Modeling and Risk Identification Method and Device for Human-Vehicle-Road Micro Traffic System

Abstract

This application discloses a method and device for modeling and risk identification of a human-vehicle-road micro traffic system, in which the method includes: building a vehicle-lane dynamic interaction action model, obtaining potential accident consequences caused by vehicle-road interaction, building a human-vehicle dynamic interaction action model, obtaining the behavioral uncertainty caused by human-vehicle interaction, building a human-lane dynamic interaction action model, obtaining the driver's risk sensitivity difference in the human-road interaction process, using the idea of ​​clustering to characterize the characteristic discipline and difference of the driver's driving habits, obtaining the driver's personalized characteristics, and building a human-vehicle-road closed-loop dynamic system based on the potential accident consequences caused by vehicle-road interaction, the behavioral uncertainty, the driver's risk sensitivity difference in the human-road interaction process, and the driver's personalized characteristics, to generate a risk identification result, thereby characterizing the influence of the driver's factors, vehicle motion state, and road environment information interaction process on the system safety state, and realizing system risk identification and graded warning.

Description

Technical Field

[0001] This application claims priority to the Chinese patent application No. 202111131805.6, titled "Modeling and Risk Identification Method and Device for Human-Vehicle-Road Micro Traffic System", filed by Tsinghua University on September 26, 2021.

[0002] This application relates to the field of autonomous driving technology, and particularly to a method and device for modeling and risk identification of a human-vehicle-road micro traffic system.

Background Art

[0003] When a vehicle is in motion, the road environment is complex and diverse, and its safety state is affected by many factors. Generally speaking, there are three major factors: people (physiological state, driving characteristics, intentions, etc.), vehicles (operating conditions), and roads (weather, road conditions, traffic flow, etc.). In addition to uncertain factors such as drivers' lack of advance preparation, delayed reactions, and driving mistakes, potential vehicle failures, difficulty in predicting the intentions of surrounding vehicles, and deterioration of the road environment make it difficult to fundamentally and completely eliminate traffic accidents. Furthermore, the characteristics of drivers, the intentions of surrounding vehicles, and the uncertainty of the road environment constantly spread and expand according to the non-linear iteration of the human-vehicle-road closed-loop system, making it difficult to predict the dynamic characteristics of vehicles. Therefore, by clarifying the interactions between humans, vehicles, and roads, accurately representing the impacts of each traffic element on the system, and quantitatively identifying the dynamic risks of complex systems, it is possible to better adapt to different conditions of drivers, vehicles, and road environments, thereby achieving the balance of the human-vehicle-road micro traffic system.

[0004] In related technologies, the modeling of the interaction effects among humans, vehicles, and roads, especially the interactions between humans and vehicles, vehicles and roads, and humans and roads, has already received extensive research. The three elements of the driver (human), vehicle (car), and traffic environment (road) and their interactions constitute the basic unit of the microscopic traffic system centered around the host vehicle. In current research, it is often extended from a single object to the perspective of the interactions between humans and vehicles, humans and roads, and vehicles and roads. That is, it rarely starts from the perspective of the integration of humans, vehicles, and roads. Specifically, although the research on the combination of vehicles and roads in the past has become increasingly in-depth, the mechanism of the combination of vehicles and roads under extreme conditions has not yet been clearly understood. Currently, the research on the mechanism of vehicle-road combination is often carried out based on the action mechanism of tire dynamics-road dynamics, and it is difficult to expand and construct a closed-loop system model of humans, vehicles, and roads from a unified perspective. On the other hand, the research on human-vehicle interaction often focuses on the technical overcoming towards advanced driver assistance systems, mainly aiming at vehicle safety. There are few systems that consider the acceptance degree of drivers towards the assistance system, and there is a limit in terms of inferior comfort. It is necessary to deeply understand the driving cognitive characteristics of drivers, construct their cognitive models, break through the trust problem in the human-vehicle interaction process, and realize the cooperative driving of humans and vehicles. On the other hand, the research on the interaction mechanism between humans and roads mainly focuses on the research related to the intention of the driving behavior of a single vehicle. Since the research on the intention recognition of the interaction is not yet deep, it is necessary to construct a model of the interaction mechanism between humans and roads, clarify the interaction mechanism between the driver of the host vehicle and surrounding traffic participants, and provide theoretical support for the active control to avoid collision risks. The research on the closed-loop system of humans, vehicles, and roads often focuses within the scope of a narrow concept, and is overly simplified for certain elements. Although it covers the traffic elements of humans, vehicles, and roads, it lacks a unified and complete system description. Therefore, a new extensible integrated modeling method for humans, vehicles, and roads is strongly required to describe this complex system, and on this basis, grasp the influence rules of traffic environment changes on vehicle operation safety, and provide theoretical guidance for traffic managers or vehicle motion control.

[0005] Road traffic risks are considered to be the result of a combination of many factors, mainly manifested as risks caused by the physical and psychological limitations of drivers, the limitations of vehicle performance, inappropriate road routes, and adverse weather conditions (such as poor visibility and slippery roads). For a complex coupling object like the human-vehicle-road closed-loop system, it is difficult to describe the entire risk conversion process from the start of risk formation during driving to the occurrence of a risk collision with a single spatio-temporal distance parameter (such as the actual inter-vehicle distance, time headway, and collision time). It is necessary to comprehensively consider multiple spatio-temporal distance parameters and use more complex models and algorithms to study the operating risks of the system. However, currently, the modeling of the human-vehicle-road traffic system itself based on methods such as vehicle kinematics and collision probability is unclear, and it is difficult to effectively feedback the risk identification results to the human-vehicle-road traffic system to make safety support decisions. Therefore, in a complex human-vehicle-road traffic system, it is necessary to fully consider the multi-source multi-dimensional risk generation process in the system, recognize the vehicle driving safety state in the human-vehicle-road combined environment, construct a risk identification model for the human-vehicle-road closed-loop system, and further better realize the inherent safety of the operation of the human-vehicle-road closed-loop system. Therefore, it is necessary to develop a modeling and risk identification method for the human-vehicle-road microscopic traffic system.

Summary of the Invention

Problems to be Solved by the Invention

[0006] This application provides a modeling and risk identification method for a human-vehicle-road microscopic traffic system to solve problems such as the unclear modeling of the human-vehicle-road traffic system itself in related technologies and the difficulty of effectively feeding back the risk identification results to the human-vehicle-road traffic system to take safety support measures.

Means for Solving the Problems

[0007] An embodiment of the first aspect of the present application provides a method for modeling and risk identification of a human-vehicle-road micro-traffic system, which constructs a vehicle-road dynamic interaction model by utilizing the interaction effects between different types of vehicles and road traffic participants, and obtains the results of potential accidents caused by the interaction between the vehicle and the road; constructs a human-vehicle dynamic interaction model by using the driving trajectory distribution output by the interaction of the human-vehicle system, and obtains the uncertainty of behaviors caused by the interaction between humans and vehicles; constructs a human-road dynamic interaction model for the area of interest where the driver observes the road environment during driving by utilizing the visual characteristics of the driver, and obtains the differences in driver risk sensitivity in the human-road interaction process; characterizes the characteristic laws and differences of the driving habits of the driver by using the concept of clustering to obtain the personalized characteristics of the driver; constructs a human-vehicle-road closed-loop dynamics system based on the results of potential accidents caused by the vehicle-road interaction, the uncertainty of the behaviors, the differences in driver risk sensitivity in the human-road interaction process, and the personalized characteristics of the driver, and generates a risk identification result.

