Traffic safety evaluation method and device, electronic equipment and nonvolatile storage medium

Abstract

The invention discloses a traffic safety evaluation method and device, electronic equipment and a nonvolatile storage medium. The method comprises the steps that intersection design information of an intersection is acquired, and a simulation road network corresponding to the intersection is determined according to the intersection design information; a plurality of different traffic scenes are set for the simulation road network, vehicle track data of the vehicle model in the simulation road network under the different traffic scenes are obtained, and the different traffic scenes correspond to different intelligent network connection permeability and left turn proportions; according to the vehicle trajectory data, determining traffic conflict numbers in different traffic scenes, and according to the traffic conflict numbers, determining influence parameters of intelligent network connection permeability and / or a left turn ratio on traffic conflicts; and constructing a traffic conflict model according to the influence parameters. According to the method and the device, the technical problem of poor safety evaluation effect caused by low model fitting degree during traffic safety evaluation in related technologies is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent traffic safety, in particular to a traffic safety evaluation method and device, electronic equipment and nonvolatile storage medium. BACKGROUND

[0002] In urban traffic network, road signal intersection is one of the bottleneck points of traffic operation and high-incident areas of traffic safety, and its traffic operation characteristics are complex, which is easy to cause frequent start-stop, sudden acceleration and deceleration of vehicles, and regional congestion, and more likely to cause traffic safety problems. With the development of intelligent networked vehicle technology, the introduction of intelligent networked vehicles is expected to improve road traffic capacity and reduce traffic congestion. Intelligent networked vehicle technology can effectively improve the level of road traffic safety through vehicle-road cooperation and automatic driving, but at present, intelligent networked vehicles are still in the development stage from low penetration rate to high penetration rate, and the influence of intelligent networked vehicles on traffic safety still needs to be further quantitatively evaluated.

[0003] The traffic safety evaluation model in the related art has deficiencies in processing the influence of intelligent networked penetration rate and other factors on traffic conflicts, which limits the applicability and accuracy of the model. Especially in the case of the continuous rise of intelligent networked vehicle market penetration rate, there is a lack of effective model to quantitatively analyze the specific influence of this change on traffic safety, resulting in the technical problems of poor traffic safety evaluation effect and the like.

[0004] At present, no effective solution has been proposed to solve the above problems. SUMMARY

[0005] The embodiments of the present application provide a traffic safety evaluation method and device, electronic equipment and nonvolatile storage medium, to at least solve the technical problem of poor safety evaluation effect caused by the low fitting degree of the model used in the traffic safety evaluation in the related art.

[0006] According to an aspect of the embodiments of the present application, a traffic safety evaluation method is provided, including: obtaining intersection design information of an intersection, and determining a simulation road network corresponding to the intersection according to the intersection design information; setting a plurality of different traffic scenarios for the simulation road network, and obtaining vehicle trajectory data of a vehicle model in the simulation road network under different traffic scenarios, wherein the intelligent network penetration rate and the left turn proportion corresponding to different traffic scenarios are different, the intelligent network penetration rate is used to represent the proportion of intelligent network vehicles in the traffic flow, and the left turn proportion is the proportion of left turn vehicles in vehicles entering the intersection; determining the number of traffic conflicts under different traffic scenarios according to the vehicle trajectory data, and determining the influence parameter of the intelligent network penetration rate and / or the left turn proportion on the traffic conflict according to the number of traffic conflicts, wherein the influence parameter is used to represent the influence degree of the intelligent network penetration rate and / or the left turn proportion on the traffic conflict; and constructing a traffic conflict model according to the influence parameter, wherein the traffic conflict model is used to evaluate the traffic conflict level of the intersection under different intelligent network penetration rates and left turn proportions.

[0007] Optionally, the intersection design information includes at least one of the following: geometric design information, signal lamp design information; determining the simulation road network corresponding to the intersection according to the intersection design information includes: obtaining actual map data of the intersection, and converting the actual map data into a road network format supported by a simulation tool to obtain an initial simulation road network; adjusting the lane parameters in the initial simulation road network according to the geometric design information, so that the simulation road network is consistent with the geometric design of the intersection, wherein the geometric design information includes at least one of the following: lane width, lane speed limit, lane type; setting the phase and cycle of the signal lamp in the initial simulation road network according to the signal lamp design information, so that the signal lamp design of the simulation road network is consistent with the signal lamp design of the intersection, wherein the signal lamp design information includes at least one of the following: phase type of the signal lamp, green signal ratio, green light duration.

[0008] Optionally, setting a plurality of different traffic scenarios for the simulation road network includes: determining a reference intelligent network penetration rate and a left turn proportion interval range according to the actual traffic data of the intersection, wherein the reference intelligent network penetration rate is used to represent the initial proportion of intelligent network vehicles in the actual traffic flow, and the left turn proportion interval range is used to represent the ratio range of the left turn vehicle flow to the total vehicle flow at the intersection; determining a plurality of different levels of intelligent network penetration rates according to the reference intelligent network penetration rate, and selecting a plurality of different left turn proportions in the left turn proportion interval range; and arranging and combining the plurality of intelligent network penetration rates and the plurality of left turn proportions to obtain a plurality of different traffic scenarios.

[0009] Optionally, the obtaining the vehicle trajectory data of the vehicle model in the simulation road network under different traffic scenarios comprises: determining the vehicle model, and setting the driving parameters corresponding to the vehicle model, wherein the vehicle model corresponding to the intelligent network vehicle and the human-driven vehicle is different, and the driving parameters comprise at least one of the following: vehicle acceleration, deceleration, driver imperfection coefficient, minimum vehicle spacing, minimum vehicle headway, speed factor, and speed deviation; loading the configuration data of the simulation road network and the vehicle model by using a simulation tool to simulate the traffic under different traffic scenarios, and obtaining the vehicle trajectory data of the vehicle model in each signal light period in the simulation process, wherein the vehicle trajectory data comprises at least one of the following: vehicle position, vehicle speed, and vehicle acceleration.

[0010] Optionally, the determining the vehicle model comprises: determining a first vehicle model corresponding to the human-driven vehicle, wherein the first vehicle model is used to simulate the car-following behavior of the human-driven vehicle; and determining a second vehicle model corresponding to the intelligent network vehicle, wherein the second vehicle model comprises a first mode and a second mode; in the case that the preceding vehicle followed by the intelligent network vehicle is a non-human-driven vehicle, the second vehicle model is the first mode, and in the case that the preceding vehicle followed by the intelligent network vehicle is a human-driven vehicle, the second vehicle model is the second mode, the first mode is used to simulate the car-following behavior of the intelligent network vehicle under the cooperative driving with other intelligent network vehicles through information interaction, and the second mode is used to simulate the car-following behavior of the intelligent network vehicle when the intelligent network vehicle perceives through its own sensor.

[0011] Optionally, the determining the number of traffic conflicts under different traffic scenarios according to the vehicle trajectory data comprises: performing linear fitting according to the vehicle trajectory data to obtain a motion trajectory equation of the vehicle model, and determining potential conflict points between the vehicle models according to the motion trajectory equation; determining a conflict index value according to the potential conflict points and the vehicle trajectory data, wherein the conflict index value comprises collision time and rear intrusion time, the collision time is the time required for two vehicle models to collide if they continue to travel at the current speed on the same path, and the rear intrusion time is the time difference between the time point at which the preceding vehicle leaves the potential conflict point and the time point at which the following vehicle reaches the potential conflict; in the case that the conflict index value is less than the corresponding index threshold value, it is determined that a collision conflict event will occur between the vehicle models; and the number of collision conflict events in the signal light period is counted to obtain the number of traffic conflicts.

