They represent the displacement of moving forward with deceleration and acceleration during the preview time, respectively.

[0071] Specifically, in the above step 101e, the current vehicle is the lead vehicle ( leading ) The construction of the driver behavior model may further include the following steps 101e1 and 101e2: Step 101e1: When the driver's current vehicle is the lead vehicle, the final speed of the current vehicle after the preview time is calculated based on the speed constraint and acceleration constraint of the current vehicle, and the movement displacement of the current vehicle within the preview time is calculated based on the final speed.

[0072] The movement displacement includes: the displacement of decelerating forward movement and the displacement of accelerating forward movement.

[0073] Step 101e2: Based on the movement displacement of the current vehicle within the preview time, a first behavior model of the vehicle in a braking state and a second behavior model of the vehicle in an accelerating state are constructed.

[0074] For example, by combining the above formula 16 and formula 17, the behavior models of the preceding vehicle in the braking and acceleration states (i.e., the above first behavior model and the second behavior model) can be obtained, which can be expressed by the following formula 22 and formula 23, respectively: (Formula 22) (Formula 23) Specifically, in the above step 101e, if the current vehicle is not the lead vehicle ( rear ) may further include the following steps 101e3 to 101e5: Step 101e3: When the driver's current vehicle is not the lead vehicle, the relative distance between the leading vehicle and the current vehicle after the preview time is calculated based on the difference between the preview position of the leading vehicle after the preview time and the preview position of the current vehicle after the preview time.

[0075] Step 101e4: Calculate the predicted risk of the current vehicle approaching the leading vehicle based on the ratio of the relative distance to the speed of the leading vehicle after the preview time, and calculate the expected position of the current vehicle after the preview time when the predicted risk is equal to the driver's expected risk.

[0076] Step 101e5: Construct a third behavior model based on the difference between the expected position of the current vehicle after the preview time and the preview position of the current vehicle after the preview time.

[0077] For example, when the driver's current vehicle is not the lead vehicle (i.e., the following vehicle), during the short preview time, the driver can assume that the leading vehicle is moving at a constant speed, as shown in Formulas 22 and 23 above. Therefore, the relative distance to the leading vehicle after the preview time is calculated using Formula 27, and the risk value is calculated using Formula 28. The specific formulas are as follows: (Formula 24) (Formula 25) (Formula 26) (Formula 27) (Formula 28) in, and Respectively represent the speed of the preceding vehicle at the current moment and after the preview time, and Respectively represent the current moment and the position after the preview time, and Respectively represent the position of the following vehicle at the current moment and after the preview time, represents the relative distance between the front and rear vehicles after the preview time, L is the vehicle length, and the subscripts “front” and “rear” represent the front and rear vehicles, respectively. Equation (28) represents the predicted risk when the rear vehicle approaches the front vehicle, where and Respectively represent the effects of relative distance and preceding vehicle speed on longitudinal risk.

[0078] For example, according to the above formula 28, when the risk value is equal to When , the corresponding relative distance can be calculated by the following formula: (Formula 29) For example, after the preview time, the expected position of the current vehicle can be calculated using the following formula 30: (Formula 30) For example, based on the above formula 16, the driver's behavior model when the current vehicle is not the lead vehicle can be obtained, that is, the above third behavior model can be expressed by the following formula 31: (Formula 31) Furthermore, after obtaining the above-mentioned driver behavior model, it is necessary to embed subjective human factors into the behavior model to obtain the driver's intersection approach behavior model.

[0079] Specifically, in step 101 above, the step of constructing the driver's intersection approach behavior model may further include the following steps 101f1 and 101f2: Step 101f1: Determine the driver's perception expression and the driver's reaction time expression based on the situational perception error, determine the driver's expected risk expression based on the driver's task saturation deviation, and determine the driver's heterogeneity expression based on the maximum task capability, the minimum task capability, and the driver's task capability.

[0080] Among them, the situational perception error is: the difference between the maximum situational perception and the driver's situational perception; the task saturation deviation is: the difference between the driver's task saturation and the critical task saturation; the perception expression includes: the driver's ability to judge the relative distance between the current vehicle and the preceding vehicle and the driver's ability to judge the relative speed between the current vehicle and the preceding vehicle; the relationship between the driver's situational perception and task saturation is determined based on the task saturation and situational awareness relationship model.