[0008] Optionally, in one embodiment of the present application, the human-vehicle-road closed-loop dynamics system is JPEG0007717409000001.jpg54140 represented as Among them, JPEG0007717409000002.jpg511 is the human-vehicle-road closed-loop dynamics system, JPEG0007717409000003.jpg611 is an effective response in the differences in driver risk sensitivity in the human-road interaction process, JPEG0007717409000004.jpg55 is the personalized characteristics of the driver, JPEG0007717409000005.jpg610 is the steering angle of the vehicle, JPEG0007717409000006.jpg832 is the uncertainty of the behaviors, JPEG0007717409000007.jpg64 is the radius of the driving interest area in the difference of the driver risk sensitivity in the above-mentioned human-road interaction process, JPEG0007717409000008.jpg75 is the result of potential accidents due to the above-mentioned vehicle-road interaction, JPEG0007717409000009.jpg34 is the steering, JPEG0007717409000010.jpg414 represents that the vehicle is going straight, JPEG0007717409000011.jpg514 represents that the vehicle turns left, JPEG0007717409000012.jpg514 represents that the vehicle turns right, JPEG0007717409000013.jpg57 is the increase amount of the steering angle within one period Δt, JPEG0007717409000014.jpg46 is the host vehicle JPEG0007717409000015.jpg42 is the interaction matrix where the host vehicle JPEG0007717409000016.jpg52 interacts with road traffic participants JPEG0007717409000017.jpg519 and JPEG0007717409000018.jpg515 are respectively the mass and speed of the host vehicle i and another road traffic participant j, JPEG0007717409000019.jpg410 is the collision avoidance time, JPEG0007717409000020.jpgLet it be 636, JPEG0007717409000021.jpg34 is the length of the major axis of the ellipse, JPEG0007717409000022.jpg53 is the length of the minor axis of the ellipse, JPEG0007717409000023.jpg611 is the speed of the host vehicle i, JPEG0007717409000024.jpg713 is the driver's viewing angle function.

[0009] Optionally, in one embodiment of the present application, to construct a vehicle-road dynamic interaction model by using the interaction effects between different types of vehicles and road traffic participants, and to obtain the results of potential accidents due to the interaction between the vehicle and the road, the steps include calculating the number of interactions between the host vehicle in the scene and different road traffic participants, calculating the results of the interaction between two objects in the traffic environment, and based on the number of interactions and the results of the interaction, in the process of multiple traffic participants interacting with the host vehicle, the results of potential accidents due to the vehicle-road interaction obtained by superimposing the cumulative interaction results JPEG0007717409000025.jpg75: generating JPEG0007717409000026.jpg1165, and Among them, JPEG0007717409000027.jpg46 is an interaction matrix in which the host vehicle i interacts with the road traffic participant j, JPEG0007717409000028.jpg519 and JPEG0007717409000029.jpg515 are respectively the mass and speed of the host vehicle i and another road traffic participant j.

[0010] Optionally, in one embodiment of the present application, the step of constructing a human-vehicle dynamic interaction action model using the driving trajectory distribution output by the interaction of the human-vehicle system and obtaining the uncertainty of the behavior caused by the interaction between the human and the vehicle includes calculating an equivalent linear two-wheeled vehicle model based on the vehicle kinematic model and the turning radius, and calculating a predicted position based on the commanded steering angle; predicting the uncertain motion of the driver-vehicle system based on the collected actual driving experiment data; obtaining the Gaussian normal distribution of the turning angle of the steering angle of the vehicle based on the predicted position and the uncertain motion, and performing parameter determination based on the Gaussian normal distribution to determine the uncertainty of the behavior caused by the interaction between the human and the vehicle. JPEG0007717409000030.jpg823: JPEG0007717409000031.jpg1491 and includes the step of determining the above. Among them, JPEG0007717409000032.jpg610 is the steering angle of the vehicle, JPEG0007717409000033.jpg34 is the dispersion of the data distribution following the normal distribution, JPEG0007717409000034.jpg44 is the mean value of the random variable following the normal distribution.

[0011] Optionally, in one embodiment of the present application, the step of constructing a human-road dynamic interaction action model for the region of interest where the driver observes the road environment during driving using the visual characteristics of the driver and obtaining the difference in the driver's risk sensitivity during the human-road interaction process includes calculating the effective response of the driver during normal driving using the driver's visual range; using the elliptical distribution characteristic of the driver's vision to calculate the length of the equipotential line of the driver's dynamic visual range. Calculating the effective response of the driver during normal driving using the driver's visual range; Using the elliptical distribution characteristic of the driver's vision to calculate the length of the equipotential line of the driver's dynamic visual range. JPEG0007717409000035.jpg65: The step of obtaining JPEG0007717409000036.jpg1182, Among them, JPEG0007717409000037.jpg64 is the radius of the size of the region of interest, JPEG0007717409000038.jpg34 is the length of the major axis of the ellipse, JPEG0007717409000039.jpg53 is the length of the minor axis of the ellipse, JPEG0007717409000040.jpg34 and The values of JPEG0007717409000041.jpg53 are related to the vehicle speed of the host vehicle JPEG0007717409000042.jpg42, JPEG0007717409000043.jpg611 is related to the vehicle speed of the host vehicle JPEG0007717409000044.jpg713 is the driver's viewing angle function, and Using the influence of the change in the driver's viewing angle on the driver, sensing the relative speed of the vehicle, obtaining the influence of the driver's dynamic perception of the visual field, and the radius of the size of the region of interest for observing the environment around the road during driving: JPEG0007717409000045.jpg19147 The step of generating, Among them, JPEG0007717409000046.jpg410 is the collision avoidance time, Taking JPEG0007717409000047.jpg628, JPEG0007717409000048.jpg611 is the speed of the host vehicle i, and Based on the effective response and the radius of the size of the region of interest, determining the difference in the driver risk sensitivity in the human-road interaction process.

[0012] Optionally, in one embodiment of the present application, the step of characterizing the characteristic laws and differences of the driving habits of drivers using the concept of clustering to obtain the personalized characteristics of drivers includes distinguishing the similarity of operation behaviors and trajectories of various types of drivers in the same scene, obtaining the personalized expressions of drivers based on the non-supervised clustering method, obtaining the characteristic expressions of each driver based on the personalized expressions of the drivers, obtaining the driving characteristics based on the distance from the clustering center, and determining the personalized characteristics of the drivers.

[0013] The embodiment of the second aspect of the present application provides a modeling and risk identification device for a human-vehicle-road micro-traffic system, including a first modeling module for constructing a vehicle-road dynamic interaction model by utilizing the interaction effect between different types of vehicles and road traffic participants and obtaining the results of potential accidents caused by the interaction between the vehicle and the road, a second modeling module for constructing a human-vehicle dynamic interaction model by using the driving trajectory distribution output by the interaction of the human-vehicle system and obtaining the uncertainty of behaviors caused by the interaction between humans and vehicles, a third modeling module for constructing a human-road dynamic interaction model for the area of interest where the driver observes the road environment during driving by utilizing the visual characteristics of the driver and obtaining the differences in driver risk sensitivity in the human-road interaction process, a characterization module for characterizing the characteristic laws and differences of the driving habits of drivers using the concept of clustering to obtain the personalized characteristics of drivers, and an identification module for constructing a human-vehicle-road closed-loop dynamics system based on the results of potential accidents caused by the vehicle-road interaction, the uncertainty of behaviors, the differences in driver risk sensitivity in the human-road interaction process, and the personalized characteristics of the drivers and generating a risk identification result.