[0012] Optionally, constructing the traffic conflict model according to the influence parameter comprises: establishing a multivariate Poisson-lognormal distribution regression model according to the traffic conflict numbers corresponding to different traffic scenarios, and the intelligent network penetration rate and the left-turn proportion corresponding to the traffic scenarios, wherein the multivariate Poisson-lognormal distribution regression model is used to quantify the influence of different intelligent network penetration rates and left-turn proportions on the traffic conflicts, and the influence parameter is a model parameter to be estimated in the multivariate Poisson-lognormal distribution regression model; an uninformed prior probability distribution is used to deduce a posterior distribution of the model parameter, and the Markov chain Monte Carlo simulation method is used to determine the model parameter to be estimated in the multivariate Poisson-lognormal distribution regression model based on the posterior distribution, so as to obtain the traffic conflict model.

[0013] According to another aspect of the embodiments of the present application, a traffic safety evaluation device is also provided, comprising: a road network simulation module, configured to obtain intersection design information of an intersection, and determine a simulation road network corresponding to the intersection according to the intersection design information; a data collection module, configured to set a plurality of different traffic scenarios for the simulation road network, and obtain vehicle trajectory data of a vehicle model in the simulation road network under different traffic scenarios, wherein the intelligent network penetration rate and the left-turn proportion corresponding to different traffic scenarios are different, the intelligent network penetration rate is used to represent the proportion of intelligent network vehicles in the traffic flow, and the left-turn proportion is the proportion of left-turn vehicles among vehicles entering the intersection; a conflict detection module, configured to determine the number of traffic conflicts under different traffic scenarios according to the vehicle trajectory data, and determine an influence parameter of the intelligent network penetration rate and / or the left-turn proportion on the traffic conflicts according to the number of traffic conflicts, wherein the influence parameter is used to represent the influence degree of the intelligent network penetration rate and / or the left-turn proportion on the traffic conflicts; and a model evaluation module, configured to construct a traffic conflict model according to the influence parameter, wherein the traffic conflict model is used to evaluate the traffic conflict level of the intersection under different intelligent network penetration rates and left-turn proportions.

[0014] According to still another aspect of the embodiments of the present application, an electronic device is also provided, comprising a memory and a processor, wherein the processor is configured to run a program stored in the memory, and the program is configured to execute a traffic safety evaluation method when running.

[0015] According to still another aspect of the embodiments of the present application, a non-volatile storage medium is also provided, comprising a stored computer program, wherein a device in which the non-volatile storage medium is located executes a traffic safety evaluation method by running the computer program.

[0016] According to still another aspect of the embodiments of the present application, a computer program product is also provided, comprising a computer program, wherein the computer program is executed by a processor to implement the steps of a traffic safety evaluation method.

[0017] In the embodiment of the present application, the intersection design information of the intersection is acquired, and a simulation road network corresponding to the intersection is determined according to the intersection design information; a plurality of different traffic scenes are set for the simulation road network, and vehicle trajectory data of a vehicle model in the simulation road network under different traffic scenes is acquired, wherein the intelligent network penetration rate and the left turn proportion corresponding to different traffic scenes are different, the intelligent network penetration rate is used to represent the proportion of intelligent network vehicles in the traffic flow, and the left turn proportion is the proportion of left turn vehicles among vehicles entering the intersection; the number of traffic conflicts under different traffic scenes is determined according to the vehicle trajectory data, and the influence parameter of the intelligent network penetration rate and / or the left turn proportion on the traffic conflict is determined according to the number of traffic conflicts, wherein the influence parameter is used to represent the influence degree of the intelligent network penetration rate and / or the left turn proportion on the traffic conflict; the traffic conflict model is constructed according to the influence parameter, wherein the traffic conflict model is used to evaluate the traffic conflict level of the intersection under different intelligent network penetration rates and left turn proportions, the influence of different left turn proportions and intelligent network penetration rates on the traffic conflict is quantitatively analyzed by setting traffic scenes with different intelligent network penetration rates and left turn proportions, and the purpose of improving the accuracy of traffic safety evaluation is achieved, thereby solving the technical problem of poor safety evaluation effect caused by low model fitting degree in the related art when performing traffic safety evaluation. BRIEF DESCRIPTION OF DRAWINGS

[0018] The drawings described herein are used to provide further understanding of the present application, and form a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application, and do not constitute an improper limitation on the present application. In the drawings:

[0019] Figure 1 It is a hardware structure block diagram of a computer terminal (or electronic equipment) for implementing a method for traffic safety evaluation according to an embodiment of the present application;

[0020] Figure 2 It is a schematic diagram of a method flow of traffic safety evaluation according to an embodiment of the present application;

[0021] Figure 3 It is a schematic diagram of a method flow of intelligent network signal intersection traffic safety evaluation based on traffic conflict prediction according to an embodiment of the present application;

[0022] Figure 4 It is a schematic diagram of a signal intersection geometric design model according to an embodiment of the present application;

[0023] Figure 5 It is a schematic diagram of a vehicle following model simulation scene according to an embodiment of the present application;

[0024] Figure 6is a schematic diagram of a vehicle-vehicle conflict type of a signal intersection according to an embodiment of the present application;

[0025] Figure 7 is a schematic diagram of a coefficient correlation analysis according to an embodiment of the present application;

[0026] Figure 8 is a schematic diagram of a change range of intelligent network penetration rate according to an embodiment of the present application;

[0027] Figure 9 is a schematic diagram of a left turn proportion distribution of a signal intersection as a whole according to an embodiment of the present application;

[0028] Figure 10a is a schematic diagram of a conflict vehicle space-time trajectory of vehicle A avoidance according to an embodiment of the present application;

[0029] Figure 10b is a schematic diagram of a conflict vehicle space-time trajectory of vehicle B avoidance according to an embodiment of the present application;

[0030] Figure 11 is a schematic diagram of a traffic conflict number under different traffic scenarios according to an embodiment of the present application;

[0031] Figure 12 is a schematic diagram of a traffic safety evaluation device according to an embodiment of the present application. DETAILED DESCRIPTION

[0032] In order to enable persons skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by persons skilled in the art without creative labor should fall within the scope of protection of the present application.

[0033] It should be noted that the terms "first", "second", and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or a chronological sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but can include other steps or units that are not clearly listed or inherent to the process, method, product or device.

[0034] The road traffic safety evaluation method in the related art mainly includes a direct safety evaluation method and an indirect safety evaluation method. The direct safety evaluation method is mainly based on a historical traffic accident data model to analyze traffic accident influencing factors, but the method has certain limitations, such as randomness and timeliness of accidents or inconsistency of accident cause reports, which will lead to incomplete accident data samples and result deviation.

[0035] The indirect safety evaluation method indirectly evaluates traffic safety through a surrogate safety analysis method, which makes up for the shortcomings of the traditional accident evaluation method. The traffic conflict technique is the most commonly used surrogate safety evaluation method, which can provide feedback more quickly than accident data collection, and can be simulated and reproduced quickly through microscopic traffic simulation technology. On the other hand, due to the limited real accident data in intelligent networked hybrid traffic safety research, the combination of traffic conflict technology and microscopic simulation technology provides technical support for the safety evaluation research of intelligent networked traffic system. Therefore, many scholars conduct traffic safety research based on the combination of microscopic simulation tools and traffic conflict technology.