[0081] For example, perception is crucial to avoid accidents, and the relative distance between the rear vehicle and the front vehicle is and relative speed Misjudgment may lead to accidents. The relative distance and relative speed can be expressed by the following formula 32 and formula 33: (Formula 32) (Formula 33) in, is the size of the perception error affected by situational awareness, that is, the situational awareness error mentioned above, The above formula 32 and formula 33 can be used to set each driver's ability to judge the relative distance between the current vehicle and the vehicle in front, as well as the relative speed between the current vehicle and the vehicle in front, during simulation.

[0082] It should be noted that in the micro-simulation model in the embodiment of the application, each driver has only one value, which is either overestimated ( =1) or underestimate ( =-1) Relative distance and speed differences. A value of 0 indicates true perception.

[0083] For example, when non-driving related events attract the driver's attention, the attention lag time may increase. , which is proportional to the reduced situational awareness and can be expressed by the following formula 34: (Formula 34) in, is the maximum attention lag time, which can be set to 2 seconds. The driver's total reaction time can be expressed by the following formula: (Formula 35) For example, through the above formula 35, the reaction time of each driver can be set during simulation.

[0084] It should be noted that the driver's physical reaction time has heterogeneous characteristics, rather than uniform. Therefore, in order to reflect the individuality of the driver's reaction, this paper assumes that the physical reaction time follows the Beta distribution (i.e. ).

[0085] For example, drivers are always driven by the need to avoid accidents, so they adapt to a larger distance between vehicles when other non-driving safety-related tasks arise. Therefore, they first reduce their desired speed, which can be specifically expressed by the following formula: (Formula 36) in, is a parameter that controls the speed reduction effect (e.g. =0.5 means the maximum expected speed is reduced by 50%) is the driver response adaptability factor (the higher the task saturation, the stronger the adaptability). In addition, the driver's approximate response characteristics can be derived based on the expected time headway, which can be specifically expressed by the following formula 37: (Formula 37) For example, through the above formula 37, the expected risk of each driver can be set during simulation.

[0086] For example, heterogeneity can be considered as differences in driving skill levels. Therefore, it can be addressed by adjusting the driver's task ability. This adjustment can be expressed by the following formula 38: (Formula 38) in, and are the maximum and minimum mission capabilities, equal to 0.8 and 1.2, respectively. = represents a normal distribution. From Equations 2 and 7 above, we can see that TC affects task saturation and indirectly leads to a deterioration in situational awareness. The reduction in task capability leads to a deterioration in situational awareness.

[0087] Step 101f2: Add the perception expression, the reaction time expression, the expected risk expression, and the heterogeneity expression to the driver behavior model to obtain the driver intersection approach behavior model.

[0088] For example, after adding the subjective human factor, a driver's intersection approach behavior model can be obtained, which can be expressed by the following formula 39: (Formula 39) Exemplarily, the model integrates the driver's subjective human factors (such as perception, reaction time, expected risk, and heterogeneity characteristics) and the expected risk formed by the risk quantification of traffic elements.

[0089] Step 102: Perform intersection simulation using the driver's intersection approach behavior model by controlling variables.

[0090] Among them, the variables controlled during the simulation process include at least one of the following: the driver's perceptual bias, the driver's reaction time, the driver's expected risk, and the driver's heterogeneity; the perceptual bias is used to characterize the driver's ability to judge the relative distance and relative speed to the vehicle in front; the expected risk is used to characterize the level of risk acceptable to the driver; and the heterogeneity is used to characterize the differences in driving ability between drivers.

[0091] For example, after completing the construction of the driver's intersection approach behavior model, the driver's intersection approach behavior model can be used for simulation. During the simulation, the variables that can be controlled include: the driver's perceptual bias, the driver's reaction time, the driver's expected risk, and the driver's heterogeneity.