[0014] Optionally, in one embodiment of the present application, the human-vehicle-road closed-loop dynamics system is JPEG0007717409000049.jpg54140 represented as Among them, JPEG0007717409000050.jpg511 is a human-vehicle-road closed-loop dynamics system, JPEG0007717409000051.jpg611 is an effective response in the difference of driver risk sensitivity in the process of the human-road interaction, JPEG0007717409000052.jpg55 is the personalized characteristic of the driver, JPEG0007717409000053.jpg610 is the steering angle of the vehicle, JPEG0007717409000054.jpg828 is the uncertainty of the behavior, JPEG0007717409000055.jpg64 is the radius of the driving interest area in the difference of driver risk sensitivity in the process of the human-road interaction, JPEG0007717409000056.jpg75 is the result of potential accidents due to the vehicle-road interaction, JPEG0007717409000057.jpg34 is the steering, JPEG0007717409000058.jpg414 represents that the vehicle goes straight, JPEG0007717409000059.jpg514 represents that the vehicle turns left, JPEG0007717409000060.jpg414 represents that the vehicle turns right, JPEG0007717409000061.jpg57 represents the increase amount of the steering angle within one period Δt, JPEG0007717409000062.jpg46 is the host vehicle JPEG0007717409000063.jpg42 is a traffic participant JPEG0007717409000064.jpg52 is an interaction matrix for interacting with the traffic participant JPEG0007717409000065.jpg744 are the mass and speed of the host vehicle i and other traffic participants j respectively, JPEG0007717409000066.jpg410 is the collision avoidance time, JPEG0007717409000067.jpg628 is used as, JPEG0007717409000068.jpg34 is the length of the major axis of the ellipse, JPEG0007717409000069.jpg53 is the length of the minor axis of the ellipse, JPEG0007717409000070.jpg611 is the speed of the own vehicle i, JPEG0007717409000071.jpg713 is the driver's viewing angle function.

[0015] An embodiment of the third aspect of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor. To realize the modeling and risk identification method of the human-vehicle-road microscopic traffic system described in the above embodiment, the processor executes the computer program.

[0016] An embodiment of the fourth aspect of the present application provides a computer-readable storage medium storing a computer program. To realize the modeling and risk identification method of the human-vehicle-road microscopic traffic system described in the above embodiment, the computer program is executed by a processor.

Advantages of the Invention

[0017] The modeling and risk identification method and device of the human-vehicle-road microscopic traffic system in the embodiments of the present application have the following advantages. 1) In the present application, the essential attributes and interaction characteristics of humans, vehicles, roads, the environment, etc. are analyzed, an integrated mathematical model of the human-vehicle-road microscopic traffic system is constructed, and the state analysis can be performed by inputting it into the traffic system as a whole. 2) This application constructs a model to characterize the impact of the driver factor, vehicle motion state, and road environment information interaction process on the system safety state, discloses the risk generation mechanism of the microscopic traffic system, realizes system risk identification and grade warning, and can further guarantee the essential safety of the human-vehicle-road traffic system. 3) Compared with other risk indicators such as Time To Collision (TTC) and Time Headway (THW), in scenes with any road topology (intersections, circular roads, highways, etc.), this application can recognize the risk situation of the human-vehicle-road microscopic traffic system and support the grading countermeasures for vehicle driving risks. It is helpful to give timely warnings and corrective assistance to drivers approaching a risk state in a complex traffic environment, and provides new ideas for the research of collision avoidance warning countermeasures and control methods.

[0018] Additional aspects and advantages of this application are given in part in the following description, will become apparent in part from the following description, or will be understood through the practice of this application.

Brief Description of the Drawings

[0019] The above and / or additional aspects and advantages of this application will become apparent and be readily understood from the description of the following embodiments in connection with the accompanying drawings.

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Embodiments for Carrying Out the Invention

[0020] Hereinafter, embodiments of the present application will be described in detail. Examples of this embodiment are shown in the drawings, and the same or similar reference numerals from the beginning to the end indicate the same or similar elements or elements having the same or similar functions. The embodiments described with reference to the accompanying drawings below are exemplary and are for explaining the present application and should not be construed as limiting the present application.

[0021] Specifically, FIG. 1 is a flowchart of a method for modeling and risk identification of a human-vehicle-road microscopic traffic system according to an embodiment of the present application.

[0022] As shown in FIG. 1, the method for modeling and risk identification of the human-vehicle-road microscopic traffic system includes the following steps.

[0023] In step S101, a vehicle-road dynamic interaction model is constructed by using the interaction effects between different types of vehicles and road traffic participants, and the results of potential accidents caused by the interaction between the vehicle and the road are obtained.

[0024] Optionally, in one embodiment of the present application, the step of constructing a vehicle-road dynamic interaction model by utilizing the interaction effects between different types of vehicles and road traffic participants and obtaining the potential accident results due to the interaction between the vehicle and the road includes calculating the number of interactions between the host vehicle in the scene and different road traffic participants, calculating the results of the interaction between two objects in the traffic environment, and generating the potential accident results due to the vehicle-road interaction obtained by superimposing the cumulative interaction results in the process of multiple traffic participants interacting with the host vehicle based on the number of interactions and the results of the interactions.

[0025] Specifically, in the operation process of the human-vehicle-road micro traffic system, there are multiple types of interaction behaviors between the host vehicle and road traffic participants. In order to characterize the possible results generated during the interaction process, modeling is performed on the interaction effects between different types of vehicles and road traffic participants, and the potential accident results due to the vehicle-road interaction Obtain JPEG0007717409000072.jpg75.

[0026] First, calculate the number of interactions between the host vehicle in the scene and different road traffic participants. In any one scene, the scene includes different road traffic participants JPEG0007717409000073.jpg642 is included, and the host vehicle JPEG0007717409000074.jpg42 and the participant JPEG0007717409000075.jpg52 interaction can be characterized by the interaction matrix JPEG0007717409000076.jpg46, and Set it as JPEG0007717409000077.jpg1954. JPEG0007717409000078.jpg621 is the host vehicle and other road traffic participants Represents the interaction with JPEG0007717409000079.jpg52, JPEG0007717409000080.jpg621 represents no interaction.

[0027] Next, calculate the result of the interaction between two objects in the traffic environment: If preventive management is not carried out in a timely manner considering the interaction process, a collision accident will occur, and energy transfer between entities will occur. That is, when a collision occurs, the resulting outcome can be regarded as the energy transfer between the colliding entities. That is, JPEG0007717409000081.jpg1460 is set as. Among them, JPEG0007717409000082.jpg55 represents the transmitted energy of the host vehicle i, JPEG0007717409000083.jpg519 and JPEG0007717409000084.jpg515 are the mass and speed (vector) of the host vehicle i and another entity j, respectively.

[0028] Finally, in the process of multiple traffic participants interacting with the host vehicle, the cumulative interaction results are superimposed to obtain the result of potential accidents due to vehicle-road interaction: JPEG0007717409000085.jpg1165 is obtained.

[0029] In step S102, a human-vehicle dynamic interaction action model is constructed using the driving trajectory distribution output by the interaction of the human-vehicle system, and the uncertainty of the behavior caused by the interaction between humans and vehicles is obtained.