[0036] Traffic conflict, as a potential traffic accident, is widely used in traffic accident prediction and control. As an indirect evaluation method of traffic safety, the traffic conflict technique replaces traffic accidents with the number of traffic conflicts per unit time and the severity of the conflict, thereby analyzing the traffic safety level at signalized intersections, which can diagnose and evaluate the traffic safety condition of signalized intersections in advance without waiting for accidents to occur. The most commonly used conflict indicators include time to collision (TTC) and post encroachment time (PET). From the perspective of PET, the smaller the PET value, the closer the two vehicles are at the data collection time, and the higher the probability of collision. From the perspective of TTC, the smaller the TTC value, the closer the two vehicles are at the data collection time, and the higher the speed of the following vehicle compared to the leading vehicle, and the higher the probability of collision. Studies have shown that traffic conflict data has the characteristics of count data, and is not continuous. Existing traffic conflict models for signalized intersections tend to ignore this characteristic, and use a single conflict indicator, while ignoring the differential performance of traffic conflict influencing factors and failing to analyze the correlation of conflict factors, resulting in biased results and low model fitting accuracy. In addition, intelligent networked vehicles are developing rapidly, but there are few conflict models considering intelligent network penetration rate (MPR), and the impact of different intelligent network penetration rates on traffic conflicts needs further research.

[0037] To solve the above problems, the related solutions are provided in the embodiments of the present application, which are described in detail below.

[0038] According to the embodiments of the present application, a method embodiment of traffic safety evaluation is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from here.

[0039] The method embodiment provided by the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 A hardware structure block diagram of a computer terminal (or electronic equipment) for implementing the traffic safety evaluation method is shown. As shown in the figure, Figure 1 The computer terminal 10 (or electronic equipment) can include one or more processors 102 (the processor 102 can include but not limited to a microprocessor MCU or a programmable logic device FPGA processing device), a memory 104 for storing data, and a transmission device 106 for communication function. In addition, it can also include a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which can be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. Those skilled in the art can understand that, Figure 1 The structure shown is only schematic, which does not limit the structure of the above-mentioned electronic device. For example, the computer terminal 10 can also include more or less components than Figure 1 shown, or have a different configuration than Figure 1 shown.

[0040] It should be noted that the one or more processors 102 and / or other data processing circuits described above can be referred to as "data processing circuits" herein. The data processing circuit can be embodied in whole or in part as software, hardware, firmware or any combination thereof. In addition, the data processing circuit can be a single independent processing module, or any one of the other elements combined into the computer terminal 10 (or electronic equipment) in whole or in part. As referred to in the embodiments of the present application, the data processing circuit as a kind of processor control (for example, the selection of variable resistance terminal path connected with the interface).

[0041] The memory 104 can be used to store software programs of application software and modules, such as program instructions / data storage devices corresponding to the traffic safety evaluation method in the embodiments of the present application, and the processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, implements the traffic safety evaluation method described above. The memory 104 can include a high-speed random access memory, and can also include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory 104 can further include a memory remotely arranged with respect to the processor 102, which can be connected to the computer terminal 10 through a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0042] The transmission device 106 is used to receive or send data via a network. Specific examples of the above-mentioned network can include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to be able to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (Radio Frequency, RF) module, which is used to communicate with the Internet in a wireless manner.

[0043] The display can be, for example, a touch screen type liquid crystal display (LCD), which can enable a user to interact with the user interface of the computer terminal 10 (or electronic device).

[0044] Under the above-mentioned operating environment, the embodiments of the present application provide a traffic safety evaluation method, Figure 2 is a schematic diagram of a method flow of traffic safety evaluation according to the embodiments of the present application, as Figure 2 shown, the method includes the following steps:

[0045] Step S202, obtaining intersection design information of the intersection, and determining a simulation road network corresponding to the intersection according to the intersection design information;

[0046] Step S204, setting a plurality of different traffic scenes for the simulation road network, and obtaining vehicle trajectory data of a vehicle model in the simulation road network under different traffic scenes, wherein the intelligent network connection penetration rate and the left turn proportion corresponding to different traffic scenes are different, the intelligent network connection penetration rate is used to represent the proportion of intelligent network connection vehicles in the traffic flow, and the left turn proportion is the proportion of left turn vehicles among vehicles entering the intersection;

[0047] In step S206, the number of traffic conflicts in different traffic scenarios is determined according to the vehicle trajectory data, and the influence parameter of the intelligent network penetration rate and / or the left turn proportion on the traffic conflict is determined according to the number of traffic conflicts, wherein the influence parameter is used to represent the influence degree of the intelligent network penetration rate and / or the left turn proportion on the traffic conflict.

[0048] In step S208, a traffic conflict model is constructed according to the influence parameter, wherein the traffic conflict model is used to evaluate the traffic conflict level of the intersection under different intelligent network penetration rates and left turn proportions.

[0049] Through the above steps, by setting different intelligent network penetration rates and left turn proportions of the traffic scene, the influence of different left turn proportions and intelligent network penetration rates on traffic conflicts is quantitatively analyzed, and the purpose of improving the accuracy of traffic safety evaluation is achieved, thereby solving the technical problem of poor safety evaluation effect caused by the low fitting degree of the model used in the related art when performing traffic safety evaluation.

[0050] The traffic safety evaluation method in steps S202 to S208 of the embodiment of the present application is further introduced below.

[0051] Figure 3 is a schematic diagram of a method flow of intelligent network signal intersection traffic safety evaluation based on traffic conflict prediction according to the embodiment of the present application, as Figure 3 shown, the embodiment of the present application can be used in an intelligent network mixed traffic environment, and a simulation tool (taking the traffic micro-simulation platform SUMO (Simulation of Urban Mobility) as an example) is used to build an intelligent network signal intersection traffic scene, and random traffic flow is simulated. From the obtained real-time simulation data of vehicles and real-time state of signal lights, the vehicle space-time trajectory information is analyzed, the dynamic traffic parameters and the number of traffic conflicts under different conflict indicators are counted in units of signal periods, the different influences of different intelligent network penetration rates and left turn proportions on traffic operation conditions are considered, the commonly used conflict indicators TTC and PET are selected, an intelligent network intersection multi-scene left turn conflict prediction method is provided, and the influence of different left turn proportions and intelligent network penetration rates on traffic conflicts is quantitatively analyzed. The specific introduction is as follows.

[0052] Firstly, the simulation road network can be constructed based on the intersection design information of the actual intersection, and the specific steps are as follows.

[0053] In some embodiments of the present application, the intersection design information includes at least one of the following: geometric design information, signal lamp design information; according to the intersection design information, determining the simulation road network corresponding to the intersection includes the following steps: obtaining the actual map data of the intersection, and converting the actual map data into a road network format supported by the simulation tool to obtain an initial simulation road network; according to the geometric design information, adjusting the lane parameters in the initial simulation road network to make the simulation road network consistent with the geometric design of the intersection, wherein the geometric design information includes at least one of the following: lane width, lane speed limit, lane type; according to the signal lamp design information, setting the phase and cycle of the signal lamp in the initial simulation road network to make the signal lamp design of the simulation road network consistent with the signal lamp design of the intersection, wherein the signal lamp design information includes at least one of the following: phase type of the signal lamp, green ratio, green light duration.

[0054] Specifically, the actual signal intersection can be selected for traffic data collection, including the geometric design information of the signal intersection, the signal lamp design information and the morning and evening peak traffic flow data, as shown in Figure 4 According to the signal intersection geometric design diagram and the signal timing strategy, the SUMO simulation scene (simulation road network) is built, and in addition, the actual map of the intersection can be obtained through the map database, and then the road speed limit, road length, road type and priority are adjusted and corrected through the NETCONVERT tool to be closer to the real road design. For example, the signal control is a four-phase scheme, and the signal timing setting can be obtained by the green ratio and the green light duration of each phase, that is, the south-north straight and right turn is 34s, and the left turn is 25s; the east-west straight and right turn is 23s, and the left turn is 29s, and the signal cycle is 124s; the intersection signal lamp phase is shown in the following table.

[0055]

[0056]

[0057] The embodiments of the present application accurately restore the geometric design and signal lamp control strategy of the intersection, ensuring the accuracy of the traffic simulation model, which is the basis for effective traffic safety evaluation. The accurate input of the geometric design information and the signal lamp design information ensures that the simulation road network can reflect the real traffic environment, so that the prediction of traffic conflicts is more reliable.