[0092] For example, Figure 8 and Figure 9 As shown, this is a simulation case when there is a perception bias. Figure 8 and Figure 9 Two cases of perceptual bias are presented, namely when all drivers underestimate and overestimate relative distance and speed differences, respectively, producing traffic situation and personal human factor results. Figure 8 Results are given when all drivers are homogeneous. Figure 9 Results are given when all drivers have different reaction times and expected risks. Figure 9 As can be seen from the figure, the vehicle trajectory is continuous, with severe congestion at the first intersection and stable diffusion, while the traffic flow is smooth at the subsequent intersections. The total travel time (TTS) is 828 minutes and the total stop time (TSD) is 242 minutes, indicating that the traffic lights effectively control the traffic flow and reduce the number of unnecessary stops. Figure 9 It can be seen that the traffic continuity is low, TTS decreases, and TSD increases to 299 min, indicating that heterogeneous drivers produce different traffic patterns.

[0093] like Figure 10 and Figure 11 As shown in the figure, there is a simulation case of perception bias and reaction time difference. When there is a perception bias, it will affect the driver's reaction time, which will further produce different traffic dynamics, such as Figure 10 and Figure 11 As shown in Figure 2, the TSD increases to 327 minutes and 334 minutes in both cases, respectively. This indicates that the dynamic change in reaction time reduces the vehicle's driving efficiency.

[0094] Furthermore, to avoid a possible collision, the driver reduces speed and increases desired headway, which is known as response adaptation. Figure 12 and Figure 13 Simulation results are presented for the presence of perceptual bias, reaction dynamics, and expected risk. As expected, the TSD is greater when reaction adaptation is present. This is because a reduction in vehicle speed means an increase in relative speed, and the increase in headway further reduces acceleration. Furthermore, drivers are more "conservative," which is safer but less efficient, leading to increased stopping time and congestion. The TSD for the homogeneous driver traffic scenario is 398 minutes ( Figure 12), the TSD of the heterogeneous driver traffic scenario is 476min ( Figure 13 ). Please note that Figure 12 The scenario shown involves a rear-end collision (marked in red).

[0095] In the last case, all human factors are considered, including perceptual bias, reaction time dynamics, response adaptation, and driver heterogeneity (i.e., task ability follows a normal distribution). The results are as follows Figure 14 and Figure 15 Compared to the base scenario, the TSD of the homogeneous driver group increased significantly, while the TTS of the heterogeneous driver group decreased. This is because the heterogeneous drivers had a physical reaction time of less than 0.2 seconds and an expected risk greater than 0.317, which allowed for higher overall driving efficiency (TTS of 832 minutes versus 758 minutes).

[0096] The intersection micro-modeling and simulation method provided in the embodiments of the present application: 1. It is capable of intrinsically simulating and predicting individual driver behavior; 2. It has explainable subjective human factors and mathematical logic for quantifying subjective risks so that large-scale simulations (for example, 300 or more vehicles approaching an intersection) can be performed; 3. It can still reproduce reliable vehicle trajectories and macro-traffic patterns; 4. It is sufficiently simplified, scalable, and easily accessible so that it can be easily expanded by further enriching complex human factors.

[0097] The intersection micro-modeling and simulation method provided in the embodiments of the present application first constructs a task saturation and situational awareness relationship model and a traffic element risk quantification model, and then constructs a driver intersection approach behavior model based on the task saturation and situational awareness relationship model and the traffic element risk quantification model. The task saturation and situational awareness relationship model is used to characterize the relationship between the driver's task saturation and situational awareness. The situational awareness is used to characterize the driver's understanding of the current situation and foresight of the future state based on the perception of internal and external environmental information. The traffic element risk quantification model is used to characterize the risk expression of each traffic element and the driver's subjective perceived risk at the intersection. The intersection approach behavior model combines the driver's subjective human factors and the expected risk of the trip after the risk quantification of each traffic element. The intersection approach behavior model is used to characterize the driver's behavioral expression at the intersection. Then, the driver's intersection approach behavior model is used to perform intersection simulation by controlling variables. In this way, the driver's intersection approach behavior model can not only inherently simulate and predict individual driver behavior, but also has explainable subjective human factors and subjective risk quantification mathematical logic, and is sufficiently simplified.

[0098] It should be noted that the intersection micro-modeling and simulation method provided in the embodiments of the present application can be executed by an intersection micro-modeling and simulation device, or a control module in the intersection micro-modeling and simulation device for executing the intersection micro-modeling and simulation method. In the embodiments of the present application, the intersection micro-modeling and simulation device executing the intersection micro-modeling and simulation method is used as an example to illustrate the intersection micro-modeling and simulation device provided in the embodiments of the present application.