[0030] Optionally, in one embodiment of the present application, the step of constructing a human-vehicle dynamic interaction action model using the driving trajectory distribution output by the interaction of the human-vehicle system and obtaining the uncertainty of the behavior caused by the interaction between the human and the vehicle includes calculating an equivalent linear two-wheeled vehicle model based on the vehicle kinematic model and the turning radius, calculating a predicted position based on the commanded steering angle, predicting the uncertain motion of the driver-vehicle system based on the collected actual driving experiment data, obtaining a Gaussian normal distribution of the turning angle of the steering angle of the vehicle based on the predicted position and the uncertain motion, and performing parameter determination based on the Gaussian normal distribution to determine the uncertainty of the behavior caused by the interaction between the human and the vehicle.

[0031] Specifically, in the process of interaction between the human and the vehicle, the vehicle corresponds to a feedback actuator mechanism, which can receive the uncertainty of the driver's intention and the incentive of the dynamically adjusted driving goal, and output the dynamic response of the vehicle driving process, thereby explaining the interaction output of the human-vehicle system with the driving trajectory distribution, and outputting the uncertainty of the behavior caused by the interaction between the human and the vehicle. JPEG0007717409000086.jpg823

[0032] First, assume that a normal driver is driving while observing traffic rules and laws. Therefore, during normal vehicle driving, mainly straight-ahead, left-turning, and right-turning steering actions are performed, accompanied by dynamic speed adjustment. According to the vehicle kinematic model, the turning radius R can be calculated as follows using the equivalent linear two-wheeled vehicle model: JPEG0007717409000087.jpg1458 Among them, K is the stability coefficient, L is the wheelbase of vehicle i, JPEG0007717409000088.jpg613 is the steering angle of the vehicle.

[0033] When vehicle i can ignore the constant-speed driving of the sideslip angle, the commanded steering angle At 54, the predicted position At 618 in JPEG0007717409000090.jpg, In JPEG0007717409000091.jpg at 3087 can be calculated, Next, based on the range of the steering angle, there are certain boundaries for the possible motion trajectories of vehicle i, and the motion state of vehicle i is absolutely stable within this boundary. When vehicle i is going straight on the road, the driver may go straight and move to the left lane or move to the right lane. After the driver operates the vehicle, in order to determine the position range output by the vehicle, based on the actual driving experiment data, the uncertain motion of the driver-vehicle system can be predicted.

[0034] Finally, the steering angles of the vehicle are statistically analyzed, and the obtained turning angle distribution basically exhibits a Gaussian normal distribution. Based on the experimental data, parameter determination is performed for the Gaussian distribution: At 1491 in JPEG0007717409000092.jpg Among them, At 610 in JPEG0007717409000093.jpg is the steering angle of the vehicle, At 34 in JPEG0007717409000094.jpg is the dispersion of the data distribution following the normal distribution, At 44 in JPEG0007717409000095.jpg is the mean value of the random variable following the normal distribution.

[0035] Please refer to Figure 2. As shown in the Gaussian distribution diagram of the steering angle in the highway scene according to the embodiment of the present application, when inputting the actual data collection results of some ring roads in Beijing, the specific parameters for fitting to obtain the Gaussian distribution: At 536 in JPEG0007717409000096.jpg is obtained, whereby the turning angle distribution is At 14153 in JPEG0007717409000097.jpg is output as In different environments, due to the differences in the characteristics of vehicle operation by different drivers, there are slight differences in the specific parameters of the Gaussian distribution, but the overall trend is consistent. That is, the steering response width in the process of a driver operating a vehicle follows a Gaussian distribution.

[0036] In step S103, using the visual characteristics of the driver, construct a human-road dynamic interaction action model for the area of interest where the driver observes the road environment during driving, and obtain the differences in the driver's risk sensitivity in the human-road interaction process.

[0037] Optionally, in one embodiment of the present application, the step of using the visual characteristics of the driver to construct a human-road dynamic interaction action model for the area of interest where the driver observes the road environment during driving and obtaining the differences in the driver's risk sensitivity in the human-road interaction process includes: calculating the effective response of the driver during normal driving using the driver's visual range; obtaining the length of the equipotential line of the driver's dynamic visual range using the elliptical distribution characteristics of the driver's vision; perceiving the relative speed of the vehicle using the influence of the change in the driver's visual angle on the driver, obtaining the influence of the driver's dynamic perception of the visual field, generating the radius of the size of the area of interest for observing the road environment during driving, and determining the differences in the driver's risk sensitivity in the human-road interaction process based on the effective response and the radius of the size of the area of interest.

[0038] Specifically, since the driver mainly relies on vision to obtain information during driving, the interaction between the driver and the road environment is mainly affected by the driver's visual effect. Based on the visual characteristics of the driver, model the area of interest where the driver observes the road environment during driving, and output the differences in the driver's risk sensitivity in the human-road interaction process (including the effective response JPEG0007717409000098.jpg611, the radius of the driving area of interest JPEG0007717409000099.jpg64).

[0039] First, calculate the effective response of the driver during normal driving. The driver's response to the surrounding environment varies depending on their viewing angle and scanning frequency, and the perception of the relative distance and relative speed between other vehicles and themselves in the road environment is the sharpest. The front of the driver is the sharp visual area of the driver, and the driver must always pay attention and react in a timely manner. On both sides of the sharp visual area, the front area of the adjacent lane is the general visual area of the driver, but during driving, the driver always pays attention because there is a possibility that a vehicle in the adjacent lane may cut in, and at the same time, they may need to move into the adjacent lane themselves. The areas on both sides, that is, the peripheral visual areas, are not paid much attention, and the rear of the vehicle is often ignored due to factors such as visual characteristics and liability determination. The driver's response to the surrounding environment is JPEG0007717409000100.jpg44, and assuming it is a function of its viewing angle JPEG0007717409000101.jpg77, JPEG0007717409000102.jpg730 is the driver's viewing angle. For example, if the driver's effective response JPEG0007717409000103.jpg647, when a situation occurred immediately before, the driver made a 100% response and ignored the following vehicle immediately after.

[0040] Next, please refer to Figure 3. As shown in the divided diagram of the driving operation attention area according to the embodiment of the present application, based on the elliptical distribution characteristics of the driver's vision, assuming that the driver is at one of the foci JPEG0007717409000104.jpg55 of the ellipse, the driver's dynamic vision is inversely proportional to the vehicle speed. The faster the vehicle speed, the narrower the vision, and the visual area directly behind is ignored. JPEG0007717409000105.jpg930 The length of the equipotential line of the dynamic vision range: JPEG0007717409000106.jpg1485 can be obtained, In the formula, JPEG0007717409000107.jpg713 is the driver's viewing angle function, From the point on the equipotential line of JPEG0007717409000108.jpg52 The distance to vehicle JPEG0007717409000109.jpg42, and The length of the major axis of the ellipse is JPEG0007717409000110.jpg34, and The length of the minor axis of the ellipse is JPEG0007717409000111.jpg53, and its value is The speed of vehicle JPEG0007717409000112.jpg42 Regarding JPEG0007717409000113.jpg611. That is, the higher the speed, the narrower the line of sight range.

[0041] Next, based on the visual characteristics of the driver, the influence of the change in the viewing angle on the driver is that the relative speed of the vehicle can be perceived, and thus the driver can perceive the field of view dynamically, that is It affects JPEG0007717409000114.jpg65. The viewing angle change rate JPEG0007717409000115.jpg711 and JPEG0007717409000116.jpg65 have the following relationship: JPEG0007717409000117.jpg1682 Among them, JPEG0007717409000118.jpg47 is the relative speed of the vehicle, and JPEG0007717409000119.jpg66 is the distance between vehicles, and JPEG0007717409000120.jpg410 is the collision avoidance time (s), and JPEG0007717409000121.jpg610 vehicle JPEG0007717409000122.jpg42 is the driving speed.