[0058] In addition, the car-following model can be designed, and the corresponding driving parameters of the vehicle model are set, as follows.

[0059] In some embodiments of the present application, determining the vehicle model comprises the following steps: determining a first vehicle model corresponding to the human-driven vehicle, wherein the first vehicle model is used to simulate the car-following behavior of the human-driven vehicle; determining a second vehicle model corresponding to the intelligent network vehicle, wherein the second vehicle model comprises: a first mode and a second mode; wherein, in the case that the preceding vehicle followed by the intelligent network vehicle is a non-human-driven vehicle, the second vehicle model is in the first mode, and in the case that the preceding vehicle followed by the intelligent network vehicle is a human-driven vehicle, the second vehicle model is in the second mode, the first mode is used to simulate the car-following behavior of the intelligent network vehicle in cooperative driving with other intelligent network vehicles through information interaction, and the second mode is used to simulate the car-following behavior of the intelligent network vehicle when it perceives through its own sensors.

[0060] Specifically, as shown in Figure 5 , the car-following models that can be selected include but are not limited to CACC (Cooperative Adaptive Cruise Control), ACC (Adaptive Cruise Control), IDM (Intelligent Driver Model), etc. In the embodiments of the present application, the car-following model of the human-driven vehicle can be set as the IDM model (i.e. the first vehicle model) in SUMO, and the driving parameters such as vehicle acceleration, deceleration, and minimum inter-vehicle distance are defined; the car-following model of the intelligent network vehicle is set as the second vehicle model, including the CACC model (i.e. the first mode) and the ACC model (i.e. the second mode). The reason for such setting is that the car-following behavior of the intelligent network vehicle will be affected by the state of the preceding vehicle, and when the intelligent vehicle follows the human-driven vehicle in the CACC model, its car-following model will degenerate into the ACC model. In the embodiments, the parameters of the three car-following model vehicles are shown in the following table.

[0061]

[0062]

[0063] In the process of simulation, the vehicle model can be controlled according to the Traci interface in SUMO, and data can be obtained through road detectors. First, the Traci interface is used for secondary development to obtain the preceding vehicle id of the vehicle in the road network. If the preceding vehicle is a human-driven vehicle, the following vehicle will be converted from the CACC mode to the ACC mode. Meanwhile, road detectors can be laid to obtain data such as road traffic and the lane where the vehicle is located.

[0064] Further, the embodiment of the present application also sets multiple different traffic scenarios for the simulation road network to obtain vehicle trajectory data of the vehicle model in the simulation road network under different traffic scenarios. The analysis and design process of different traffic scenarios is introduced below.

[0065] Firstly, the analysis process of traffic conflict influencing factors is introduced.

[0066] Traffic conflict can indirectly reflect the level of traffic safety. Traffic conflict refers to the proximity of two or more road users in time and space, i.e., the proximity to a certain limit in time and space, at which time if each party does not take appropriate risk avoidance measures, a collision may occur in the road environment. According to the location of the conflict, due to the proximity between vehicles, the traffic conflicts of vehicles driving from the main road to the signal intersection area are mostly rear-end conflicts and crossing conflicts, as shown in FIG. 1. Figure 6

[0067] It is found that the current traffic conflict influencing factors based on the signal cycle level include: straight traffic flow (V S ), left-turn traffic flow (V L ), maximum queue length (Q), shock wave speed (S), shock wave area (A), and vehicle passing rate (P), average maximum waiting time (Waiting time), the number of vehicles passing through the green light (AOG), and the influence of different intelligent network penetration rates on traffic conflicts in the intelligent network signal intersection mixed traffic environment is also proved to be significant, so the intelligent network penetration rate (MPR) is also a traffic parameter affecting traffic conflicts. The embodiment of the present application performs Spearman coefficient correlation analysis on the nine cycle dynamic parameters, and the correlation coefficient between traffic conflict influencing factors and conflict number and the correlation coefficient between variables are analyzed. The results are shown in FIG. 2. Figure 7 The first column is the correlation coefficient value between the conflict number and itself, straight traffic flow, left-turn traffic flow, intelligent network penetration rate, average maximum waiting time, vehicle passing rate, number of vehicles passing through the green light, maximum queue length, shock wave speed, and shock wave area. According to the Spearman correlation coefficient between parameters, there is a certain correlation between the nine traffic parameters and the traffic conflict number. Among them, the correlation between straight traffic flow (V S ), left-turn traffic flow (V L ), and intelligent network penetration rate (MPR) and traffic conflict is above 0.5, which indicates that V S , V L , and MPR have strong correlation with traffic conflict, and it is worth further analyzing their influence on traffic conflict.

[0068] ​Based on the above analysis, the embodiments of the present application use two parameters, intelligent network penetration rate and left turn ratio, to set different traffic scenarios, as follows.

[0069] In some embodiments of the present application, setting multiple different traffic scenarios for the simulation road network includes the following steps: determining a reference intelligent network penetration rate and a left turn ratio interval range according to the actual traffic data of the intersection, wherein the reference intelligent network penetration rate is used to represent the initial proportion of intelligent network vehicles in the actual traffic flow, and the left turn ratio interval range is used to represent the range of the ratio of left turn vehicle flow to total vehicle flow at the intersection; determining multiple different levels of intelligent network penetration rate according to the reference intelligent network penetration rate, and selecting multiple different left turn ratios within the left turn ratio interval range; and arranging and combining the multiple intelligent network penetration rates and the multiple left turn ratios to obtain multiple different traffic scenarios.

[0070] Specifically, in the mixed traffic environment of the intelligent network signal intersection, due to the intelligent vehicle information system of the intelligent network vehicle, information transmission, position sharing and information calculation and processing are realized, which can effectively improve the road capacity and reduce traffic congestion. Different intelligent network penetration rates are also proven to cause differences in traffic conflicts. In SUMO, given the intelligent network penetration rate, a mixed traffic simulation experiment is performed. According to the real-time vehicle data, it can be known that during the simulation running process, the intelligent network penetration rate is floating within the interval range of the reference intelligent network penetration rate plus or minus 15%, as shown in FIG. 2. Therefore, in the present embodiment, the MPR can be set to 20%, 50% and 80% as low intelligent network penetration rate, medium intelligent network penetration rate and high intelligent network penetration rate scenarios, respectively. Figure 8

[0071] In addition, considering the influence of different straight-left ratios (left turn ratios) on traffic conflicts, the left turn ratio interval range can be determined according to the data information of each entrance road of a signal intersection in a city. For example, assuming that there are four entrance roads, 1h data information can be obtained for the morning peak (1h / 5min) and the evening peak (1h / 5min) respectively, totaling 104 sets of vehicle trajectory information. The overall left turn ratio of the signal intersection is as shown in FIG. 3, with a distribution interval of [0.08, 0.48], and 90% of the data range interval is [0.1, 0.4]. Therefore, in the present embodiment, the left turn ratios of 0.1, 0.2, 0.3 and 0.4 can be selected to study the influence of different intelligent network penetration rates on traffic conflicts. Figure 9

[0072] ​​Therefore, in the embodiment, traffic scenarios with left turn ratios of 0.1, 0.2, 0.3, and 0.4 can be selected, and the penetration rates of intelligent networking are taken as 20%, 50%, and 80% as representatives of low intelligent networking penetration rate, medium intelligent networking penetration rate, and high intelligent networking penetration rate. Twelve traffic scenarios are divided according to different intelligent networking penetration rates and different left turn ratios, as shown in the following table.

[0073]

[0074]

[0075] The embodiment of the application scientifically constructs a plurality of traffic scenarios by setting the reference intelligent networking penetration rate and the left turn ratio interval range, covers the change trend of the intelligent networking vehicle market penetration rate from low to high and the dynamic distribution of the left turn vehicle ratio, can simulate a plurality of interaction modes of the intelligent networking vehicle and the traditional vehicle at the intersection, and provides a rich data basis for subsequent traffic conflict analysis.