[0099] It should be noted that the intersection micromodeling and simulation methods shown in the figures in the embodiments of this application are each illustrated by way of example in conjunction with one of the figures in the embodiments of this application. In specific implementations, the intersection micromodeling and simulation methods shown in the figures in the figures in the above methods may also be implemented in conjunction with any other combinable figures shown in the above embodiments, and will not be further described here.

[0100] The following describes the intersection micro-modeling and simulation device provided by the present application. The intersection micro-modeling and simulation method described below and described above can refer to each other.

[0101] Figure 16 A schematic diagram of the structure of the intersection micro-modeling simulation device provided in the embodiment of the present application is shown as follows: Figure 16 As shown, specifically including: The model construction module 1601 is used to construct a task saturation and situational awareness relationship model and a traffic element risk quantification model, and based on the task saturation and situational awareness relationship model and the traffic element risk quantification model, construct a driver's intersection approach behavior model; the task saturation and situational awareness relationship model is used to characterize the logical relationship between the driver's task saturation and situational awareness; the situational awareness is used to characterize the driver's understanding of the current situation and foresight of the future state based on the perception of internal and external environmental information; the traffic element risk quantification model is used to characterize the risk expression of each traffic element and the driver's subjective perception of risk at the intersection; the intersection approach behavior model combines the driver's The expected risk of the trip after quantification of subjective human factors and risks of various traffic elements; the intersection approach behavior model is used to characterize the driver's behavioral expression at the intersection; the simulation module 1602 is used to use the driver's intersection approach behavior model to simulate the intersection by controlling variables; wherein the variables controlled during the simulation process include at least one of the following: the driver's perceptual bias, the driver's reaction time, the driver's expected risk and the driver's heterogeneity; the perceptual bias is used to characterize the driver's ability to judge the relative distance and relative speed to the vehicle ahead; the expected risk is used to characterize the level of risk acceptable to the driver; the heterogeneity is used to characterize the differences in driving ability between drivers.

[0102] Optionally, the model building module 1601 is specifically used to obtain a set of task requirements by accumulating the basic graph of each driver's task requirements, and determine the logical relationship between the driver's task saturation and situational awareness by analyzing the impact of the accumulated task requirements on the driver's preferences and situational awareness; wherein, when the driver's task saturation is less than the critical task saturation, the driver's situational awareness takes the maximum value; when the driver's task saturation is greater than or equal to the maximum task saturation, the driver's situational awareness takes the minimum value; when the driver's task saturation is greater than or equal to the critical task saturation and less than the maximum task saturation, the driver's situational awareness takes the value of: the difference between the maximum situational awareness value and the target calculation result; the target calculation result is: the product of the target ratio and the first difference; the first difference is: the difference between the maximum situational awareness value and the minimum situational awareness value; the target ratio is: the ratio of the second difference to the third difference; the second difference is: the difference between the driver's task saturation and the critical task saturation; the third difference is: the difference between the maximum situational awareness value and the critical task saturation.

[0103] Optionally, the traffic element risk quantification model includes: a traffic participant risk field, a traffic signal risk field and a traffic marking risk field; the model construction module 1601 is specifically used to construct the traffic participant risk field, the traffic signal risk field and the traffic marking risk field, and obtain the overall risk field of the intersection traffic environment based on the traffic participant risk field, the traffic signal risk field and the traffic marking risk field; the model construction module 1601 is also specifically used to calculate the driver's subjective perceived risk value at the intersection based on the overall risk field; wherein, the traffic participant risk field is used to characterize the risk expression of traffic participants to any spatial position; the traffic signal risk field is used to characterize the risk expression of traffic signal periodic changes to any spatial position; the traffic marking risk field is used to characterize the risk expression of traffic markings to any spatial position; the overall risk field is used to characterize the risk expression of any spatial position in the intersection; the overall risk field is: obtained by superimposing the risks of various traffic elements and taking the maximum value.