[0042] Changing the above formula, Define JPEG0007717409000123.jpg628, and the equivalent radius is JPEG0007717409000124.jpg19147 can be required to be, Among them, JPEG0007717409000125.jpg410 is the collision avoidance time, set JPEG0007717409000126.jpg628 as, JPEG0007717409000127.jpg611 is the speed of the host vehicle i, that is, the radius of the size of the region of interest where the driver observes the environment around the road during driving can be defined as JPEG0007717409000128.jpg64.

[0043] In step S104, the idea of clustering is used to characterize the characteristic laws and differences of the driving habits of the driver, so as to obtain the personalized characteristics of the driver.

[0044] Optionally, in one embodiment of the present application, the step of using the idea of clustering to characterize the characteristic laws and differences of the driving habits of the driver to obtain the personalized characteristics of the driver includes distinguishing the similarity of the operation behaviors and trajectories of various drivers in the same scene, obtaining the personalized expression of the driver based on the non-monitoring clustering method, obtaining the characteristic expression of each driver based on the personalized expression of the driver, obtaining the driving characteristics based on the distance from the clustering center, and determining the personalized characteristics of the driver.

[0045] Specifically, due to the individual differences of the driver and the complexity of the situation, the active behaviors of the driver in the actual traffic scene exhibit characteristics such as high randomness and non-linearity. Therefore, the idea of clustering is used to characterize the characteristic laws and differences of the driving habits of the driver, and the personalized characteristics JPEG0007717409000129.jpg55 is output.

[0046] First, driving habits are greatly affected by different ages, genders, personalities, driving experiences, and driving proficiencies. The personalized expression of drivers is closely related to the stability of the system. Distinguish by the similarity of the operation behaviors and trajectories of various drivers in the same scene, and based on the non-monitoring clustering method, three clustering centers ( JPEG0007717409000130.jpg612, JPEG0007717409000131.jpg612, JPEG0007717409000132.jpg612) are output, and drivers are classified into three types, defined as aggressive type A, normal type B, and conservative type C.

[0047] Next, perform personalized driver characteristic expressions, represent them as the driver behaviors (e.g., operation trajectories), and perform clustering JPEG0007717409000133.jpg611. Calculate the distances from each of the three clustering centers ( JPEG0007717409000134.jpg833, JPEG0007717409000135.jpg833, JPEG0007717409000136.jpg833), and calculate the distance ratios, which are JPEG0007717409000137.jpg16125 JPEG0007717409000138.jpg16125 JPEG0007717409000139.jpg16125 and obtain that the driving characteristics of the driver are ( JPEG0007717409000140.jpg652).

[0048] In step S105, based on the potential accidents caused by vehicle-road interactions, the uncertainty of behaviors, the differences in driver risk sensitivities during the human-road interaction process, and the personalized characteristics of drivers, construct a human-vehicle-road closed-loop dynamics system and generate risk identification results.

[0049] Optionally, in one embodiment of the present application, the human-vehicle-road closed-loop dynamics system JPEG0007717409000141.jpg1584 is represented as Among them, JPEG0007717409000142.jpg511 is the human-vehicle-road closed-loop dynamics system, JPEG0007717409000143.jpg611 is an effective response in the difference of driver risk sensitivity in the human-road interaction process, JPEG0007717409000144.jpg55 are the personalized characteristics of the driver, JPEG0007717409000145.jpg610 is the steering angle of the vehicle, JPEG0007717409000146.jpg828 is the uncertainty of behavior, JPEG0007717409000147.jpg64 is the radius of the driver's interest area in the difference of driver risk sensitivity in the human-road interaction process, JPEG0007717409000148.jpg75 is the result of potential accidents due to vehicle-road interaction.

[0050] Specifically, by modeling the interaction effects of vehicle-road, human-vehicle, and human-road respectively, that is, the result of potential accidents due to vehicle-road interaction JPEG0007717409000149.jpg75, the uncertainty of behavior caused by the interaction between humans and vehicles JPEG0007717409000150.jpg823 and the difference in driver risk sensitivity in the human-road interaction process (effective response JPEG0007717409000151.jpg611, the radius of the driver's interest area JPEG0007717409000152.jpg64, the personalized characteristics of the driver By including JPEG0007717409000153.jpg55, a unified vision angle is ultimately used to construct a human-vehicle-road closed-loop dynamics system. System integration model JPEG0007717409000154.jpg511 is represented as follows: JPEG0007717409000155.jpg54140 Among them, JPEG0007717409000156.jpg511 is a human-vehicle-road closed-loop dynamics system, JPEG0007717409000157.jpg611 is an effective response in the difference of driver risk sensitivity in the human-road interaction process, JPEG0007717409000158.jpg55 are the personalized characteristics of the driver, JPEG0007717409000159.jpg610 is the steering angle of the vehicle, JPEG0007717409000160.jpg828 is the uncertainty of behavior, JPEG0007717409000161.jpg64 is the radius of the driver's area of interest in the difference of driver risk sensitivity in the human-road interaction process, JPEG0007717409000162.jpg75 is the result of potential accidents due to vehicle-road interaction, JPEG0007717409000163.jpg34 is steering, JPEG0007717409000164.jpg414 represents that the vehicle is going straight, JPEG0007717409000165.jpg514 represents that the vehicle turns left, JPEG0007717409000166.jpg514 represents that the vehicle turns right, JPEG0007717409000167.jpg57 is the increase in the steering angle within a period Δt, JPEG0007717409000168.jpg46 is the host vehicle JPEG0007717409000169.jpg42 and road traffic participants JPEG0007717409000170.jpg52 is an interaction matrix in which they interact, JPEG0007717409000171.jpg519 and JPEG0007717409000172.jpg515 are respectively the mass and speed of the own vehicle i and another road traffic participant j, JPEG0007717409000173.jpg410 is the collision avoidance time, JPEG0007717409000174.jpg628 is taken as, JPEG0007717409000175.jpg34 is the length of the major axis of the ellipse, JPEG0007717409000176.jpg53 is the length of the minor axis of the ellipse, JPEG0007717409000177.jpg611 is the speed of the own vehicle i, JPEG0007717409000178.jpg713 is the driver's viewing angle function.

[0051] Please refer to FIG. 4. As shown in the schematic framework diagram of the coupling relationship of the human-vehicle-environment elements according to the embodiments of the present application, in a complex situation, the driving environment of a motor vehicle is complex, the driving states are diverse, the driving styles and driving experiences of drivers are different, and the driver, vehicle, and road environment are combined with each other to form a complex and broad sense dynamic system, and the safety of the system is affected by the interaction of the driver, vehicle, and road. The human-vehicle-road closed-loop system has strong non-linearity, coupling, and time-varying characteristics.

[0052] Overall, the vehicle corresponds to a control system, and the movement of the vehicle is like the operation of a controller. When driving the vehicle, the driver observes the surrounding environment, such as road conditions and objects visible within the field of vision, and sends operation signals corresponding to the vehicle, that is, the operation signals applied to the steering wheel, brakes, and accelerator pedal. Subsequently, the driver receives the state feedback signal from the vehicle through sensory organs such as the eyes and ears, and adjusts the control based on it. At the same time, the environment also affects the driver and the vehicle. If the road conditions are bad, it is possible that the vehicle cannot respond accurately to the control operation, or the driving behavior of the driver may change. Therefore, the environment is part of the noise and interference that affects the vehicle control system.