[0076] Then, the vehicle trajectory data of the vehicle model in the simulation road network under different traffic scenarios can be obtained, as follows.

[0077] In some embodiments of the application, obtaining the vehicle trajectory data of the vehicle model in the simulation road network under different traffic scenarios includes the following steps: determining the vehicle model and setting the driving parameters corresponding to the vehicle model, wherein the vehicle models corresponding to the intelligent network vehicle and the human-driven vehicle are different, and the driving parameters include at least one of the following: vehicle acceleration, deceleration, driver imperfection coefficient, minimum vehicle spacing, minimum vehicle headway, speed factor, and speed deviation; using a simulation tool to load the configuration data of the simulation road network and the vehicle model to simulate the traffic situation under different traffic scenarios and obtain the vehicle trajectory data of the vehicle model in each signal light period of the simulation process, wherein the vehicle trajectory data includes at least one of the following: vehicle position, vehicle speed, and vehicle acceleration.

[0078] The embodiment of the application loads these configuration data through the simulation tool, can simulate the dynamic driving process of the vehicle at the signal intersection, including the start, acceleration, deceleration, parking, and turning of the vehicle, which provides detailed dynamic data for traffic conflict analysis. For example, in the embodiment, 281585 pieces of vehicle trajectory information are extracted in 240 signal periods of the signal intersection.

[0079] Further, the number of traffic conflicts under different traffic scenarios can be determined according to the vehicle trajectory data, as follows.

[0080] In some embodiments of the present application, the number of traffic conflicts in different traffic scenarios is determined according to the vehicle trajectory data, including the following steps: linear fitting is performed according to the vehicle trajectory data to obtain a motion trajectory equation of the vehicle model, and potential conflict points between the vehicle models are determined according to the motion trajectory equation; a conflict index value is determined according to the potential conflict points and the vehicle trajectory data, wherein the conflict index value includes: a collision time and a post-encroachment time, the collision time is the time required for two vehicle models to collide if they continue to travel at the current speed on the same path, and the post-encroachment time is the time difference between the time point at which the front vehicle leaves the potential conflict point and the time point at which the rear vehicle reaches the potential conflict; in the case that the conflict index value is less than the corresponding index threshold value, it is determined that a collision conflict event will occur between the vehicle models; the number of collision conflict events in a signal light period is counted to obtain the number of traffic conflicts.

[0081] Specifically, the conflict determination indexes: time to collision (TTC) and post encroachment time (PET) can be selected for the judgment of vehicle conflict events. In this embodiment, the TTC threshold is 3s, and the PET threshold is 5s, that is, when TTC≤3s or PET≤5s, it is a conflict event.

[0082] Wherein, the collision time TTC is defined as: the time required for two vehicles to collide if they continue to travel at the current speed on the same path, Figure 10a The space-time trajectory and conflict time information of two vehicles in the case of vehicle A avoiding are shown, it can be seen that vehicle A and vehicle B start to discover that traffic conflict may occur at time t1, after the reaction time of the driver, vehicle A decides to take the strategy of slowing down, and vehicle B decides to take the strategy of accelerating forward, and finally the two sides avoid collision; if the two vehicles do not take any avoiding behavior according to their original driving dynamics, vehicle A and vehicle B will collide at t4, at this time the value of TTC is equal to t4-t1. This is a theoretical value of collision time calculated by a simple method. However, in actual traffic operation, the vehicle has its own length, and the vehicle trajectory cannot be regarded as a mass point, so the vehicle length should be considered when calculating the collision time, and the calculation formula is:

[0083]

[0084] In the formula, x L,t and x F,t are the positions of the front and rear vehicles at time t, v L,t and v F,t correspond to the speeds of the front and rear vehicles at time t, and L is the length of the front vehicle.

[0085] The post-entry time PET is defined as the time difference between the leading vehicle leaving the conflict area and the trailing vehicle arriving at the conflict area. Figure 10b As shown in the figure, vehicles A and B realize a potential collision at time t1 and take different actions. Vehicle A maintains its original speed, while vehicle B decides to slow down and yield. As a result, vehicle A reaches the potential collision point at t3, and vehicle B reaches the predicted collision point at t5. Therefore, the post-intrusion time (PET) is numerically equal to t5 - t3. PET calculation is relatively simple and does not require speed or position information. PET can be expressed as:

[0086] PET t =t F,t -t L,t

[0087] Where, t F,t is the time when the following vehicle reaches the conflict line, t L,t The time it takes for the car ahead to leave the conflict line.

[0088] For example, after analyzing the vehicle trajectory data extracted in this embodiment, there are 14,129 TTC conflicts and 973 PET conflicts. The results of traffic conflicts affected by the left-turn ratio under different intelligent network penetration rates are as follows: Figure 11 As shown in the figure, the penetration rate of intelligent connected vehicles significantly affects traffic conflicts. Traffic conflicts are minimal at an 80% penetration rate, and the total number of traffic conflicts is minimal at a 20% penetration rate. Furthermore, different straight-left ratios (left-turn ratios) at signalized intersections have different impacts on traffic conflicts. Across the three penetration rates, a left-turn ratio of 0.1 results in the fewest traffic conflicts. At low and medium penetration rates, a left-turn ratio of 0.3 results in the most traffic conflicts. At high penetration rates, a left-turn ratio of 0.4 results in the most traffic conflicts. Therefore, conflicts vary significantly across different traffic scenarios, and it is necessary to comprehensively consider the impact of different intelligent connected vehicle penetration rates and left-turn ratios on traffic conflicts.

[0089] In summary, based on the 12 traffic scenarios, the number of traffic conflicts under different traffic scenarios, as well as the statistical description of through traffic flow and left-turn traffic flow, are shown in the following table.

[0090]

[0091] Afterwards, a traffic conflict model based on the multivariate Poisson-lognormal distribution can be constructed. The Markov chain Monte Carlo (MCMC) simulation method is used to analyze the results, quantify the influencing parameters of the intelligent network penetration rate and / or the left-turn ratio on traffic conflicts, and obtain the traffic conflict model, as shown below.

[0092] In some embodiments of the present application, the traffic conflict model is constructed according to the influence parameter, including the following steps: a multivariate Poisson-lognormal distribution regression model is established according to the traffic conflict number corresponding to different traffic scenes, and the intelligent network penetration rate and the left turn proportion corresponding to the traffic scene, wherein the multivariate Poisson-lognormal distribution regression model is used to quantify the influence of different intelligent network penetration rates and left turn proportions on traffic conflicts, and the influence parameter is a model parameter to be estimated in the multivariate Poisson-lognormal distribution regression model; the posterior distribution of the model parameter is derived by using the non-informative prior probability distribution, and the model parameter to be estimated in the multivariate Poisson-lognormal distribution regression model is determined based on the posterior distribution by using the Markov chain Monte Carlo simulation method, so as to obtain the traffic conflict model.

[0093] Specifically, the multivariate Poisson-lognormal distribution regression model (MVPLN) can solve the problem that traffic conflicts are different and related to each other under different traffic scenes.