[0104] Optionally, the model building module 1601 is specifically used to calculate the risk value of the current vehicle based on the risk field of the current vehicle, and calculate the risk value of the traffic signal of the current road based on the traffic signal risk field, when the current vehicle driven by the driver is the leading vehicle at the intersection, and normalize the risk value of the current vehicle and the risk value of the traffic signal of the current road to obtain the driver's subjective perceived risk value at the intersection; the model building module 1601 is also specifically used to calculate the risk value of the current vehicle based on the risk field of the current vehicle, and calculate the risk value of the preceding vehicle based on the risk field of the preceding vehicle of the current vehicle, and normalize the risk value of the current vehicle and the risk value of the preceding vehicle to obtain the driver's subjective perceived risk value at the intersection, when the current vehicle driven by the driver is not the leading vehicle at the intersection.

[0105] Optionally, the model construction module 1601 is specifically used to determine the expected position of the vehicle after the preview time and the preview position after the preview time by using the risk steady-state theory; the expected position and preview position of the vehicle are determined based on the driver's subjective perceived risk; the model construction module 1601 is also specifically used to construct a driver behavior model based on the difference between the expected position of the vehicle after the preview time and the preview position after the preview time, and embed subjective human factors into the driver behavior model to obtain the driver intersection approach behavior model.

[0106] Optionally, the model building module 1601 is specifically used to calculate the final speed of the current vehicle after the preview time based on the speed constraint and acceleration constraint of the current vehicle when the driver's current vehicle is the lead vehicle, and calculate the movement displacement of the current vehicle within the preview time based on the final speed; the movement displacement includes: the displacement of decelerating forward movement and the displacement of accelerating forward movement; the model building module 1601 is also specifically used to construct a first behavior model of the vehicle in a braking state and a second behavior model of the vehicle in an accelerating state based on the movement displacement of the current vehicle within the preview time.

[0107] Optionally, the model building module 1601 is specifically used to calculate the relative distance between the front vehicle and the current vehicle after the preview time based on the difference between the preview position of the front vehicle after the preview time and the preview position of the current vehicle after the preview time when the driver's current vehicle is not the lead vehicle; the model building module 1601 is also specifically used to calculate the predicted risk of the current vehicle when approaching the front vehicle based on the ratio of the relative distance to the speed of the front vehicle after the preview time, and calculate the expected position of the current vehicle after the preview time when the predicted risk is equal to the driver's expected risk; the model building module 1601 is also specifically used to construct a third behavior model based on the difference between the expected position of the current vehicle after the preview time and the preview position of the current vehicle after the preview time.

[0108] Optionally, the model construction module 1601 is specifically used to determine the driver's perception expression and the driver's reaction time expression based on the situational perception error, determine the driver's expected risk expression based on the driver's task saturation deviation, and determine the driver's heterogeneity expression based on the maximum task capability, the minimum task capability and the driver's task capability; the model construction module 1601 is also specifically used to add the perception expression, the reaction time expression, the expected risk expression and the heterogeneity expression to the driver behavior model to obtain the driver's intersection approach behavior model; wherein, the situational perception error is: the difference between the situational perception maximum value and the driver's situational perception; the task saturation deviation is: the difference between the driver's task saturation and the critical task saturation; the perception expression includes: the driver's ability to judge the relative distance between the current vehicle and the preceding vehicle and the driver's ability to judge the relative speed between the current vehicle and the preceding vehicle; the relationship between the driver's situational perception and task saturation is determined based on the task saturation and situational awareness relationship model.

[0109] The intersection micro-modeling and simulation device provided by the present application first constructs a task saturation and situational awareness relationship model and a traffic element risk quantification model, and then constructs a driver's intersection approach behavior model based on the task saturation and situational awareness relationship model and the traffic element risk quantification model. The task saturation and situational awareness relationship model is used to characterize the relationship between the driver's task saturation and situational awareness. The situational awareness is used to characterize the driver's understanding of the current situation and foresight of the future state based on the perception of internal and external environmental information. The traffic element risk quantification model is used to characterize the risk expression of each traffic element and the driver's subjective perceived risk at the intersection. The intersection approach behavior model combines the driver's subjective human factors and the expected risk of the trip after the risk quantification of each traffic element. The intersection approach behavior model is used to characterize the driver's behavioral expression at the intersection. Then, the intersection simulation is performed using the driver's intersection approach behavior model by controlling variables. In this way, the system can not only inherently simulate and predict individual driver behavior, but also has explainable subjective human factors and subjective risk quantification mathematical logic, and is sufficiently simplified.