[0053] By adding a controller to the vehicle, it is possible to adjust situations where the vehicle fails or is difficult to control. Through environmental input, the controller grasps the driving environment and road conditions, and through driver input, the controller grasps the driving state of the driver. Usually, the controller is also required to supply signals to the vehicle to adjust the driver's behavior according to the driving situation of the vehicle itself, such as the steering angle, brakes, and traction torque. Therefore, the entire human-vehicle-road system can be represented as shown in Figure 4.

[0054] As shown in Figure 5, the driving risk is the buckling in a broad sense of the human-vehicle-road closed-loop system, and is usually determined by the driver, road conditions, and the dynamic characteristics of the vehicle itself. Taking the scene of multi-lane lane change and obstacle avoidance as an example, at the initial point, the driver drives the vehicle and drives normally on a flat road with good road conditions, and the system state at this stage is JPEG0007717409000179.jpg614. When an obstacle occurs in front of the vehicle driving lane, after the driver notices the front obstacle, the driver releases the accelerator pedal or steps on the brake pedal to decelerate the vehicle and maintain a relative safe distance from the front obstacle vehicle. The change in the human-vehicle-road system state is It is JPEG0007717409000180.jpg614. Then, while the driver continues to decelerate, the vehicle speed continuously decreases, the distance between the host vehicle and the obstacle vehicle shortens, and the state of the human-vehicle-road system changes from state to JPEG0007717409000181.jpg614 When the state of the human-vehicle-road system changes to state JPEG0007717409000182.jpg614 and the driver suddenly makes a mistake, the state of the human-vehicle-road system changes from state JPEG0007717409000183.jpg614 to state JPEG0007717409000184.jpg614, and an accident occurs.

[0055] The method for modeling and risk identification of the human-vehicle-road micro traffic system according to the embodiments of the present application characterizes the influence of driver elements, vehicle motion states, and road environment information interaction processes on the system safety state, discloses the risk generation mechanism of the micro traffic system, and realizes system risk identification and grade warning. The human-vehicle-road integrated model analyzes the essential attributes and interaction characteristics of humans, vehicles, roads, environments, etc. In scenes with any road topology (intersections, circular roads, highways, etc.), it can recognize the risk situation of the human-vehicle-road micro traffic system and assist in formulating grade countermeasures for vehicle driving risks. The method helps to provide timely warnings and correction assistance to drivers approaching a risk state in a complex traffic environment, and provides new ideas for the research of collision avoidance warning countermeasures and control methods.

[0056] Next, with reference to the drawings, the human-vehicle-road micro traffic system modeling and risk identification device according to the embodiments of the present application will be described.

[0057] FIG. 6 is an exemplary diagram of the human-vehicle-road micro traffic system modeling and risk identification device according to the embodiments of the present application.

[0058] As shown in FIG. 6, the person-vehicle-road microscopic traffic system modeling and risk identification device 10 includes a first modeling module 100, a second modeling module 200, a third modeling module 300, a characterization module 400, and an identification module 500.

[0059] Among them, the first modeling module 100 is used to construct a vehicle-road dynamic interaction model by utilizing the interaction between different types of vehicles and road traffic participants, and to obtain the results of potential accidents caused by the interaction between vehicles and roads. The second modeling module 200 is used to construct a person-vehicle dynamic interaction model by using the driving trajectory distribution output by the interaction of the person-vehicle system, and to obtain the uncertainty of behaviors caused by the interaction between people and vehicles. The third modeling module 300 is used to construct a person-road dynamic interaction model for the area of interest where the driver observes the road environment during driving by utilizing the visual characteristics of the driver, and to obtain the differences in driver risk sensitivity during the person-road interaction process. The characterization module 400 is used to characterize the characteristic laws and differences of the driving habits of drivers by using the concept of clustering, so as to obtain the personalized characteristics of drivers. The identification module 500 is used to construct a person-vehicle-road closed-loop dynamics system based on the results of potential accidents caused by vehicle-road interaction, the uncertainty of behaviors, the differences in driver risk sensitivity during the person-road interaction process, and the personalized characteristics of drivers, and to generate risk identification results.

[0060] Optionally, in one embodiment of the present application, the person-vehicle-road closed-loop dynamics system is JPEG0007717409000185.jpg54140 represented as Among them, JPEG0007717409000186.jpg511 is the person-vehicle-road closed-loop dynamics system, JPEG0007717409000187.jpg611 is an effective response in the difference of drivers' risk sensitivity during the human-road interaction process. JPEG0007717409000188.jpg55 are the personalized characteristics of the driver. JPEG0007717409000189.jpg610 is the steering angle of the vehicle. JPEG0007717409000190.jpg828 is the uncertainty of behavior. JPEG0007717409000191.jpg64 is the radius of the driving interest area in the difference of drivers' risk sensitivity during the human-road interaction process. JPEG0007717409000192.jpg75 is the result of potential accidents due to vehicle-road interaction. JPEG0007717409000193.jpg34 is the steering. JPEG0007717409000194.jpg414 represents that the vehicle is going straight. JPEG0007717409000195.jpg514 represents that the vehicle turns left. JPEG0007717409000196.jpg514 represents that the vehicle turns right. JPEG0007717409000197.jpg57 represents the increase in the steering angle within one period Δt. JPEG0007717409000198.jpg46 is the host vehicle JPEG0007717409000199.jpg42 is the road traffic participant JPEG0007717409000200.jpg52 is the interaction matrix for interacting with JPEG0007717409000201.jpg519 and JPEG0007717409000202.jpg515 are the mass and speed of the host vehicle i and the other road traffic participant j respectively. JPEG0007717409000203.jpg410 is the collision avoidance time. Set JPEG0007717409000204.jpg to 628, JPEG0007717409000205.jpg34 is the length of the major axis of the ellipse, JPEG0007717409000206.jpg53 is the length of the minor axis of the ellipse, JPEG0007717409000207.jpg611 is the speed of vehicle i, JPEG0007717409000208.jpg713 is the driver's viewing angle function.

[0061] Optionally, in one embodiment of the present application, the first modeling module specifically calculates the number of interactions between the host vehicle in the scene and different road traffic participants, calculates the result of the interaction between two objects in the traffic environment, and based on the number of interactions and the result of the interaction, in the process of multiple traffic participants interacting with the host vehicle, the potential accident result due to the lane interaction obtained by superimposing the interaction cumulative results JPEG0007717409000209.jpg75: JPEG0007717409000210.jpg1165 is used to generate.

[0062] Among them, JPEG0007717409000211.jpg46 is the interaction matrix of vehicle i interacting with road traffic participant j, JPEG0007717409000212.jpg519 and JPEG0007717409000213.jpg515 are the mass and speed of vehicle i and other road traffic participant j respectively.

[0063] Optionally, in one embodiment of the present application, the second modeling module specifically calculates an equivalent linear two-wheeler model based on a vehicle kinematic model and a turning radius, calculates a predicted position based on a commanded steering angle, predicts the uncertain motion of the driver-vehicle system based on the collected actual driving experiment data, and based on the predicted position and the uncertain motion, obtains a Gaussian normal distribution of the turning angle of the steering angle of the vehicle, and performs parameter determination based on the Gaussian normal distribution to determine the uncertainty of the behavior caused by the interaction between the person and the vehicle JPEG0007717409000214.jpg823: JPEG0007717409000215.jpg1491 is used to Among them, JPEG0007717409000216.jpg610 is the steering angle of the vehicle, JPEG0007717409000217.jpg34 is the dispersion of the data distribution following the normal distribution, JPEG0007717409000218.jpg44 is the mean value of the random variable following the normal distribution.