[0094] Let Y ik represent the traffic conflict number of the i th sample under the k th traffic scene, k = 1, 2, …, K, K is the total number of traffic flow scenes, and it is assumed that y i obeys an independent distribution, y ik obeys a Poisson distribution with parameter μ ik , denoted as:

[0095] Y ik | θ ik ~ Poisson (θ ik )

[0096]

[0097] ln (θ ik ) = ln (μ ik ) + ε ik

[0098] wherein the MVPLN model of the error term ε ik is:

[0099] ln (μ ik ) = β k0 + β k1 X k1 + β k2 X k2 + β k3 X k3

[0100] In the formula, θ ik represents the expected value of the traffic conflict number of the i th sample under the k th traffic scene; μ ikXik represents the number of traffic conflicts at the i-th sample under the k-th traffic flow scenario, i.e., the expected traffic conflicts in a signal cycle; X k1 k2 k3 respectively represent the logarithm ln(V S ), the logarithm ln(V L ) of the left-turn traffic volume, and the intelligent network penetration rate MPR; β k0 , β k1 , β k2 , β k3 represent the coefficients of the corresponding explanatory variables, i.e., the model parameters (impact parameters) to be estimated. ε ik is the error term of the i-th sample under the k-th traffic scenario, ε ik obeys a multi-dimensional normal distribution, i.e., ε i ~N K (0,Σ).

[0101]

[0102] Let X represent the covariate matrix; β represents the corresponding coefficient variable, i.e., β = {β1, β2, …, β m}, and m is 3 in this embodiment. Given the matrix (X, β, Σ), θ i obeys an independent distribution, and its probability density function is

[0103]

[0104] The following non-informative prior probability distribution is adopted:

[0105]

[0106] where 0 K represents a 0 vector matrix of Kx1; I K represents a unit matrix of KxK; P and r represent the Wishart distribution parameters, and r≥K is the degree of freedom, P = I K and r = K are taken in this paper. The probability density function of the Wishart distribution is:

[0107] f(Σ -1 |P, r) = |P| K / 2 |Σ -1 | (r-K-1) / 2 exp{-0.5Tr(PΣ -1 )}

[0108] For the parameters [β, Σ], let the parameters For the y vector, y = (y1, y2, …, y n ). Then, given X, the joint probability distribution of (y, θ, β, Σ) is:​​

[0109]

[0110] wherein,

[0111]

[0112] According to the Bayesian principle, the posterior probability distribution of each estimated parameter is:

[0113]

[0114] wherein,

[0115]

[0116] The posterior probability density distribution of unknown parameters (i.e. influence parameters) in a multivariate Poisson-lognormal distribution regression model is obtained by a Markov chain Monte Carlo (MCMC) Bayesian method, and a traffic conflict model is obtained.

[0117] After obtaining the traffic conflict model, model result analysis and experimental cross-validation can be further performed.

[0118] Specifically, the Deviance Information Criterion (DIC) can be used as a statistical goodness-of-fit index to compare the established Bayesian models. When the model converges, the DIC value of the model can be obtained by continuing to iterate the simulation of the model. The smaller the DIC value, the better the fitting effect of the model on the conflict. The DIC can be expressed as:

[0119]

[0120] wherein, D represents the model deviation; is the posterior mean of the model deviation, which is-2 times the logarithmic likelihood function; represents the point estimate of each parameter; p D is an index indicating the number of effective parameters in the explanatory model. The smaller the DIC value, the higher the model fitting degree.

[0121] The model fitting degree can be determined by the DIC value, the model is adaptively verified by selecting an actual urban signal intersection section, and different intelligent network penetration rates and straight left turn traffic flow ratios are set to verify the effectiveness of the model.

[0122] In the study of traffic safety at signalized intersections in the existing intelligent network environment, due to the limited disclosure of real vehicle data and other reasons, there is a lack of real accident data resources in the study of intelligent network mixed traffic safety. The combination of traffic conflict technology and micro-simulation technology provides technical support for the safety evaluation of intelligent network traffic system. Therefore, in order to make the traffic scene more real and specific, the embodiments of the application select the actual urban road signal intersection section as the main research scene, take the mixed traffic simulation environment composed of human-driven vehicles and intelligent network vehicles as the research object, and establish an intelligent network intersection multi-scene straight-left conflict model for the influence of the straight-left ratio on the intelligent network signal intersection traffic conflict. Quantitative analysis of traffic conflicts in different traffic scenarios can quantitatively analyze the influence of different intelligent network penetration rates and straight-left ratios on traffic conflicts, and further quantitatively evaluate the traffic safety of signalized intersections.

[0123] The application scheme takes the signal period as the statistical unit, comprehensively considers multiple conflict types, and performs conflict statistics by using two conflict judgment indexes TTC and PET (which can simultaneously evaluate longitudinal (rear-end) conflicts and lateral (lane-changing) conflicts), thereby avoiding the deviation caused by actual observation and the subjective feeling of drivers in estimating traffic conflicts. Compared with the traditional single conflict judgment index, the application scheme has higher authenticity and more reliable data. In addition, the application scheme provides a multi-straight-left conflict model of intelligent network mixed traffic at signalized intersections (traffic conflict model) in an intelligent network vehicle mixed traffic environment. The straight-left conflict model of the intersection multi-scene is established by considering different intelligent network penetration rates and straight-left ratios. The experiment and the traffic conflict model in a single traffic scene are compared. The influence of different left-turn ratios and intelligent network penetration rates on traffic conflicts is quantitatively analyzed, and the adaptability of the model is proved. The application scheme provides a new idea for the market penetration rate of intelligent network vehicles in the future and the influence of the market penetration rate on traffic safety. By analyzing the communication demand of intelligent network vehicles at intersections, network resource allocation can be optimized to ensure stable and low-delay communication services in high-density traffic areas.

[0124] According to the embodiments of the application, an embodiment of a traffic safety evaluation device is also provided. Figure 12 As shown in FIG. 1, the device includes: Figure 12

[0125] The road network simulation module 120 is configured to acquire intersection design information of the intersection, and determine a simulation road network corresponding to the intersection according to the intersection design information.

[0126] ​The data collection module 122 is configured to set a plurality of different traffic scenarios for the simulation road network, and obtain vehicle trajectory data of the vehicle model in the simulation road network under different traffic scenarios, wherein the intelligent network penetration rate and the left-turn proportion corresponding to different traffic scenarios are different, the intelligent network penetration rate is used to represent the proportion of intelligent network vehicles in the traffic flow, and the left-turn proportion is the proportion of left-turn vehicles in vehicles entering the intersection.

[0127] The conflict detection module 124 is configured to determine the number of traffic conflicts under different traffic scenarios according to the vehicle trajectory data, and determine the influence parameter of the intelligent network penetration rate and / or the left-turn proportion on the traffic conflicts according to the number of traffic conflicts, wherein the influence parameter is used to represent the influence degree of the intelligent network penetration rate and / or the left-turn proportion on the traffic conflicts.

[0128] The model evaluation module 126 is configured to construct a traffic conflict model according to the influence parameter, wherein the traffic conflict model is used to evaluate the traffic conflict level of the intersection under different intelligent network penetration rates and left-turn proportions.

[0129] Optionally, the intersection design information includes at least one of the following: geometric design information, signal lamp design information; determining the simulation road network corresponding to the intersection according to the intersection design information includes: obtaining actual map data of the intersection, and converting the actual map data into a road network format supported by a simulation tool to obtain an initial simulation road network; adjusting lane parameters in the initial simulation road network according to the geometric design information, so that the simulation road network is consistent with the geometric design of the intersection, wherein the geometric design information includes at least one of the following: lane width, lane speed limit, lane type; setting phases and cycles of signal lamps in the initial simulation road network according to the signal lamp design information, so that the signal lamp design of the simulation road network is consistent with the signal lamp design of the intersection, wherein the signal lamp design information includes at least one of the following: phase type of the signal lamp, green signal ratio, green light duration.

[0130] Optionally, setting a plurality of different traffic scenarios for the simulation road network includes: determining a reference intelligent network penetration rate and a left-turn proportion interval range according to actual traffic data of the intersection, wherein the reference intelligent network penetration rate is used to represent the initial proportion of intelligent network vehicles in the actual traffic flow, and the left-turn proportion interval range is used to represent the ratio range of the left-turn vehicle flow to the total vehicle flow at the intersection; determining a plurality of different levels of intelligent network penetration rates according to the reference intelligent network penetration rate, and selecting a plurality of different left-turn proportions in the left-turn proportion interval range; and arranging and combining the plurality of intelligent network penetration rates and the plurality of left-turn proportions to obtain a plurality of different traffic scenarios.