[0110] Figure 17 An example of a physical structure diagram of an electronic device is shown below. Figure 17As shown, the electronic device may include: a processor 1710 , a communication interface 1720 , a memory 1730 and a communication bus 1740 , wherein the processor 1710 , the communication interface 1720 and the memory 1730 communicate with each other via the communication bus 1740 . Processor 1710 can call logic instructions in memory 1730 to execute an intersection micro-modeling simulation method, which includes: first, constructing a task saturation and situational awareness relationship model and a traffic element risk quantification model, and constructing a driver's intersection approach behavior model based on the task saturation and situational awareness relationship model and the traffic element risk quantification model; the task saturation and situational awareness relationship model is used to characterize the relationship between the driver's task saturation and situational awareness; the situational awareness is used to characterize the driver's understanding of the current situation and foresight of the future state based on the perception of internal and external environmental information; the traffic element risk quantification model is used to characterize the risk expression of each traffic element and the driver's subjective perceived risk at the intersection; the intersection approach behavior model combines the driver's subjective human factors and the expected risk of the trip after the risk quantification of each traffic element; the intersection approach behavior model is used to characterize the driver's behavioral expression at the intersection; thereafter, by controlling variables, the driver's intersection approach behavior model is used to perform intersection simulation. In this way, it is not only possible to inherently simulate and predict individual driver behavior, but also has explainable subjective human factors and subjective risk quantification mathematical logic, and is sufficiently simplified.

[0111] In addition, the logical instructions in the above-mentioned memory 1730 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present application, or the part that contributes to the existing technology, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program code.

[0112] On the other hand, the present application also provides a computer program product, which includes a computer program stored on a computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the intersection micro-modeling simulation method provided by the above methods, the method comprising: first, constructing a task saturation and situational awareness relationship model and a traffic element risk quantification model, and based on the task saturation and situational awareness relationship model and the traffic element risk quantification model, constructing a driver's intersection approach behavior model; the task saturation and situational awareness relationship model is used to characterize the driver's task The relationship between saturation and situational awareness; situational awareness, which characterizes the driver's understanding of the current situation and foresight of future states based on their perception of internal and external environmental information; the traffic element risk quantification model, which characterizes the risk expression of each traffic element and the driver's subjective perceived risk at an intersection; the intersection approach behavior model, which combines the driver's subjective human factors and the expected risk of the trip after risk quantification of each traffic element; and the intersection approach behavior model, which characterizes the driver's behavioral expression at an intersection. Subsequently, the driver's intersection approach behavior model is used to perform intersection simulation by controlling variables. This approach not only inherently simulates and predicts individual driver behavior, but also incorporates explainable subjective human factors and the mathematical logic of subjective risk quantification, while being sufficiently simplified.

[0113] On the other hand, the present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the above-mentioned intersection micro-modeling and simulation methods, the method comprising: first, constructing a task saturation and situational awareness relationship model and a traffic element risk quantification model, and constructing a driver's intersection approach behavior model based on the task saturation and situational awareness relationship model and the traffic element risk quantification model; the task saturation and situational awareness relationship model is used to characterize the relationship between the driver's task saturation and situational awareness; the situational awareness is used to characterize the driver's understanding of the current situation and foresight of the future state based on the perception of internal and external environmental information; the traffic element risk quantification model is used to characterize the risk expression of each traffic element and the driver's subjective perceived risk at the intersection; the intersection approach behavior model combines the driver's subjective human factors and the expected risk of the journey after the risk quantification of each traffic element; the intersection approach behavior model is used to characterize the driver's behavioral expression at the intersection; thereafter, by controlling variables, the driver's intersection approach behavior model is used to perform intersection simulation. In this way, it is not only possible to inherently simulate and predict individual driver behavior, but also has explainable subjective human factors and subjective risk quantification mathematical logic, and is sufficiently simplified.