[0064] Optionally, in one embodiment of the present application, the third modeling module specifically calculates an effective response during normal driving of the driver by using the driver's visual range, uses the elliptical distribution characteristic of the driver's vision to obtain the length of the equipotential line of the driver's dynamic visual range JPEG0007717409000219.jpg65: JPEG0007717409000220.jpg1485 and obtains Among them, JPEG0007717409000221.jpg64 is the radius of the size of the region of interest, JPEG0007717409000222.jpg34 is the length of the major axis of the ellipse, The length of the minor axis of the ellipse is 53 in JPEG0007717409000223.jpg, and in JPEG0007717409000224.jpg is 34 and the value of 53 in JPEG0007717409000225.jpg is related to the vehicle speed of 42 in JPEG0007717409000226.jpg and is related to 611 in JPEG0007717409000227.jpg, where 713 in JPEG0007717409000228.jpg is the driver's viewing angle function, utilize the impact of the change in the driver's viewing angle on the driver to sense the relative speed of the vehicle, obtain the impact of the driver's dynamically perceived field of vision, and the radius of the size of the region of interest for observing the environment around the road during driving: is 19147 in JPEG0007717409000229.jpg and generate Among them, 410 in JPEG0007717409000230.jpg is the collision avoidance time, is set as 628 in JPEG0007717409000231.jpg, 611 in JPEG0007717409000232.jpg is the speed of the own vehicle i, and it is used to determine the difference in the driver's risk sensitivity during the human-road interaction process based on the effective response and the radius of the size of the region of interest.

[0065] Optionally, in one embodiment of the present application, the characterization module is specifically used to distinguish the similarity of the operation behaviors and trajectories of various types of drivers in the same scene, obtain the personalized expression of the driver based on the non-monitoring clustering method, obtain the characteristic expression of each driver based on the personalized expression of the driver, obtain the driving characteristics based on the distance from the clustering center, and determine the personalized characteristics of the driver.

[0066] In addition, the interpretations and explanations for the above embodiments of the modeling and risk identification of the human-vehicle-road microscopic traffic system are also applicable to the modeling and risk identification device of the human-vehicle-road microscopic traffic system in the embodiments, and will not be repeated here.

[0067] The modeling and risk identification device of the human-vehicle-road microscopic traffic system according to the embodiments of the present application is used to characterize the influence of driver elements, vehicle motion states, and road environment information interaction processes on the system safety state. It discloses the risk generation mechanism of the microscopic traffic system and realizes system risk identification and grade warning. The human-vehicle-road integrated model analyzes the essential attributes and interaction characteristics of people, vehicles, roads, the environment, etc. In scenes with any road topology (such as intersections, circular roads, highways, etc.), it can recognize the risk situation of the human-vehicle-road microscopic traffic system and assist in formulating countermeasures for the grade of vehicle driving risks. It helps to provide timely warnings and corrective assistance to drivers approaching a risk state in a complex traffic environment, and provides new ideas for the research of collision avoidance warning countermeasures and control methods.

[0068] Figure 7 is a structural schematic diagram of an electronic device according to an embodiment of the present application. The electronic device may include a memory 701, a processor 702, and a computer program stored in the memory 701 and running on the processor 702. When the processor 702 executes the program, it realizes the modeling and risk identification method of the human-vehicle-road microscopic traffic system according to the above embodiments.

[0069] Furthermore, the electronic device further includes a communication interface 703 used for communication between the memory 701 and the processor 702.

[0070] The memory 701 is used to store the computer program running on the processor 702.

[0071] The memory 701 may include a high-speed RAM memory, or may include a non-volatile memory such as at least one disk memory for example.

[0072] When the memory 701, the processor 702, and the communication interface 703 are independently realized, the communication interface 703, the memory 701, and the processor 702 can be connected to each other via a bus to complete mutual communication. The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus can be divided into an address bus, a data bus, a control bus, and the like. For the sake of illustration, only thick lines are shown in FIG. 7, but it does not mean that there is only one or one type of bus.

[0073] Optionally, specifically in implementation, when the memory 701, the processor 702, and the communication interface 703 are integrated on one chip, the memory 701, the processor 702, and the communication interface 703 can complete mutual communication via an internal interface.

[0074] The processor 702 may be a Central Processing Unit (CPU), or a specific Application Specific Integrated Circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0075] This embodiment further provides a computer-readable storage medium in which a computer program is stored, and when the program is executed by a processor, the above-described method for modeling and risk identification of a human-vehicle-road microscopic traffic system is realized.

[0076] In the description of this specification, descriptions such as reference terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" mean that specific features, structures, materials, or characteristics described by combining the embodiments or examples are included in at least one embodiment or example of the present invention. In this specification, the above exemplary descriptions do not necessarily target the same embodiment or example. Also, the specific features, structures, materials, or characteristics to be described can be combined in an appropriate manner in any one or more embodiments or examples. Note that when there is no conflict with each other, those skilled in the art can combine and combine different embodiments or examples and the features of different embodiments or examples described in this specification.

[0077] Note that terms such as "first" and "second" are only for the purpose of explanation and cannot be regarded as indicating relative importance implicitly or explicitly, or implicitly indicating the number of technical features. Therefore, features limited by "first" and "second" can explicitly or implicitly include one or more of the features. In the description of this application, unless specifically and clearly limited, the concept of "N" is at least two, for example, two or three.

[0078] Any process or method description in the flowchart or described here in other ways can be understood as indicating one or more modules, fragments, or parts including executable instruction codes for realizing custom logic functions or process steps. And the scope of the preferred embodiment of this application includes other realizations, including basically performing functions in the same or reverse order based on related functions, and it is not necessary to perform functions in the shown or discussed order, which should be understood by those skilled in the art.

[0079] It should be understood that each part of the present application can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented by software or firmware stored in a memory and executed by an appropriate instruction execution system. When implemented by hardware, as in other embodiments, it can be implemented by any of the techniques known in the art, such as a discrete logic circuit having a logic gate circuit that realizes a logical function for a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a programmable gate array (PGA), a field programmable gate array (FPGA), etc., or a combination thereof.

[0080] Those skilled in the art can understand that the implementation of all or part of the steps described in the above method embodiments can be completed by instructing the relevant hardware by a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can be understood that it includes one or a combination of the steps of the method embodiments.