[0131] Optionally, the obtaining the vehicle trajectory data of the vehicle model in the simulation road network under different traffic scenarios comprises: determining the vehicle model, and setting the driving parameters corresponding to the vehicle model, wherein the vehicle model corresponding to the intelligent network vehicle and the human-driven vehicle is different, and the driving parameters comprise at least one of the following: vehicle acceleration, deceleration, driver imperfection coefficient, minimum vehicle spacing, minimum vehicle headway, speed factor, and speed deviation; loading the configuration data of the simulation road network and the vehicle model by using a simulation tool to simulate the traffic under different traffic scenarios, and obtaining the vehicle trajectory data of the vehicle model in each signal light period in the simulation process, wherein the vehicle trajectory data comprises at least one of the following: vehicle position, vehicle speed, and vehicle acceleration.

[0132] Optionally, the determining the vehicle model comprises: determining a first vehicle model corresponding to the human-driven vehicle, wherein the first vehicle model is used to simulate the following behavior of the human-driven vehicle; and determining a second vehicle model corresponding to the intelligent network vehicle, wherein the second vehicle model comprises a first mode and a second mode; in the case that the preceding vehicle followed by the intelligent network vehicle is a non-human-driven vehicle, the second vehicle model is the first mode, and in the case that the preceding vehicle followed by the intelligent network vehicle is a human-driven vehicle, the second vehicle model is the second mode, the first mode is used to simulate the following behavior of the intelligent network vehicle in the cooperative driving with other intelligent network vehicles through information interaction, and the second mode is used to simulate the following behavior of the intelligent network vehicle when the intelligent network vehicle perceives through its own sensor.

[0133] Optionally, the determining the number of traffic conflicts under different traffic scenarios according to the vehicle trajectory data comprises: performing linear fitting according to the vehicle trajectory data to obtain a motion trajectory equation of the vehicle model, and determining potential conflict points between the vehicle models according to the motion trajectory equation; determining a conflict index value according to the potential conflict points and the vehicle trajectory data, wherein the conflict index value comprises collision time and rear intrusion time, the collision time is the time required for two vehicle models to collide if they continue to travel at the current speed on the same path, and the rear intrusion time is the time difference between the time point at which the preceding vehicle leaves the potential conflict point and the time point at which the following vehicle reaches the potential conflict; in the case that the conflict index value is less than the corresponding index threshold value, it is determined that a collision conflict event will occur between the vehicle models; and the number of collision conflict events in the signal light period is counted to obtain the number of traffic conflicts.

[0134] Optionally, constructing the traffic conflict model according to the influence parameter comprises: establishing a multivariate Poisson-lognormal distribution regression model according to the traffic conflict numbers corresponding to different traffic scenes, and the intelligent network penetration rate and the left-turn proportion corresponding to the traffic scenes, wherein the multivariate Poisson-lognormal distribution regression model is used to quantify the influence of different intelligent network penetration rates and left-turn proportions on the traffic conflict, and the influence parameter is a model parameter to be estimated in the multivariate Poisson-lognormal distribution regression model; an uninformed prior probability distribution is used to deduce the posterior distribution of the model parameter, and the Markov chain Monte Carlo simulation method is used to determine the model parameter to be estimated in the multivariate Poisson-lognormal distribution regression model based on the posterior distribution, so as to obtain the traffic conflict model.

[0135] It should be noted that each module in the traffic safety evaluation device described above can be a program module (for example, a set of program instructions that implement a certain specific function) or a hardware module. For the latter, it can be in the following form, but is not limited to this: the form of each module described above is a processor, or the functions of each module described above are implemented by a processor.

[0136] It should be noted that the traffic safety evaluation device provided in the embodiment can be used to execute the traffic safety evaluation method shown in Figure 2 Therefore, the related explanations and descriptions of the traffic safety evaluation method described above also apply to the embodiments of the present application, and will not be repeated here.

[0137] The non-volatile storage medium provided in the embodiments of the present application includes a stored computer program, wherein the device in which the non-volatile storage medium is located executes the following traffic safety evaluation method by running the computer program: obtaining intersection design information of an intersection, and determining a simulation road network corresponding to the intersection according to the intersection design information; setting a plurality of different traffic scenes for the simulation road network, and obtaining vehicle trajectory data of a vehicle model in the simulation road network under different traffic scenes, wherein the intelligent network penetration rate and the left-turn proportion corresponding to different traffic scenes are different, the intelligent network penetration rate is used to represent the proportion of intelligent network vehicles in the traffic flow, and the left-turn proportion is the proportion of left-turn vehicles among the vehicles entering the intersection; determining the number of traffic conflicts under different traffic scenes according to the vehicle trajectory data, and determining the influence parameter of the intelligent network penetration rate and / or the left-turn proportion on the traffic conflict according to the number of traffic conflicts, wherein the influence parameter is used to represent the influence degree of the intelligent network penetration rate and / or the left-turn proportion on the traffic conflict; and constructing a traffic conflict model according to the influence parameter, wherein the traffic conflict model is used to evaluate the traffic conflict level of the intersection under different intelligent network penetration rates and left-turn proportions.

[0138] The embodiment of the present application further provides a computer program product comprising a computer program, which, when executed by a processor, implements the steps of the traffic safety evaluation method described in various embodiments of the present application: obtaining intersection design information of an intersection, and determining a simulation road network corresponding to the intersection according to the intersection design information; setting a plurality of different traffic scenes for the simulation road network, and obtaining vehicle trajectory data of a vehicle model in the simulation road network under different traffic scenes, wherein the intelligent network penetration rate and the left-turn proportion corresponding to different traffic scenes are different, the intelligent network penetration rate is used to represent the proportion of intelligent network vehicles in the traffic flow, and the left-turn proportion is the proportion of left-turn vehicles among vehicles entering the intersection; determining the number of traffic conflicts under different traffic scenes according to the vehicle trajectory data, and determining an influence parameter of the intelligent network penetration rate and / or the left-turn proportion on the traffic conflicts according to the number of traffic conflicts, wherein the influence parameter is used to represent the influence degree of the intelligent network penetration rate and / or the left-turn proportion on the traffic conflicts; and constructing a traffic conflict model according to the influence parameter, wherein the traffic conflict model is used to evaluate the traffic conflict level of the intersection under different intelligent network penetration rates and left-turn proportions.

[0139] The above sequence numbers of the embodiments of the present application are only for description, and do not represent the advantages or disadvantages of the embodiments.

[0140] In the above embodiments of the present application, the description of each embodiment has its own emphasis, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0141] In the several embodiments provided by the present application, it should be understood that the disclosed technology can be implemented in other ways. Of course, the unit embodiment described above is only schematic. For example, the division of the units can be different, and each unit or some features can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual coupling or direct coupling or communication connection between each represented or discussed unit can be indirect coupling or communication connection through some interface, electrical or other form.

[0142] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, that is, they can be located in one place, or they can be distributed on multiple units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0143] In addition, each function unit in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software function unit.

[0144] When the integrated unit is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application, essentially or in the form of a contribution to the prior art, or all or part of the technical solutions 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 causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The foregoing storage medium includes various media that can store program codes, such as a U disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a mobile hard disk, a magnetic disk or an optical disk, etc.

[0145] The above is only the preferred embodiment of the present application, and it should be pointed out that, for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, and these improvements and refinements should be considered as the protection scope of the present application.