[0114] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0115] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0116] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A microscopic modeling and simulation method for an intersection, characterized in that: include: Constructing a task saturation and situational awareness relationship model and a traffic element risk quantification model, and constructing a driver's intersection approach behavior model based on the task saturation and situational awareness relationship model and the traffic element risk quantification model; The task saturation and situational awareness relationship model is used to characterize the relationship between the driver's task saturation and situational awareness; Situational awareness is used to represent the driver's understanding of the current situation and foresight of future states based on the perception of internal and external environmental information; The traffic element risk quantification model is used to characterize the risk expression of each traffic element and the driver's subjective perceived risk at the intersection; the intersection approach behavior model combines the driver's subjective human factors and the expected risk of the trip after the risk of each traffic element is quantified; the intersection approach behavior model is used to characterize the driver's behavioral expression at the intersection; Performing intersection simulation using the driver's intersection approach behavior model by controlling variables; Among them, the variables controlled during the simulation process include at least one of the following: the driver's perceptual bias, the driver's reaction time, the driver's expected risk, and the driver's heterogeneity; the perceptual bias is used to characterize the driver's ability to judge the relative distance and relative speed to the vehicle in front; the expected risk is used to characterize the level of risk acceptable to the driver; and the heterogeneity is used to characterize the differences in driving ability between drivers.

2. The method according to claim 1, characterized in that The construction of the relationship model between task saturation and situational awareness includes: The task requirement set is obtained by accumulating the basic graph of each driver's task requirements. The logical relationship between the driver's task saturation and situational awareness is determined by analyzing the impact of the accumulated task requirements on the driver's preferences and situational awareness. Among them, when the driver's task saturation is less than the critical task saturation, the driver's situational awareness takes the maximum value; when the driver's task saturation is greater than or equal to the maximum task saturation, the driver's situational awareness takes the minimum value; when the driver's task saturation is greater than or equal to the critical task saturation and less than the maximum task saturation, the driver's situational awareness is taken as: the difference between the maximum situational awareness value and the target calculation result; the target calculation result is: the product of the target ratio and the first difference; the first difference is: the difference between the maximum situational awareness value and the minimum situational awareness value; the target ratio is: the ratio of the second difference to the third difference; the second difference is: the difference between the driver's task saturation and the critical task saturation; the third difference is: the difference between the maximum situational awareness value and the critical task saturation.

3. The method according to claim 1 or 2, characterized in that The traffic element risk quantification model includes: traffic participant risk field, traffic signal risk field and traffic marking risk field; The construction of the traffic element risk quantification model includes: Constructing a traffic participant risk field, a traffic signal risk field, and a traffic marking risk field, and obtaining an overall risk field of the intersection traffic environment based on the traffic participant risk field, the traffic signal risk field, and the traffic marking risk field; Calculating a driver's subjective perceived risk value at the intersection based on the overall risk field; Among them, the traffic participant risk field is used to characterize the risk expression of traffic participants to any spatial location; the traffic signal risk field is used to characterize the risk expression of traffic signal periodic changes to any spatial location; the traffic marking risk field is used to characterize the risk expression of traffic markings to any spatial location; the overall risk field is used to characterize the risk expression of any spatial location in the intersection; the overall risk field is: obtained by superimposing the risks of various traffic elements and taking the maximum value.

4. The method according to claim 3, characterized in that Calculating the driver's subjective perceived risk value at the intersection based on the overall risk field includes: When the driver's current vehicle is the leading vehicle at an intersection, a risk value for the current vehicle is calculated based on the risk field of the current vehicle, and a risk value for the traffic signal on the current road is calculated based on the traffic signal risk field. The risk values of the current vehicle and the traffic signal on the current road are normalized to obtain the driver's subjective perceived risk value at the intersection. or, When the driver's current vehicle is not the first vehicle at an intersection, the risk value of the current vehicle is calculated based on the risk field of the current vehicle, and the risk value of the preceding vehicle is calculated based on the risk field of the preceding vehicle. After normalizing the risk values of the current vehicle and the preceding vehicle, the driver's subjective perceived risk value at the intersection is obtained.