Description of Reference Signs

[0081] 100 First Modeling Module 200 Second Modeling Module 300 Third Modeling Module 400 Characterization Module 500 Identification Module 701 Memory 702 Processor 703 Communication Interface

Claims

A method for modeling and risk identification of a human-vehicle-road microscopic traffic system executed by an electronic device, comprising: Constructing a vehicle-road dynamic interaction model by utilizing different types of interaction effects between vehicles and road traffic participants, and obtaining the results of potential accidents caused by the interaction between the vehicle and the road; Constructing a human-vehicle dynamic interaction model by using the vehicle driving trajectory distribution output by the interaction of the human-vehicle system, and obtaining the uncertainty of behaviors caused by the interaction between humans and vehicles; Constructing a human-road dynamic interaction model for the area of interest where a driver observes the road environment during driving by utilizing the visual characteristics of the driver, and obtaining the differences in driver risk sensitivity during the human-road interaction process; Characterizing the characteristic laws and differences in driving habits of drivers by using the concept of clustering to obtain the personalized characteristics of drivers; Constructing a human-vehicle-road closed-loop dynamics system based on the results of potential accidents caused by the interaction between the vehicle and the road, the uncertainty of the behaviors, the differences in driver risk sensitivity during the human-road interaction process, and the personalized characteristics of the driver, and generating a risk identification result. A method for modeling and risk identification of a human-vehicle-road microscopic traffic system, characterized by comprising the above steps. **Claim 2** The human-vehicle-road closed-loop dynamics system is: Expressed as: Among them: Is the human-vehicle-road closed-loop dynamics system; Is the effective response in the differences in driver risk sensitivity during the human-road interaction process; Is the personalized characteristics of the driver; Is the steering angle of the vehicle; Is the uncertainty of the behaviors; Is the radius of the area of interest in the differences in driver risk sensitivity during the human-road interaction process; Is the result of potential accidents caused by the interaction between the vehicle and the road; Is the steering; Indicates that the vehicle is going straight; Indicates that the vehicle turns left; Indicates that the vehicle turns right; Is the increase in the steering angle within a certain period Δt; Is the host vehicle Interacts with road traffic participants Is the interaction matrix; And Are the mass and speed of the host vehicle i and other road traffic participants j respectively; Is the collision avoidance time; Set as: Is the length of the major axis of the ellipse; Is the length of the minor axis of the ellipse; Is the speed of the host vehicle i; The method according to claim 1, characterized in that it is a driver's visual angle function.

3. The step of constructing a vehicle-road dynamic interaction model by using different types of interaction effects between the vehicle and road traffic participants and obtaining the potential accident results due to the interaction between the vehicle and the road is as follows: Calculating the number of interactions between the host vehicle and different road traffic participants within the scene; Calculating the result of the interaction between two objects in the traffic environment; Based on the number of the interactions and the result of the interaction, during the process of multiple traffic participants interacting with the host vehicle, Generating the potential accident results due to the interaction between the vehicle and the road obtained by superimposing the cumulative interaction results, : including the step of: Among them, is the interaction matrix in which the host vehicle i interacts with the road traffic participant j, and are the mass and speed of the host vehicle i and the other road traffic participant j respectively. The method according to claim 1 is characterized in that.

4. The step of constructing a human-vehicle dynamic interaction model by using the driving trajectory distribution of the vehicle output by the interaction of the human-vehicle system and obtaining the uncertainty of the behavior caused by the interaction between the human and the vehicle is as follows: Calculating an equivalent linear two-wheeled vehicle model based on the vehicle kinematic model and the turning radius, and calculating the predicted position based on the commanded steering angle; Predicting the uncertain motion of the driver-vehicle system based on the collected actual driving experiment data; Based on the predicted position and the uncertain motion, obtaining the Gaussian normal distribution of the turning angle of the steering angle of the vehicle, and performing parameter determination based on the Gaussian normal distribution, and the uncertainty of the behavior caused by the interaction between the human and the vehicle : including the step of determining; Among them, is the steering angle of the vehicle, is the dispersion degree of the data distribution following the normal distribution, is the mean value of the random variable following the normal distribution. The method according to claim 1 is characterized in that.

5. The step of constructing a human-road dynamic interaction model for the area of interest where the driver observes the road surrounding environment during driving by using the visual characteristics of the driver and obtaining the difference in the driver's risk sensitivity during the human-road interaction process is as follows: Calculating the effective response of the driver during normal driving by using the visual range of the driver; A step of obtaining the length of the equipotential line of the driver's dynamic visual range by using the elliptical distribution characteristics of the driver's vision : comprising: wherein is the radius of the size of the region of interest, is the length of the major axis of the ellipse, is the length of the minor axis of the ellipse, and are related to the vehicle speed of the host vehicle and is a step that is the driver's visual angle function, Using the influence of the change in the driver's visual angle on the driver to sense the relative speed of the vehicle, obtaining the influence of the driver's dynamic perception of the visual field, and the radius of the size of the region of interest for observing the environment around the road during driving: generating wherein is the collision avoidance time, set as and is the speed of the host vehicle i, Determining the difference in the driver risk sensitivity in the human-road interaction process based on the effective response and the radius of the size of the region of interest, the method according to claim 1, characterized by comprising this step.

6. The step of characterizing the characteristic laws and differences of the driving habits of drivers using the concept of clustering to obtain the personalized characteristics of drivers distinguishing the similarity of the operation behaviors and trajectories of various types of drivers in the same scene, and obtaining the personalized expressions of drivers based on the non-monitoring clustering method, obtaining the characteristic expressions of each driver based on the personalized expressions of the drivers, obtaining the driving characteristics based on the distance from the clustering center, and determining the personalized characteristics of the drivers, the method according to claim 1, characterized by comprising this step.

7. A human-vehicle-road microscopic traffic system modeling and risk identification device, A first modeling module for constructing a vehicle-road dynamic interaction action model by using different types of interaction actions between the vehicle and road traffic participants, and obtaining the results of potential accidents caused by the interaction between the vehicle and the road A second modeling module for constructing a human-vehicle dynamic interaction action model by using the driving trajectory distribution of the vehicle output by the interaction of the human-vehicle system, and obtaining the uncertainty of the behaviors caused by the interaction between the human and the vehicle A third modeling module for constructing a human-road dynamic interaction action model for the region of interest where the driver observes the environment around the road during driving by using the visual characteristics of the driver, and obtaining the differences in the driver risk sensitivity in the human-road interaction process Characterize the characteristic laws and differences of drivers' driving habits using the concept of clustering to obtain a characterization module for drivers' personalized characteristics, and Based on the potential accident results of the interaction between the vehicle and the road, the uncertainty of the behavior, the differences in drivers' risk sensitivity during the human-road interaction process, and the personalized characteristics of the driver, construct a human-vehicle-road closed-loop dynamics system to generate risk identification results. An identification module, characterized in that it includes a modeling and risk identification device for a human-vehicle-road micro-traffic system.

8. The human-vehicle-road closed-loop dynamics system is represented as Among them, is the human-vehicle-road closed-loop dynamics system, is an effective response in the differences in drivers' risk sensitivity during the human-road interaction process, is the personalized characteristics of the driver, is the steering angle of the vehicle, is the uncertainty of the behavior, is the radius of the driver's area of interest in the differences in drivers' risk sensitivity during the human-road interaction process, is the result of potential accidents due to the interaction between the vehicle and the road, is the steering, represents that the vehicle is going straight, represents that the vehicle turns left, represents that the vehicle turns right, is the increase in the steering angle within a certain period Δt, is the host vehicle interacts with a traffic participant and are the mass and speed of the host vehicle i and another traffic participant j respectively, is the collision avoidance time, is defined as is the length of the major axis of the ellipse, is the length of the minor axis of the ellipse, is the speed of the host vehicle i, is the driver's visual angle function, characterized in that the device according to claim 7.

9. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program so as to implement the modeling and risk identification method of the human-vehicle-road micro-traffic system according to any one of claims 1 to 6. An electronic device characterized by this.

10. A computer-readable storage medium storing a computer program, wherein the computer program is executed by a processor so as to realize the method for modeling and risk identification of the human-vehicle-road microscopic traffic system according to any one of claims 1 to 6.

Images