Claims

1. A traffic safety evaluation method, characterized in that: include: Obtaining intersection design information of the intersection, and determining a simulated road network corresponding to the intersection based on the intersection design information; Setting a plurality of different traffic scenarios for the simulated road network, and obtaining vehicle trajectory data of the vehicle model in the simulated road network under the different traffic scenarios, wherein the different traffic scenarios correspond to different intelligent connected vehicle penetration rates and left-turn ratios, the intelligent connected vehicle penetration rate being used to characterize the proportion of intelligent connected vehicles in the traffic flow, and the left-turn ratio being the proportion of left-turning vehicles among vehicles entering the intersection; Determining the number of traffic conflicts in different traffic scenarios based on the vehicle trajectory data, and determining, based on the number of traffic conflicts, an impact parameter of the intelligent network penetration rate and / or the left-turn ratio on the traffic conflicts, wherein the impact parameter is used to characterize the degree of impact of the intelligent network penetration rate and / or the left-turn ratio on the traffic conflicts; Based on the influencing parameters, a traffic conflict model is constructed, wherein the traffic conflict model is used to evaluate the traffic conflict level at intersections under different intelligent network penetration rates and left-turn ratios.

2. The traffic safety evaluation method according to claim 1, characterized in that: The intersection design information includes at least one of the following: geometric design information, signal light design information; Determining a simulated road network corresponding to the intersection according to the intersection design information includes: Acquiring actual map data of the intersection, and converting the actual map data into a road network format supported by a simulation tool to obtain the initial simulated road network; Adjusting lane parameters in the initial simulated road network based on the geometric design information so that the simulated road network is consistent with the geometric design of the intersection, wherein the geometric design information includes at least one of the following: lane width, lane speed limit, and lane type; Based on the traffic light design information, the phase and period of the traffic lights in the initial simulated road network are set so that the traffic light design of the simulated road network is consistent with the traffic light design of the intersection, wherein the traffic light design information includes at least one of the following: the phase type of the traffic light, the green-to-signal ratio, and the green light duration.

3. The traffic safety evaluation method according to claim 2, characterized in that: Setting a plurality of different traffic scenarios for the simulated road network includes: Determine a baseline intelligent network penetration rate and a left-turn ratio range based on actual traffic data at the intersection, wherein the baseline intelligent network penetration rate is used to represent the initial proportion of intelligent network vehicles in the actual traffic flow, and the left-turn ratio range is used to represent the ratio of left-turn vehicle flow to total vehicle flow at the intersection; Determining a plurality of different levels of the intelligent connected vehicle penetration rate based on the benchmark intelligent connected vehicle penetration rate, and selecting a plurality of different left-turn ratios within the left-turn ratio range; Arrange and combine multiple intelligent network penetration rates and multiple left-turn ratios to obtain multiple different traffic scenarios.

4. The traffic safety evaluation method according to claim 1, characterized in that: Acquiring vehicle trajectory data of the vehicle model in the simulated road network under different traffic scenarios includes: Determining a vehicle model and setting driving parameters corresponding to the vehicle model, wherein the vehicle model corresponding to an intelligent network vehicle and a human-driven vehicle is different, and the driving parameters include at least one of the following: vehicle acceleration, deceleration, driver imperfection coefficient, minimum vehicle spacing, minimum headway, speed factor, and speed deviation; A simulation tool is used to load the configuration data of the simulated road network and the vehicle model to simulate traffic conditions under different traffic scenarios, and obtain the vehicle trajectory data of the vehicle model in each traffic light cycle of the simulation process, wherein the vehicle trajectory data includes at least one of the following: vehicle position, vehicle speed, and vehicle acceleration.

5. The traffic safety evaluation method according to claim 4, characterized in that: Determining the vehicle model includes: Determining a first vehicle model corresponding to a human-driven vehicle, wherein the first vehicle model is used to simulate a car-following behavior of the human-driven vehicle; Determining a second vehicle model corresponding to the intelligent network vehicle, wherein the second vehicle model includes: a first mode and a second mode; Among them, when the preceding vehicle followed by the intelligent network vehicle is not a human-driven vehicle, the second vehicle model is the first mode; when the preceding vehicle followed by the intelligent network vehicle is a human-driven vehicle, the second vehicle model is the second mode. The first mode is used to simulate the following behavior of the intelligent network vehicle under collaborative driving through information interaction between other intelligent network vehicles, and the second mode is used to simulate the following behavior of the intelligent network vehicle when perceiving through its own sensors.

6. The traffic safety evaluation method according to claim 4, characterized in that: Determining the number of traffic conflicts in different traffic scenarios based on the vehicle trajectory data includes: Performing linear fitting based on the vehicle trajectory data to obtain motion trajectory equations of vehicle models, and determining potential conflict points between vehicle models based on the motion trajectory equations; Determining a conflict index value based on the potential conflict point and the vehicle trajectory data, wherein the conflict index value includes: a collision time and a rear intrusion time, wherein the collision time is the time required for two vehicle models to collide if they continue to travel on the same path at the current speed; and the rear intrusion time is the time difference between the time when the leading vehicle leaves the potential conflict point and the time when the trailing vehicle arrives at the potential conflict point; When the conflict index value is less than the corresponding index threshold, determining that a collision conflict event will occur between the vehicle models; The number of collision conflict events in the traffic light cycle is counted to obtain the number of traffic conflicts.

7. The traffic safety evaluation method according to claim 4, characterized in that: Based on the influencing parameters, constructing a traffic conflict model includes: A multivariate Poisson-lognormal distribution regression model is established based on the number of traffic conflicts corresponding to different traffic scenarios, as well as the intelligent network penetration rate and the left-turn ratio corresponding to the traffic scenarios. The multivariate Poisson-lognormal distribution regression model is used to quantify the impact of different intelligent network penetration rates and left-turn ratios on traffic conflicts, and the influencing parameters are model parameters to be estimated in the multivariate Poisson-lognormal distribution regression model. The posterior distribution of the model parameters is derived using an uninformative prior probability distribution, and the model parameters to be estimated in the multivariate Poisson-lognormal distribution regression model are determined based on the posterior distribution using a Markov chain Monte Carlo simulation method to obtain the traffic conflict model.

8. A traffic safety evaluation device, characterized in that: include: A road network simulation module is used to obtain intersection design information of the intersection and determine a simulated road network corresponding to the intersection based on the intersection design information; a data collection module, configured to set a plurality of different traffic scenarios for the simulated road network and obtain vehicle trajectory data of the vehicle model in the simulated road network under the different traffic scenarios, wherein the different traffic scenarios correspond to different intelligent connected vehicle penetration rates and left-turn ratios, the intelligent connected vehicle penetration rate being used to characterize the proportion of intelligent connected vehicles in the traffic flow, and the left-turn ratio being the proportion of left-turning vehicles among vehicles entering the intersection; a conflict detection module, configured to determine, based on the vehicle trajectory data, the number of traffic conflicts in different traffic scenarios, and, based on the number of traffic conflicts, determine an impact parameter of the intelligent network penetration rate and / or the left-turn ratio on the traffic conflict, wherein the impact parameter is used to characterize the degree of impact of the intelligent network penetration rate and / or the left-turn ratio on the traffic conflict; A model evaluation module is used to construct a traffic conflict model based on the influencing parameters, wherein the traffic conflict model is used to evaluate the traffic conflict level at intersections under different smart network penetration rates and left-turn ratios.

9. An electronic device, characterized in that: include: A memory and a processor, wherein the processor is used to run a program stored in the memory, wherein the traffic safety evaluation method according to any one of claims 1 to 7 is executed when the program is run.

10. A non-volatile storage medium, characterized in that: The non-volatile storage medium includes a stored computer program, wherein the device where the non-volatile storage medium is located executes the traffic safety evaluation method according to any one of claims 1 to 7 by running the computer program.

11. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the traffic safety evaluation method according to any one of claims 1 to 7 are implemented.

Images