5. The method according to claim 1 or 2, characterized in that The constructing of a driver's intersection approach behavior model based on the task saturation and situational awareness relationship model and the traffic element risk quantification model includes: Using the risk steady-state theory, the vehicle's expected position after the preview time and the preview position after the preview time are determined; the vehicle's expected position and preview position are determined based on the driver's subjective perception of risk; A driver behavior model is constructed based on a difference between a desired position of the vehicle after a preview time and a preview position after the preview time, and subjective human factors are embedded into the driver behavior model to obtain the driver intersection approach behavior model.

6. The method according to claim 5, characterized in that The driver behavior model is constructed based on the difference between the expected position of the vehicle after the preview time and the preview position after the preview time, including: When the driver's current vehicle is the lead vehicle, the final speed of the current vehicle after the preview time is calculated based on the speed constraint and acceleration constraint of the current vehicle, and the movement displacement of the current vehicle during the preview time is calculated based on the final speed; the movement displacement includes the displacement of decelerating forward movement and the displacement of accelerating forward movement; Based on the movement displacement of the current vehicle within the preview time, a first behavior model of the vehicle in a braking state and a second behavior model of the vehicle in an accelerating state are constructed.

7. The method according to claim 5 or 6, characterized in that The driver behavior model is constructed based on the difference between the expected position of the vehicle after the preview time and the preview position after the preview time, including: When the driver's current vehicle is not the lead vehicle, calculating the relative distance between the leading vehicle and the current vehicle after the preview time based on the difference between the preview position of the leading vehicle after the preview time and the preview position of the current vehicle after the preview time; calculating a predicted risk of the current vehicle approaching the leading vehicle based on a ratio of the relative distance to the speed of the leading vehicle after the preview time, and calculating an expected position of the current vehicle after the preview time when the predicted risk equals the driver's expected risk; A third behavior model is constructed based on a difference between the expected position of the current vehicle after the preview time and the preview position of the current vehicle after the preview time.

8. The method according to any one of claims 5 to 7, characterized in that The embedding of subjective human factors into the driver behavior model to obtain the driver intersection approach behavior model includes: Determine the driver's perception expression and reaction time expression based on situational perception error, determine the driver's expected risk expression based on the driver's task saturation deviation, and determine the driver's heterogeneity expression based on the maximum task capability, minimum task capability and the driver's task capability; Adding the perception expression, the reaction time expression, the expected risk expression, and the heterogeneity expression to the driver behavior model to obtain the driver intersection approach behavior model; Among them, the situational perception error is: the difference between the maximum situational perception and the driver's situational perception; the task saturation deviation is: the difference between the driver's task saturation and the critical task saturation; the perception expression includes: the driver's ability to judge the relative distance between the current vehicle and the preceding vehicle and the driver's ability to judge the relative speed between the current vehicle and the preceding vehicle; the relationship between the driver's situational perception and task saturation is determined based on the task saturation and situational awareness relationship model.

9. A microscopic modeling and simulation device for an intersection, characterized in that: The device comprises: A model construction module is used to construct a task saturation and situational awareness relationship model and a traffic element risk quantification model, and based on the task saturation and situational awareness relationship model and the traffic element risk quantification model, a driver's intersection approach behavior model is constructed; the task saturation and situational awareness relationship model is used to characterize the logical relationship between the driver's task saturation and situational awareness; the situational awareness is used to characterize the driver's understanding of the current situation and foresight of the future state based on the perception of internal and external environmental information; the traffic element risk quantification model is used to characterize the risk expression of each traffic element and the driver's subjective perceived risk at the intersection; the intersection approach behavior model combines the driver's subjective human factors and the expected risk of the trip after the risk of each traffic element is quantified; the intersection approach behavior model is used to characterize the driver's behavioral expression at the intersection; A simulation module, configured to perform intersection simulation using the driver's intersection approach behavior model by controlling variables; Among them, the variables controlled during the simulation process include at least one of the following: the driver's perceptual bias, the driver's reaction time, the driver's expected risk, and the driver's heterogeneity; the perceptual bias is used to characterize the driver's ability to judge the relative distance and relative speed to the vehicle in front; the expected risk is used to characterize the level of risk acceptable to the driver; and the heterogeneity is used to characterize the differences in driving ability between drivers.

10. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the intersection micro-modeling and simulation method according to any one of claims 1 to 8 are implemented.

Images